LIBRAR Y OF CONGR ESS. (il^qi (S0pi|nrjl}l 1?o-. Shell' UNITED STATES OF AMERICA. r^y- 'j*\? '•^^' iiP; '■'laf WEST'S MOULDERS' TEXT-BOOK: BEING PART II. OF AMERICAN FOUNDRY PRACTICE. PRESENTING ORIGINAL METHODS AND RULES FOR OBTAINING GOOD, SOUND, CLEAN CASTINGS; AND GIVING DETAILED DESCRIPTION FOR INIAKING MOULDS REQUIRING SKILL AND EXPERIENCE. ALSO CONTAINING A PRACTICAL TREATISE UPON THE CONSTRUCTION OF CRANES AND CUPOLAS, AND THE MELTING OP IRON AND SCRAP-STEEL IN IRON FOUNDRIES. BY ^^^ THOMAS D. WEST, PBACTICAL IRON MOULDER AND FOUNDRY FOREMAN, MEMBER OP THE AMERICAN SOCIETY OP MECHANICAL ENGINEERS, AND OP THE CIVIL ENGINEERS' CLUB OP CLEVELAND, OHIO. FJTLLY ILLUSTRATED. NEW YORK ; JOHN WILEY & SONS, 15 AsTOR Place. 1885. \\ ^ \ii CorviiKiHT, 18S5, By THOMAS D. AVEST. ^'?7^9 ELKCTnoTYPKIJ AND PIlINTKl) UX UANU, AVKKY, AND COMl'ANV, BOSTON. PREFACE. Although it is more than two years since the appearance of the first volume, the author cannot refrain from here tender- ing his most sincere thanks to the press and public of America and England for the cordial reception given his first book. Also, to American foundrymen and moulders the author is greatly indebted for the universally rapid introduction of his work among them. The compliments which were so kindl}' tendered the first volume have encouraged and stimulated the author to write this second book. Many of the original articles here submitted, as in vol. i., appeared in " The American Machinist," and have been revised for this volume. Also, many of these articles have had valua- ble additions made to them. The subjects of Cupolas and Melting, also those of mould- ing in green sand, in dry sand, and in loam, are extensively treated; and this volume, in connection with vol. i., it is thought affords a thorough presentation of each subject. The author received many communications regretting the lack of a treatise upon cranes in the first volume : hence he has endeavored to present, in this, the practical and essential j y, riiEFArE ft'.'iliiros t(i Ito ooiisi(l(M'0(l in pioporly oonRtrnctiiip; tliciii for fi>iiii(liv iisf. .Iil», post, and tnivclliiij; rinncs arc (n-alfil, so llial ideas of piactical value may lie ohtaineil, eillicr frr \'fiils I'A.iK . 101 . KtS ri;()(ri;i\(; fi-KAV fixisiiki) rAsTiXfiS from rjiiEEX- NAND Mori'.DS. fasting Finishod Work Horizontally . . . . Heavy ami Light AVork Skiniininj^-tialfS 'l"()l>-lM)urinL; (Jatcs, and Swi-fping a Lathe Face-Platc 114 120 12!) METHODS AND KULES FOR (;REE\-SAND AND GENERAL MOULDING. Small Castings, — The Mould-Roanl and Flask-Hinge IMpes, Orei'n-Sand Cons and Hollow I'ipc TattL-rns r.cdding-in and Rolling-over Coping, Venting, and Jointing Green-Sand Moulds Dnuving and Making Patterns .... Skin-Drying Green-Sand Moulds .... Setting and Centring Cores Improper Setting and Wedging of Chaplets. Momentum and Kules for Weighting down Copes and Cores 1.34 140 140 ir>-> 164 1G9 173 179 187 MLSCELLANEOUS CHAPTERS. Elements and Manufacture of Foundry Facing . ■\Velding Steel to Cast-Iron, and Mending Cracked Castings Foundry Addition, — Oven and Pits .... Ladle and Casting Carriage combined .... Making Chilled Rolls, and Roll I-'lask, Runners and Gates ^Moulding-Machines ....... ]".<>lis. IikI. (ar," nipola) .•{.>! ( hicap., 111. {iU;' X 4:^ cupola) . :Vm (Jalr.shuri;, III. {•.10" cupola) . -.m I'.floit, Wis. {40'' cupola) . :i:,l Minn«'ai.. Fig. ^. This gives a body which, l»3' means of a leeding-rod, and by occasionally pouring hot iion in the feeding-head, will remain in a lluid slate as long as the heavy portion. This accom- SOUND CASTING. 3 plished, it can be readily seen tliat tlie formation of a cavity, as at A, Fig. 1, is prevented. (It must be, however, understood, that enlarging a section, as in Fig. 3, is only recommended in cases where it is not practicable to attach independent feeding-heads. Where such a section, as in Fig. 3, is at the outer portion of a mould, and the heavy part to be fed is below the joint of the mould, it ma}^, in many cases, be fed by feeders placed from G" to 8" from the surface of the mould. Connections running from the feeding- heads, if the case would not admit of branch gates being drawn inward or outward, could easily be formed with cores having holes of the size required.) Now, it is by no means practicable to attain soundness in all castings by the above means ; for there are many moulds in which the intended form of the casting would be made almost unrecognizable, were the}' to have all their heavy sections thus reached and fed by risers. Attending this is often the imprac- ticability of placing over three or four feeders upon a mould ; for often the bars of the cope, chaplets, binders, and weights will not permit the use of any more. Then, again, were it prac- ticable to have a cope filled with feeding-heads, there are many castings, which, in order to be sound, would require that more men be taken off from the work of "running off the heat" than foundries at casting-time can generally spare. It is very evident, from the shapes of existing patterns and castings, that but little thought has been given to this element involved in obtaining an entirely sound casting. The best place to study this error is at the scrap-pile. There one can find the shrinkage-hole in many forms. Often fillets which were intended as factors for strength will be found to be exactly the reverse. The greater part of machinery castings inade are more or less filleted ; and some designers have the idea that the larger the fillet, the greater the strength given. In cases where the fillet is fed by other metal than that contained in its 4 SOl'NI) CASTINt;. cfiitral IkhIv, this may lu' triic. Often fillets arc so hituatcil, tliev eaiiiiot Ite {yi\ \>\ oilier than tile metal euiitaiiied within their ^" balls, 3^" diameter; for Gf b:dl, 4" diameter; for 1U2" ball, ')" diameter. For the liist three sizes, the height of the head from the llask-joint up to the top of gate was i)", and for the l'>jj" balls the gale-head was 12". The gates which admitted the metal into the moulds were cut broad and very thin, in uider that SOUND CASTING. O they should freeze a few moments after the nioiilcl became full, thereby insuring that metal did not enter through the pouring- gates to supply au}^ shrinkage. In pouring these balls, the iron was medium hot, and tlie gates were filled up to the heights given. The balls having the feeding-heads were "churned" until they solidified. In cleaning the castings, the feeding- heads were chipped off, so as to preserve the spherical form of the balls as much as possible. This statement with reference to the manner of moulding and casting the balls is simply given to show the conditions under which the tests were made. The following is a table giving the weights of the balls, and the difference between the fed and the unfed balls : — FIRST HEAT. Mixture of Iron. 200 lbs. ordinary No. 2 pitr and 400 tbs. scrap. inAMKTER OK Balls. Fed. Unfed. SUUINKAGE FOIINP. Pkbcentauf, op Shrinkage. 4" 5s" or' 103" 8 tbs. 12 OZ. 20 tbs. 11 oz. . .39 tt)8. 10^ oz. 150 tbs. 8 tbs. 10 oz. 20 lbs. 8 oz. 39 lbs. 4 oz. 147 lbs. 15 oz. 2 oz. 3 oz. 6^oz. 33 oz. 1.428 0.906 1.024 1.375 SECOND HEAT. Mixture of lion. 100 lbs. No. 1, Bessemer. A strong colje iron. 100 tbs. No. 1, Hubbard. A strong colce iron. 100 lbs. No. 1, Pine Grove. A strong ctiarcoal iron. 300 lbs. Machinery scrap. DIAMETEK OF Balls. Fed. Unfed. SHRINK AGE FOUND. PERrENTAOE OF Shrinkage. 4" 5i" 63" lOi" 8 lbs. 131 oz. 20 tbs. 13 oz. 39 tbs. 11^ OZ. 149 tbs. 12 OZ. 8 lt)S. 12 oz. 20 tbs. 9 oz. 39 tbs. 6 oz. 148 tbs. 7 oz. IJ OZ. 4 OZ. Si OZ. 21 OZ. 1.060 1.201 0.805 0.876 This second heat was poured with middling fluid iron. SOI'NI) CASTINO. TJinil) IIKAT. Mirliirr .n HtH. No. 1, lliitibiird. A KlroriK rokc- iron. 2iKi |Im«. MiicliiniTy hit«ii Iron. DiAJdKTI K F»:i.. I'MF.I". SlIBINKAtiK K'lUSIl. ruiKKNTACK OK 61IUINKAUB. 4" f'J" loi" 8 Rxi. 14] oz. '2u ttiM. 14 oz. :i'J lbs. l-.'l oz. 149 n«». 8 oz. 8 Dm. 124 oz. •><) III". '.*{ oz. :ju ll>rt. 7i «z. 14S Dm*. C oz. 21 oz. 41 oz. .'> (iZ. IS nz. 1.272 0.7S.-) o.T'iJ 111 this thinl lu-at, with the exception of the lo^" halls, they \\v\v :ill poured with a more fluid metal than was used in the two iip|ier lu'ats. This I would assiirn as the reason for the (>3", Tii^". and l" balls lieing heavier than in any of tlie other two ln'ats shown. In elassiiiii one heat against another, the mixture of the iron must he taken into cousidi'iation. Balls from each of the respective heats were split in order to learn, if possible, the cau.se (jf the dissimilarity of weight most noticeable in the smaller sizes. f'^D UNFED Fig. 5. The cut at Fig. 5 partly illustrates the fracture of the split balls. The smallest-sized unfed balls showed a very o\)QVi grain SOUND CASTING. 7 at their centres, gradually increasing in density towards the shell. Tlie uufed 10^" balls were not only ver}- porous at their centres, but contained large holes as well. The flat place seen at K shows about how the top part of the unfed balls looked. This was, of course, formed while the crust remained fluid enough to supply shrinkage. After the crust became set, the balance of shrinkage was then drawn from the innermost fluid portion of the balls, as proved by the porousness and holes found when the balls were split open. The fed balls were the most dense in the middle ; the most porous part of them being about midway between the shell and centre, as seen in the cut. The density of some of the fed balls at the centre was remark- able, and was a clear explanation of the cause of their variation in weight. This centre density was, no doubt, mainly caused by the pressure exerted by the feeding-rod, and the occasional supplying of the feeding-heads with hot iron. When feeding a casting, the feeding-rod at the latter end is more or less enlarged, caused by molten metal sticking to it. This may be knocked off, or a new rod used ; but, whichever way is used, there will exist variations in the manipulations of feeding, suflS- cient to cause the dissimilarity in weights seen. It seems reasonable to assert that a thick feeding-rod should exert more of a pressure and disturbance than a thinner rod, and that, the smaller the ball, the more effect could be produced. In moulding these balls, I was very careful in all the manip- ulations performed. The ramming, venting, drawing of the pattern, and gating were as near alike as study and care could make them. In feeding, attention was given to the procuring of solid castings. The 10|" ball would occupy from fifty to sixty minutes to be fed solid ; and, although these largest balls show about the lowest percentage in shrinkage, they no doubt give the nearest approximation that it would be practical to assign to shrinkage in the general run of castings, which, if estimated at one pound for every hundred pounds of casting^ would not be far out of the way. SOUND CASTINO. While it is fsscntial tliat .1 ciistinji; should Ito fi-d solid, to lie stiiiii;^, the (ciiiiM ralmc of (he iioii tist-d is also a I'aflur for rousidi'ialion. Some time ap) I madi' the asserti(jn, that metal poured at a dull heat would produce the strongest iron (an opini<;n then held liy others lieside the writer). Ilavin;^ made this assertion, there coulfl l)e' no one more anxious than myself to have seen this kept a n):iiutained fart. Mr. (iurdiner, foreman of Pratt & AVhitney's foundry at Hartford, Conn., has iuformeil me, that, Fig. 6. — Testing-machine tIirout2;h experiments which he had made with test-bars poured dull and poured hot, he found the hot-poured 1»ars the strongest. Thinkinii; that I might be in error, from the fact that the tests I had made were but few and crudely performed (as can be seen from the description then given), I desired to give the question another and a more thorough test. Having no testing- machme, I devised the simple affair shown in Fig. G for the purpose of dealing with the subject. In using this machine, bars 1" square X 21" long were tested. In all tests, the hot- poured bars stood the greatest load. To make sure that my machine was working correctly, and to know what the results SOUND CARTING. 9 would lie, were heavier liai's used than 1" square, I had some l)atterus made, measuring 4]", '2-\'\ and i\'' square by 24" loug. When cast they were taken into the machine-shop, and accurately planed up to the respective sizes, 4", 3 J", 2", and 1" square. The following table shows the strength of the duU- and hot-poured bars, as found by tests taken by an Olseu machine at the Otis Steel Works, Cleveland, O. : — Section of Babs 24" Long. BUEAKINti Load. Section of B.m:s i'4" LoNU. Bkeakinu Load. 4" square. 4" not . . Dull. . 56,130 49,S30 2" square. 2" Hot . . Dull. . 9,-520 6,400 3r " 3r " Hot . . Dull. . 38,470 36,960 1" 1" Hot . . Dull. . 1,050 1,020 2" 2" Hot . . Dull. . '.."jeo 6,340 1" 1" Hot . . Dull. . 1,130 900 2" 2" Hot . . Dull. . 8,650 6,810 The above bars all showed a perfect fracture, with the excep- tion of the 3|" dull bar, which showed a hone3'combed centre. These S^" bars were intended for 4" : but as soon as the skin was broken when planing the dull bar, blow-holes were seen ; and, thinking that were the bar planed down they might disap- pear, the machinist was instructed to make the bars S^" square. As every cut revealed fresh holes, it was found no cleaner at 3A" than at the 4" square. These blow-holes were readily accounted for by tlie fact that the iron with which this bar was poured was so dull that it would hardly flow out of the ladle. It was purposely so poured in order to learn how it would stand for strength. The result as shown will no doubt be a surprise to many, as it was to me ; for, although this bar showed such a bad fracture, we see that 10 SOUND PASTINO. it stood within 1,')!'^ j)oiin machine is only about eighty pounds: and anyone who may choose to give it a trial would, I think, be pleased with its workings, especially in view of the amount it would cost to make one (which should not exceed six dollars). The machine is best adapted for testing foundry mixtures of iron, and new l)rands of pig-iron. As seen, it will record the three essential points which foundrymen ought to know al)out their iron : — SOUND CASTING. 11 The first is the contraction of tlae iron ; , The seeoud, its clejlectioii ; The third, its strength. lu obtaiuiug the coutraction, the pattern A, from which the test-bars are to be made, should be just the length of tlie dis- tance between the standpoints BB. Then, Avhen the bars are cast, all that is necessary after one is set in place is to keep it tight to one end, and the space at the other will give the contraction. For obtaining tlie deflection, a piece at F has a slot through which a thumb set-screw binds it against the stand H. Before commencing to screw down upon the bar A, the piece F is set down upon the ratchet-wheel K; and, being secured bj' means of the thumb-screw above mentioned, it will, of course, remain stationary. Then, when the bar A breaks, its deflection can be told by the space between J' and the top of the ratchet-wheel. The two arms which F is seen to have are for the purpose of holding a small 2" iron rule, divided into fifty or a hundred parts ; and there are slots in the arms for the purpose of hold- ing the rule. To obtain the strength, the load is applied by means of the screw E, which is 1^", having nine threads to the inch. In the bottom of the screw, there is a steel pin having a bear- ing-surface of about y. The ratchet-wheel K is, of course, secured to the screw E, and a part of the screw projects up aliove it so as to leave a pin for the ratchet-lever D to work up on. The lever D is provided with a ratchet-pawl, so that the operator can stand in the same place while working the screw. Behind the pawl is a spring so as to force it into the teeth of the ratchet. At /S' is a sliding band, which, when pulled back, releases the hold of the spring upon the pawl, thereby allowing the ratchet-wheel, or screw, to be turned back without removing the lever D. At the end of the lever is a common twenty-five-ceut spring-balance scale. Across its face, 12 SOI'MI CASTINf}. at /i', is (Ittcil :i thin iiiccc <»f luass or copper plato. A wire is inscrlccl ill a hiiiall liulr wliicli is ''^«^'=»=Aj^ fif ^\ i':-VvjVTiii-i-.:: Fig. 7. that the principle is used in a varietj^ of forms, made to suit different moulds and conditions. The value of skimming-gates is often lost through the moulder's not using judgment in mak- ing the gates or runners 5, jP, and E^ having a proper relation IS DKFKCTS IN SlUrcri KAI. CASIINfJS. to cMfli (itlitT. /VlioiiliI nlwMVs lie I lie l:ULr«'sl, in onlcr to alTonl room Toi- till- dill to rise. 1', .should he l:ir<;cr tliaii tliat slio'.vn liclow E. If K wc'iL- lar', it would l)e ii dillic-ull iiiatttT to kfC'i) the dirt from itas.siiig into the mould ; for the .simple reason tliat E would lalu' iron fasti-r tlian 7i, tlicrcliy not allow- ing tlic dirt-riser heads I) and F to lie kept full, which must lie doni' in order to collect and hold the imi)urities as shown. KeepinLi; the riser /-'' full is not always a <:;uaranty that the impurities are being collected: the How of metal may he too fast to give the impurities a chance to be held. A point to l»e ke[)t in view is, the longer that metal can be practically mainfained in F and D, before passing into the mould, the more puri^ficd it shordd be. In this cnt shown, the E 1^" gate would be better if it were not so nearly under the dirt-riser F, as shown. The farther away E can practically be carried from F, the more effective will such a skimming-gate be iu catching and holding the dirt. The gates or runners, ZJ, F, and E>, are supposed to be round ; and the sizes shown represent about what relation such skim- ming-gates should bear to each other. The sketch marked '• whirl " shows one of the zrrinkles sometimes used. The con- nection I), if cut from B to F in the manner shown, ■will cause the metal to whirl in F, thereby assisting the dirt or impurities to rise up, as shown at S. The greater the whirl, the better the results. Another plan, sometimes practised to catch and hold impurities from going into a mould, is as shown at N. This is commonly called a skimming-core : it is built or set into the main basin, from 2" to 6" lower than the reservoir's bottom, J'J*; below this core is made a basin, as shown at X; when this basin is tilled with metal, the ladle's dirt is held as shown at F, and the clean iron flows through at X. The amount of dirt or impurities that a well-contrived skimming-gate or basin will gather and keep from going into a easting is often remark- able. The section marked " Direct " shows the method practised DEFECTS IN STRUCTURAL CASTINGS. 19 in ordinaril}' jxnted moulds, in which tlioro is nothing to prevent the dirt or impurities from passing into the mould, except what is held up by keeping the basin full of metal while pouring. I should like to here treat upon other forms of gates and runners in their relation to special forms of eastings ; hut as my time to prepare even what I have was very short, I shall have to dispense with much that should be brought out in order to fully discuss such a subject as the title implies. It is not intended here to convey the impression, that, by having a well- planned skimming-gate, the casting is sure to be free from de- fects. In some cases, where the making of the mould is in the hands of a first-class moulder, it might be so ; but, as a general thing, the skimming-gate is but a small factor. About all that can be said of it is, that it aids in collecting the iynjnirities of iron before it 2^asses into the motild. The engineer has other im- perfections that he often needs to be more watchful of than the impurities of the iron ; consisting of scabs, blow-holes, cold-shuts, misplaced cores, improper feeding, etc. Any one of these could form a hidden defect that would reduce a casting to one-twelfth of what should be its ultimate working strength, and maybe greatly exceed that, going from twelfths to twentieths. Such defects cannot be bridled with mathematics in any form : they are infinite, treacherous, and beyond human reason to define. A scab is part of the mould-surface flaked off, the depth of which varies from ^y up to G" in thickness. When a scab is over \" in thickness, there will be generally more or less visible blowing. Sometimes this mould or casting blowing will become so violent as to tear a mould all to pieces, thereby making the exact form of the intended casting nnrecognizable. Such defects as this will, as a general thing, leave but little doubt as to the casting's future use ; they, being too apparent to deceive, must necessarily be introduced to the scrap-pile. When a mould scabs, the sand mingles with the iron ; some of it ma}' be visible, while some may not ; the sand being specifically lighter than iron, it 20 ni;i'i:crs i\ siurci ikal {'Asiivris nnlurally rises until sl(iiii>c(l ]>y coiitac-l with (mucs e of an even thickness all around. I remember, some twenty-two years back, insi)ectors testing a lot of columns (they might have been pipes; Jiut tiie princii)le involved is the same). ^ The shop where this inspection or testing occnrri'd was at the I'ort- land Locomotive Company's foundry, Poitland. .Me. in th«' testing of these castings, two rails were placed parallel ; and, after being levelled, the castings were raised, one at a time, and set ui)on them. The inspectors would tlien rotate them about one-half their circumference; and, after coming to a stand, they would then be allowed to find their own centre of gravity. B}' this process, any uneveuness of thickness was quickly detected ; and if any of the castings, in revolving to their centre of gravity, went faster thau the allowed speed, they were condemned. To the best of mj' memory, the castings were made in green sand, and cast horizontally. Now, the question in my mind is, AVas not the test somewhat in error? In the horizontal casting of any cylindrical-shaped mould, the cope, or top part of the casting, cannot be as sound as the sides or bottom, for the rea- son that it will be more porous, and coutaiu more dirt than any other portion of the casting. « Mr. A. C. Ootclu'll Plated, during Uic very interesting discnssion wliich followed llic reading of tlie aiillior's paper, (liiil, being in rorlland wlien tlie tests were made, ho r<'ineml)ered lliat liie eolumnii or pipes referred to as tested at tlie I'ortland Locomotive Wurlis were pipes cast borizontall}', and that about onc-lliird of Ibern were eoudeiuued. DEFECTS IN STRUCTURAL CASTINGS. 21 I think it is a safe assertion to make, that if a horizontally cast pipe or column, tliat was found to stay in equililjrium at any point when being tested upon rails, were given an even internal tension, or end compression strain, until it would burst or rupture, the point of first fracture would be the co[)e part of the casting. I should like to hear of such tests being made ; for I do think it would result in opening the minds of many to an important factor in the casting of structural work, which is as follows : Where it is reasonable to expect dirt or jMrousness in castincjs, make that section thicker or heavier than the design woidd other- wise call for, in order to coxinterbalance the tveakening effect caused through the mingling of dirt or impurities with the iron. As to how much thicker the cope section of pipes or columns should be than the sides and bottom, this would be rather a dif- ficult question to answer ; as it would great!}' depend upon the combination of lengths, diameters, and thicknesses, and also facilities for moulding. However, I would say, that, with a pipe or column 12" diameter, f" thick, and fourteen feet long, one-quarter of its thickness added to the cope, as represented by the dotted line H in the column-section cut, would then not always be a guaranty of its holding its own in a testing- machine. In structural castings, the question of proportion, contraction, and qualit}' of metals, contains three very important elements that require careful consideration. But as my limited time would not allow me to now do justice to the discussion of them, I will close with the remark, that to figure for strength in cast- ings is one thing : to know if you have obtained it, is quite another. The former is the work of rules and tables : the latter is only assisted by observation, investigation, and practical experience. P.S. Shrinkage occurs when metal is liquid ; contraction^ when it is cooling off in a solid state. NOVELTIES IN EOrNDUY rUACTICE. NOVELTIES IN FOUNDRY PRACTICE. In the I^itcnt-Offioo buildin.ns at Wasliin^ton, arc many nov- rltios, some good and somk' of vcrv liltlc value. ^lany of them are applianees for mochaiiieal trades, and have been iieralded liefore the i)id)lic. In this line, the moulder's trade has not Iteen very prominent ; where!)}' the public have been led to believe that, to do moulding, no inventive talent was required. There are many tools in a foundry that at one time were just as patentable, and, in fact, far more so, than other things that have been patented. One reason why foundiy novelties are not patented to any extent is because it would not pay. The greatest novelties in foundry practice are generally got up for some special job, which, perhaps, is not made in a half-dozen foundries in the United States, and even those could generally invent other ways to accomplish the end if they desired. Even if a man has something novel, that every foundry could use, be could seldom make it i)ay to attempt its introduction : all would look, lint few would buy. They would look to steal the idea, from which, in many cases, they could get up something else to answer their j^rpose. "When a moulder gets up a new tool or rigging, he seldom thinks of getting it patented. There are some things in foun- diies that reiiuire the highest inventive qualities to originate ; and it is wrong to suppose, that, because the foundry is not extensively represented in the Patent Odice, no invention is required there. If thi' getting-up of something never before known is patentable, tlien there arc foundrymen Avho every year of their lives could be applicants for patent honors. NOVELTIES IN FOUNDRY TRACTICE. 23 There are many who patent things which eventnall}' they would be glad to give awa}^, in view of their experience at a later date. It is one thing to ^'■get up a patent,'' but quite another to get it introduced, and have it earn mone}^ for the inventor: at least, that was the author's experience when he Mas new at this patent business. Ever}' tool or rigging now used in a foundry was at one time more or less of a novelty. Many moulders seem to have the idea that the trade was originated as they found it, and that all that is reqtiired of them is to do as they see others do. The habits and customs of the shop in which the}' learned their trade are theirs : they get to think that there is only one way that a jol) can be done, and that is the way they were taught. AVhat a deplor- able condition the moulder's trade would be in, were there no exceptions to this rule ! Once in a while we come across men who are original. They have, to our views, odd ways of working ; and, if we are fortu- nate enough to be their shopmates for some years, we will often see them adopt new modes of working. Such a man cares nothing for tchat he teas taxight to do : to him it is only a step- ping-stone. Once under way, he begins to forget what he was taught to do, and commences to do that which he learns by his own experience and study. " "What is that John is getting up now? " says some one. " Oh ! something to draw the boss's attention," replies some jealous sore-head. Almost every advancement in a foundr}' is met with more or less ridicule. A progressive moulder is not always welcomed, but is often a target for abuse, especially when he starts in a new shop. It is astonishing how afraid some are of new-com- ers showing or introducing any novelty into a shop : no matter whether it is original or borrowed, if it is a novelt}' to them, tliey will try to ride it down. It is not only the men, but often the foreman as well, that will deride the iutroductiou of any new or stranjic feature. 24 NOVKLTIKS IN FOUNDRY rilACTICE, A moiildcr, in trnvollincc to soo nnd loam, may po tlirongli a (lo/.i'ii .slioi».s, and sei- nt;tliini^ very new or hlraiigc in them. He may soe difFcrcnt classes of work made, Init, for all that, see notliini^ novel to liiin in the way it is made. It will seem as if one master tan^^lit them all. When first starting to work at the tia GEOMETRY IN THE FOUNDRY. 33 centre of any circle to its circumference, as B A, Fig. 8, will divide the circumference into six equal parts F^ Z>, A, P, E, and S. To divide the circumference into three, erase every other one of the six points. To divide it into twelve equally, divide each of the six parts. To divide it into four parts, describe a line through the centre, as /i 0, Fig. 9. From the points where KO intersects the circumference, with tram- mels or dividers set at more than two-thirds the diameter, describe arcs cutting each other, as at N, A line then drawn from N through the centre divides the circle into four parts. Fig. 8. Fig. 9. For eight parts, bisect each of the four. Many divide the circle into four parts by the use of a square and straight-edge, as shown at Fig. 12, instead of by describing the arcs as at N, Fig. 9. The division of circles into odd, or even equal, sections, can be done as follows. At H and jB, Fig. 10, the radius, or one- sixth of the circumference, is set off. This radius is then divided, by trial, into the same number of sections into which it is desired to divide the whole circle. With the trammels or dividers set to the chord of six of the divided radius points, shown in the radius // and 72, space off the circumference. This will divide the circumference, or circle, into the same number of sections as in the radius, which in radius H and R is seven. Should it fail to do so, the fault is yours. It must 34 CJKOMKTUY IN TIIK I'OINDICY 1k' nincinlxTcd, Hint lo r.ra»7/y divide the (•ircmiifiiciicc rc(iuiri's vi'iy fine iii:ini|nil:iti()ii. 'i'licre an* Itiit few iiitii lliiil can go urotiiid the circU' twice, and eoine out exactly alike. Chords one-ninth, one-tenth, and onc-i'li-venth are simply shown to fnrther illustrate the rule. To divide the circle into liflhs, the radins is divided into live ; and, in order to have tho « 8 3 3 Fig. 10. six points, one is added to the radins length, as shown. If by this rule the circle were to be divided into four equal sections, the radius chord would be divided into four sections ; and, in ordiT to have the six points, two would be addetl to the chord's len many convey ideas of lavintj out that may l»e useful. The IteU ljeini)i'r portion of the cyliiidcr B will contain more or less invisible dirt, or the iron will n^>^'-^^^^^^^^^^^ far less than that around the cylindeiv centre core. By pouring the casting as shown at XX, the metal is admitted into the thinnest sections, from which it runs to the heaviest, thereby ■\\ MAKINC CVI.INDKUS AND CASTINfiS 'in FINISH. jxivinii ii far mon' cvfii tiiiipciatiiii' all over tlic iiioiiM than wcif till' iron lirsl niii iiilo llic licavicst, ami all()\vt\ thus doing, it cuts up the dirt or scum in such a manner as to kci'p it ui)on the top, and keep it from gathering in lumps or rolling up against the side of the core or mould. In this way tlu- dirt is kept fioaling ui)on the U)[) of the rising metal, and thereby is MAKING CYLINDERS AND CASTINGS TO FINISH. 45 brought np into the dirt-catcher, or riser-head, shown at W. Cylinders poured from the top generally have a rougher sJdn than those poured from the bottom, caused by the agitation of the rising metal against the mould's surface. They are also more liable to be scabbed and cut than those j^oured from the bottom; but nevertheless they will, as a general thing, finish up clean. It is often surprising how much a casting poured entirely from the top may scab, and still be clean when finished up. Were the same scabs upon a casting poured from the bottom, the chances are ten to one it would present so man}- holes as to disgust any one even to count them, especially if the casting were poured b}' having the metal poured in at the main body of the cylinder through a gate situated similarly to that shown at H. A cylinder poured from the bottom, to Jinish clean, must at least be free from scabs. The plan here shown of having a gate running down to the bottom of the mould, with top runners attached to it in order to pour from the top as well as the bottom, is often a good one for cylinders, as by it you can start slowly to cover over the bottom of the mould, after which the gates can be filled, and the metal be made to enter at the top as well as at the bottom. By first covering over the bottom of the mould, we prevent it from being cut by the falling iron. Some may say this is not necessary ; for how is it that long water or gas pipes can be poured altogether from the top, and yet the bottoms not be cut? In pouring such castings, the iron is prevented from falling directly from the top to the bottom by the thinness of the space between the core and mould ; the iron, in dropping, going from one side to the other, its friction decreases its velocity, and the force of its fall. When such thin castings are scabbed, it is generally the sides of the mould, and not the bottom. With cj'linders or pipes over one inch in thickness, the iron has a freer chance to fall directly upon the bottom, and thereby cut or scab it. 4t) MAKINC; CYI.INDKUS AND rASTINfiS TO riNISII. ('yIukUts siiiiil:ir to llic locomotive sulc-saiUUc ones, tli:it li:ive large foreij^ii altatlniients cast on them, are often belter east by having most of the metal g(j in at the bottom. Should such cylinders as these be poured altogether from the top, the sides of tile mould all the way \i\) would lie lialili' to be eut or seabbfd, or present a very rough body skin, caused by the agi- tation of the metal against the surface (jf the iiKJuld, and the length of time re(iuired for the metal to rise above any given point. The falling iron, instead of directly filling the body of the cylinder, runs away to fill up the side-saddle or the large attachments ; and in the mean time there is danger of the agi- tation of the metal, causing scabbing of the mould's surface. In jioaring any castings, the sooner the agitation of the vietal against the moukVs surface ceases, the better for the casting. With reference to the general plan adopted iu pouring cylin- ders in loam, they are usually poured from the top, while iu dry sand the reverse is true. Often in both instances the top and bottom methods are combined ; especially so when the cylinder is over four-foot stroke. There are other points aljout cylinders that require to be as perfect as the bore, some of which will be found iu the following article, "Moulding and Casting Cylinders." It might be well here to notice the question of unsound riser- heads. Many vertical-cast cylinders have, when their riser-head was cut off, presented a flanged surface full of holes, some of which are often larger than a marble. The writer recalls the ease of a foundry wIumc he worked, that had experienced nuich trouble from this cause. They had tried in every way imagin- able to stop the trouble; but when "working in the dark," there is greater liability of aggravating the difliculty than of remedying it. The whole trouble lay in having too large a corner at K, and tc cMst out of ii;ir:illfl. tli(Til>y often requiring niueli ciiiitpiiig to get llieui true, or [terliaits [treveutiug the intended width of port openings. The cylinder here shown, the author lately cast iu the Cuyahoga AVorks ; and it well illus- CASTING WHOLE OR IN PARTS. 67 trates the advantage of making chest and port cores as described, when practicable. For convenience in handling and making cores, and also to save pattern-making, the steam-chest and belt cores were made in what might be termed half-core boxes. This will be better understood by reference to T T, Fig. 18. The belt core at B B, Fig. 18, has no prints ; this end being held in place by top and bottom chaplets, seen at P P, Fig. 20. The black squares on belt core, Fig. 18, represent chaplets, and show in what order they were placed above and below the core. The steam-chest core was held in place by chaplets at H and S, Fig. 20. The core iV, below F, was made independent of the steam-chest, and was the first core set. As it was to cut through to steam-chest, the opening made a very good bearing for the chest coi'e to rest upon. The lower half of the chest core being set, the next to follow was the belt core, after which the upper half of the chest core was set. In setting the halves of the chest core, care was taken to see that the valve-face portion was true and in line. Of course, had the chest core been whole, or not parted, as shown at T T, this care would not have been re- quired, nor the chaplets H and S needed. After these cores were set, as described, the cope, or upper- cheek portion of the mould, was lowered to place. Then a bolt as seen at TF, near Fig. 19, was placed in order to firmly hold the port core back against the chaplets. The belt core was then chapleted to hold it down : this completed the setting of these cores. After securing the vents, the next operation was ramming up the mould in the pit. This having been done, the mould was cleaned out, and preparation made for lowering in the centre core. The preparation consisted mainly in pla- cing the set screws as seen in the securing-pit at D D, the purpose of which is described in the article "Moulding a Jacket Cylinder," p. 60. Six short pieces of candles being lighted, and placed upon the bottom flange, the centre core was M CASTING WIIOIJ-: OK IN I'AItTS. tlicn lowcrcil into its print. ('I'lic in:iiiiior in wliidi this oore was lift('(l will he understood iiy " Ki'volvin^ Core ami I'ndrr- Surfacc Swcfpin*; " iirtick', p. (i'>.) The core havin, the space between the bottom phite and core rinjj, at 3f, was then securely packed with mud and bricks, in order to iirevent any chance run-out. After this, the eye- l)olt was placed f(n* the purpose of assisting in holdinji; down the core. The space F being now filled with moulding-sand, the G" core was centred and set. Twelve chaplets — three for each of the four cap cores — now being set, the cap cores were placed upon them in position to have their arms come square with the mould, and the outer circumference kept so as to give the r('([uircd tliickni'ss. Tliis completed the setting of all cores. A straight dry-sand cope was then set on, and the vent-holes made. Then, being closed for good, the mould was finished and got ready for casting. In making the chest and port cores, vents were formed, as shown at X, Fig. 20, which clearly represents the manner in which they are arranged. The vents were obtained by the use of straight rods, and the core irons were welded frames. The vents of the chest and port cores were taken off at the print end, the same as shown for the belt core at Y T. The belt core was vented by partly filling the space between the cast-iron pricked frames used for the core rods, as seen at A A, with fine cinders. By this plan there are no joint vents. This is something I always try to avoid as much as possil)lo. Taking otT vents through the joints of cores is always more or less risk}'. Hot iron is about as bad as steam for penetrating cracks or joints. In building up this mould, sweeps were used for the plain cylindrical portions, and patterns for the chest, etc., such as are ordinarily used. In building under the steam-chest, cinders were laid to carry off the vent from core N. Between the lower flange and steam-chest, rods were laid to assist the bricks in CASTING WHOLE OR IN PARTS. 59 firmly holdins; the small body C. The body over the chest was held and lifted by being secured with the rods and plates shown. The style adopted in building up around the steam- chest was different from that generally employed : instead of oiling the pattern, and building it up in soft loam, the bricks were kept back about two inches from the face of the pattern, as shown ; and a drj'-sand mixture, similar to that which would be used for dr3'-sand moulding, rammed between the bricks and the face of the pattern, the bricks having their faces rubbed with a little wet loam in order to make it certain the sand would stick to them. This plan is us^. with much of our work, and it gives good results. With this explanation, the practical loam- moulder will be able to account for the building-iu of the rods over the chest and at C, as shown. In making the centre core, the brickwork being built up to the height showii, the top-plate, after being set on, was partly filled up with cinders, over which a mixture of dry sand was rammed to form the toj) portion of the core. The corners, GG^ are well nailed to assist in holding the projection seen, and to help prevent the falling iron, when pouring the mould, from cutting the sand. In pouring the cylinder, we let about two thousand pounds go in at the bottom gate, shown by dotted entrance Z, Fig. 20. When about fifteen hundred pounds had been poured in, we then started pouring in from the top, through the eleven gates shown in plan, Fig. 17. The size of these gates was |"xl|". At K is represented the feeding-head, which was placed over the steam-chest side. The casting did not present any scabs or sand-holes : the skin was a dark-blue color, and as smooth as a piece of stove-plate. This cylinder is not shown to repre- sent large work, but simply because its making involved points thought to be of a;eueral interest. liU MOULULNG A JACKKTKI) CVLINDKK. MOULDING A JACKETED CYLINDER. At till' left of the engraving (Kig. 21) is shown a section of a jacketed cjlinder, ^vhich will l)e recognized by the practised moulder as being a somewhat difllcult casting to make. The outside of the casting is a simple affair enough ; the dilficulties being confined almost entirely to the centre core, which is shown in section on the right, together with the sweep and other ac- cessories used in its making. In making the mould, the outside was made first, not l>ecause it is customary to do so, but because we had to luiild it up and dry it in n pit, from a lack of oven-room. At K is shown the bottom plate ; also a holding-down hook, of which there are four. The plate was set on a solid sand foundation ; and, in order to leave a pit below, a cast-iron ring F was used. This pit was required to provide room for a man to operate screws for centring the centre core, as will be explained farther on. In making the outside mould, there were five G" round blocks distributed so as to equally divide the circumference, for making vent-holes, as at H. At Y is represented a plate placed upon the top of the mould to stiffen and hold the brick- work together. After the mould was finished and blacked, it was then prepared for diying by laying four railroad-bars across, so they would rest upon the pit about 4" above the top of the mould. On top of these were placed sheet-iron ])lates, and the open portions of the pit, between the rails, were biicked up to prevent the escai>e of heat. Charcoal and coke were used for drying, the charcoal lieing on the outside and the coke on the inside. For the first twenty-four hours, there was a fire l>i.itional } iLW vf Mould MOULDING A JACKETED CYLINDER. 61 upou the outside only, because both fires would, at the begin- uing, have beeu likely to blister the mould. The coke fire was made in a perforated boiler-iron kettle, about 18" diameter and 24" deep ; the kettle having an open top and fire-grate bottom, and being let down until the top of the fire was about even with the bottom of the mould. The pit was originally some fifteen feet in depth, having been made for other purposes, and then filled up so as to be eight feet deep. The bottom of the small inner pit F was three feet below the bottom of the large pit, the diameter of the small pit being 42". The diameter of the large pit was 13 feet. Upon the bottom of the large pit, a boiler-iron curb was placed. This was to make the pit smaller at and towards the bottom, to confine the fire, and also to save work in ramming up. The distances of the mould from the pit, given in engraving, are not the actual ones, but are those it would be desirable to have for convenience of operations with such a mould. It may be asked, Why, if there is room enough at the bottom of the pit, should it not be made the same size at the top ? In answer, it may be said that the sand requires harder ram- ming at the bottom than at the top of a mould, and sand can more readily be rammed solid in a small space than in a large one. Besides, while it is practicable to ram the small space at the bottom, if this space were continued to the top there would not be room to work to advantage. In fitting up old pits for drying moulds, where a natural draught cannot be had, a blast-pipe may be laid all around the bottom, having a branch E passing up to the top, through which connection is made with a blower or fan. The 4" brick wall seen upon the outside of the blast-pipe E is the pit's wall. "While it is only shown as of a small height, it is, of course, to be understood that its depth is about the height of the mould. In firing up on the outside of this mould, six to eiglit bushels of charcoal were evenly distributed ou top of the blast-pipe /iT, Srctiunal ficur of CiUttmg RejuUi li ftj Scrrip ) ■^' Sectional Vino of Mould Fig. 21. {]2 Mon.DlNf; A JACKKTKI) ('YI,IN1)I:R. which had siiiall holes ])ore(l in it. The fire was started \)y throwing hot coals on the charcoal and pntting the Mast on. After the fire was well under way, the blast was shut off. The inonld was uncovered every twenty-four hours, and fresh fuel adiled, until it was found to he dry. In making the centre core of this mould, tiie sweep was in sections, so that parts could be detached as required. In commencing, the sweep was secured to the arms A and J{, as shown. The bottom plate having been levelled and centred, tile core was then built up tottoni plate, the core was adjusted to show equal space all around the riser-head 6'. After the core was centred, the space between the plates, as at 37, was carefully and solidly packed with brick and loam, as a safeguard against run-outs. At 3.S and .'{'J are shown a ])lan and section of nuts for the rciiiilating-screws. Tlie two views, .S'.', show a Itlock with a conical hole to allow the point of the screw t(j ni(jvc. thereby preventing throwing the bottom of the core out of tiie centre of tlie mould when clear of the print. After the bottom j<^int had been secured as described, the upper section of core from r.i was hoisted, and all the side chaplets 41 and 12 set. There bemg five cores to set so as to leave a thickness of iron between them all, the chaplets required to be divided equallj*. The bottom chaplets, shown at 40, were set in an iron stand, which fits into still another stand that is cast with plate 20. This plan makes the moving of chaplets impossible. The chaplets had f" stems, with plates 4" x G", and for each core three bottom chaplets were used. Four side chaplets were used for the back or each core, two of which are shown at 41 and 42. A half-inch bolt, as at 43 and 44, was used to hold the cores against the chaplets. At 45 is represented a vent-hole con- nected with cinders, and at 4G a tube -H" diameter, with the end tapered to lit tightly in the core vent-hole. The space //, be- tween the tube and mould, was rammed all around with sand to prevent the metal getting mto the vent-tube. At 47 is a [)late 3" wide, V' thick, and 18" long, placed m the core, when being made, to give support to the bolt-head 43. The space in front of the bolt-head was filled with beer-sand, and made level with the surface of the core. The cores being secured, the next operation was to airange for chapleting down the cores, as at 48, which was done as fol- lows : On top of each core, three clay balls were set. The upper jointed section of the centre core was lowered down, its MOULDING A JACKETED CYLINDER. 65 joint being at 49. This section was then hoisted, and cbaplets were made ^^" shorter than each elay ball measured. All the clay balls were numbered, and only two or three removed at a time, so as to insure against getting them mixed, as any blun- dering in this respect would probably result in losing the cast- ing. When all the chaplets were placed upon the exact spots previously occupied by the clay balls, and a little flour put on to insure their touching, the upper section of the centre core was lowered to its place, after which the riser-head S was covered with segment cores, as at JST. The runner-basin X was not covered over as represented, until after the core was dried. The segments covering cores 51 were dry when set ; but in order to dry out the course of loam and bricks at 50, the core was given one night's firing, to expel any dampness the course of lirick 50 might contain. At 52 is shown an iron ring, used in combination with 36 for wedging down the core against the high-head pressure. The pouring-basin had one runner gate, 4" diameter, leading to basin X, as seen at 53. The cylinder weighed a little more than eight tons ; there being flanges that are not shown, as they would serve to confuse the subject. The whole mould was secured by a cross-beam and slings, chains coming down to four hooks, one of which is shown at K. Enough iron was poured in at the bottom of the mould to fill it above 40 before any was poured in at the top. After the mould was poured, and sufficiently cool, the basin X was uncovered, and the basm iron broken up to assist the shrinkage as much as possible. The casting, when finished, was clean and without perceptible flaw. t;0 UMVOLVINC; COKE AND UNDEU-SUIIFACE SWEKPINO. REVOLVING CORE AND UNDER-SURFACE SWEEPING. Ik vol. i., p. 187, is an illustrated article upon "Sweeps and Spindles." The engraving shown is that of a rigging for nnder-loain surface sweeping. Loam cores arc often of such shape that some such rigging is almost inilispcnsahle. Having in our foundry a very siini)le arrangement, that is not only ada[)ted to under-sweeping, but to other purposes as well, and which is in some respects superior to the rigging previously shown, it is thought a description of its workings ma}' be of value. The advantage of this rigging over the common run of spin- dles could seldom be better displayed than in making the con- denser core seen in Fig. 23. The spindle is so designed, that after the core is swept it is then hoisted by the same spindle, as shown in Fig. 23. This spindle having a collar F, and a key- hol(! (r, Fig. 24, provides for securing to it any i)late or ring, as seen at // and A", Figs. 2.3 and 2G, "When the ring K is wedged up tight against the collar, by the keys 3/, the buildiug founda- tion of the condenser core is formed. The steel pin N being set in the step P, the spindle is then set up and secured by the top centring and holding-arm R. This arm is so constructed that it can in no way be sprung. The elevation and plan view of arm show its construction in outline. After the spindle has been secured, the next opera- tion is that of securing the sweep. In setting the sweep, an arm is bolted to steji P. as seen at V' (The cap of this arm is not shown, in order to show the BB Stationary Stitep and lUiolung Core Spindle Ste2> and Steadying JUgying REVOLVING CORE AND UNDER-SURFACE SWEEPING. 67 step and steel pin more fully.) This arm is set so as to be parallel witli the top of swee[)-liolding- arm Y. The manner of bolting the sweep to this arm is more fully shown at AA, in plan of arm. The bottom of sweep being secured by the bolt liB, and the sweep found to be gauged right, all is then ready for buildiug up the core, which is done as follows : Pieces of bricks being built up as high as No. 2, a thin cast-iron ring is then laid on, after which the core is built up to No. 3, the plate there shown being then laid on ; bricks are then built up, and, the two plates at No. 4 being set, bricks are laid up to the end of the core, on top of which is a ring No. 5. This is used for the purpose of bk)cking upon, to hold down the core when cast- ing. The inside of the core was filled with clean, small cinders, lightly rammed, as it was built up. The core, being completed, Avas hoisted up and lowered on a plate FF, Fig. 23. On this plate was set an iron ring BB: this was packed with sand. The bottom of the spindle, where it projects through the plate FF, was clasped by a cap having laps as seen at KK. This being firml}- bolted around the spindle, the core and its attach- ments were then hoisted and set on the oven carriage. While many m.ay never have such a core to move, the plan shown will no doubt be worth remembering, for the principle is applicable to other work. This condenser core is one which practical loam-moulders will concede to be rather a difficult core to make. Had the core been larger, ]^\e risks in making it would have been greatly lessened. The form of the casting made with the above core is seen in the section, Fig. 22. The outside portion of mould was jointed in two parts, at the respective heights A and B. The casting was run entirely from the bottom, the metal going in through two gates at the flange C. At D, upon the riser head, is a feeder. The situation of the gates will, of course, show that the mould was cast verticall}'. The dots at E repre- sent the print, which was swept in the mould for the print seen CS J:i:VOLVlN(i C'OUK and I'NDKR srUI'ACE SWKKriNO. on core lo set into. This fjiiidcd and centred the Ixtttoni. The top was held (•ciitrally !•}' ihri-c dtained. As a general thing, loam-cores are swept by having the sweep revolve. I doubt if there could lie found six foundries in the United States that do not follow this practice ; in fact, so far as I know, our shop is the only one that makes a practice of sweeping cores with stationary sweeps, as shown. The plan was established long before I ever saw the shop ; and, as I find it a good one, it is still used. The advantage of having the sweep stationary is, that the core is certain to come to whatever diameter the sweep is set to ; also, there is no raising or lower- ing of spindle-arms to clear the brickwork as it«is built up. as is often necessary when the core is stationary and the sweep revolves. Having cores come larger or smaller than intended, or one end not right with the other in size, is no uncommon occurrence. Having to change the position of the arm, as is often necessary with revolving sweejis in sweeping long cores, one is apt to move the sweep ; and as the brickwork is more or less between the sweej) and spindle, there is no handy means of ascertaining it. We^of course, can cali^ier the core after it is REVOLVING CORE AND UNDER-SURFACE SWEEPING. 69 swept up ; but to then change the sweep's diameter is often objectionable, for loam scraped off or put on in thin layers }nay cause surface scabs. I might sa}^, "• Sweep only two small spots to test the diameter, then, if found right, sweep up the core." This, in manj- cases, is a good plan, and should be practised when exact sizes are wanted. But, as a general thing, wlieu a moulder sets the sweep, he thinks of nothing but driving ahead ; and if the core is not found to be the right size when set into the mould, he often can easily make himself and others believe that the right sizes or gauges to set the sweep by were not obtained. In our shop, all cores are calipered with long, wooden-legged calipers, simply to make sure that our gauges or measurements were right when setting the sweep to sweep them. Our presi- dent, J. F. HoUoway, is very particular in knowing that all parts have the thickness the drawing calls for ; and, did they not come so, an intelligent reason would have to be given. Did the receivers of swept-up loam-castings know how often the intended thickness is increased or decreased in the castings tliey receive, they might be surprised. In jobbing loam-work, no one can, da}' in and day out, sweep all his moulds so as to measure to ^V of what the draught calls for. As little as -^^" off or on a thickness is but a small matter with the general run of work ; but when it comes to adding or subtracting from \" to •|", the value of establishing ways of insuring correct thickness is seen. "While the cores generally need the most attention, the outside part often requires measuring to insure correctness. A good plan to insure the thickness wanted is, to take the size of the first part swept upon a narrow stick when the mould or core is finished or blacked ; then, when sweeping the second part, gauge the mould or core, as the case may be, by the meas- urement taken from the lirst part. By this means, if one does not get the first part the size called for, he has at least a chance to insure obtaining the proper thickness. 70 UKV()I.VIN(i CORK AM) UNDKU-SUUl'ACE S^V1•:K^I^•0. A third a(lvanta«;c the i)hin of swoepinp; shown has ovor the R'volviiiii swci'|) is, that the inouUhT can stand still ; therehy savin;^ hihor and the making of a eircus-pedfstrian of himself. A straiitjer to this plan would he snrprised to see with what ease heavy cores can be revolved. Cores as large as nine feet in dhimeter, and seven feet high, have been swept here by revolving them. For heavy or high cores there are two plans shown in Fig. 25, one of which it is sometimes found necessary to adopt in order to steady the core when l)eing swept np with loam. One of these is to use steadying-bolts, as shown, or swivel-screws. Three or four such bolts or screws may be used, niiiniiig from the bottom ring up to a top-steadying flange, as shown. This is a good plan to adopt for cores of large diameter. The brickwork seen inside of these bolts is to rep- resent a' high core being held steady In' a top brace bolted to the spindle, as seen at /S, and then wedged. Such braces are generally required when cores 18" to 72" diameter are longer than four to Qve feet. It should be remembered that the braces are not used during the bricking-up of the mould, but only during the rubbing-on of the loam. It keeps the core rigid, so as to assist in its being swept true. The size of the spindle given, Fig. 25, is that for cores ranging from four feet up to nine feet diameter. The spmdle at Fig. 24 is for work under the above sizes. The floor-level and pit marked shows how we use the arrange- ment. A pit of the dei)th shown is very handy for our general run of work. The diameter of the pit being about ten feet, there is room to walk around. The mud and bricks are kept on the floor, so that, in flrst starting to build, there is no stooping clown to reach material. Then, when the core is built two or three feet, the pit is readily planked over to enable reaching np higlier. The pit was originally made in order to jirocure more height for hoisting : of course, where the crane is high enough, one could dispense with the pit if it were desirable. Before REVOLVING CORE AND UNDER-SURFACE SWEEPING. 71 removing the top arm for placing on plates, or to hoist the core, it is generally necessary to have the under side of core blocked. For this purpose wooden horses come very handy. The one seen at Fig. 30 will be suggestive of how they may be placed. Fig. 27 is a plan view of //, Fig. 25. The long arm X is attached, simply to show that the same rig can be used for larger cores, by extending the lugs. It was by such a rigging that the centre core shown in the article upon "Casting Whole or in Parts," p. 5G, was made and lowered in the mould. When the core was trued by the set-screws there described, the keys at TT were taken out ; and after this, those at W, which let the plate H fall down to the bottom of the pit. The spindle, now being released, was hoisted out, and the hole in top of the core filled up as there described. For that job the 3" spindle was the one used. 72 SWEEPING GllOOVEU-CONK URLMS. s\vi:i:rixG orooved-coxe diu'.ms. TiiK inac'hino sliowii (p. 73) is intc'ii(U-d for swc'cpiiiji cither right or left liaiul grooved drums of cyliudrieal, couieal, or curved shape, and of any pitch desired. The originator of this device, S, B. Whiting, M.E., of Potts- ville, Penu., first used it in ISGT or I'SGH, since which time a great number and variety- of grooved drums have been reported as made with it. At the right arc sections of what were no doubt very large cone-castings to make. The one in Fig. 32 was of twelve and twenty feet diameters, with a height of about six feet. Our trade is under obligations to Mr. Whiting. Among our best moulders, but very few have any knowledge or idea of cone-drum sweeping ; and upon reading this many will, like the author, feel like tendering Mr. Whiting thauks for allowing the publication of his device. The engravings are from photographs taken from a model, therefore the proportions will differ somewhat from those of a full-sized working machine ; and, while to many the three views will give a clear idea of the machine, there are those for whom it might be well to give a detailed descrii)tion. A, Fig. 33, is -a spindle that is held stationary in the base. Fitted to work upon and around this spindle is a sleeve A". To guide and hold the arm E at a right angle to A", is the cross- head centre P. The arms R and X are firmly secured to sleeve A", and therefore will cause the latter to rotate around the spindle A when operating the machine. The cross-head P slides up or down upon the sleeve A", being controlled in its motion by the screw D. The bar E (carrying at its end the former or sweep F) slides in the cross-head guides SS, and is controlled in its movements, lengthwise, by the bar or former SWEEPING GROOA^ED-CONE DRUMS. 73 r, which may be set at any angle, and may be straight, curved, or of au irreoular form. The gear H being fast to the spindle S.Ji.\yhitiny^s me Drum Sweeping Macliine A, the screw D will be turned whenever the bar E is swept around the mould. By changing the gear on the screw D and 74 SWEEPING GROOVED-CONK DUTMS. spindle ^1, iuiy pitch may l»e obUiim-d ; and, liy inKcrtinjx t-ithcr one or two inteiinediato {rears, a riglit or left hand pitcli may be ol)tained. The opposite side-view of Fig. IV.i is shown in Fig. 34. As will be seen at iV", the bar E is there guided, as well as at SS^ Fig. 33. Fig. 35 is a plan view of the interme- diate and principal gears. As there are four wheels, the nse of the two, 1' and il/, may not appear plain. These gears ( Y and M) neither increase nor decrease speed, but are simply for making cither right or left drums or pitches. Were wliccls required for one-hand sweeping only, then these gears would not be required, nnli'ss the centres A and V were so far apart that they were neccssaiy to transmit motion. As shown, the intermediate gear J/ is the one engaged with the large wheel //, and will produce a right-hand drum or pitch. To produce a left-hand drum, the thumb-screw ir, Fig. 34, is loosened, and the gear Y is made to engage direct with the large wheel H and pinion V. The screw D should, according to the diameter of the mould, be set to balance the bar E, in about the relation to the centres between the sleeve K and bar T, as here shown. In other words, the screw D should be set so as to balance, at an aver- age, the bar E in its up or downward movement. When arranging gears for moulds of large diameters, the gears Y and J/ could, to save using a large centre-wheel and pinion, be reversed so as to stand between the pinion V aud large wheel H. To make an opposite hand drum, the two pinions would require to be replaced by three smaller ones. For making small-sized moulds of right-hand pitch, only the gear //and pinion F may be required. In liguring the relation of gears to give desired-sized moulds, the pitch of the leading screw D will be the regulator. As an example, we will suppose the leading screw D to have ^" pitch. (IJy pitch is meant that every time the screw revolves once, the thread would cause a nut to rise in height i".) Now, suppos- ing there was to be a cone made having a 2" pitch, as seen in SWEEPING GROOVED-CONE DRUMS. 75 section, Fig. 36 : the arrangement should be such as to cause the sweep F^ Fig. 33, to rise in height 2" every time the sweep revolves once. Knowing that the leading-screw rises \" ever}' revolution, the gears must be made so that, in every revolution of the sweep, the leading-screw will revolve four times, in order to raise the sweep 2" in one revolution. Now, knowing that the leading-screw pinion V must revolve four times in order to raise the sweep 2" in one revolution, it will be seen that the large wheel H must contain four times the number of teeth that the pinion has ; therefore, if the pinion has, say, twelve teeth, the large wheel must have forty-eight teeth. Did one wish to make a mould having grooves of 3" pitch, by using the same ^" pitch leading-screw, the gears V and H would require changing so as to cause the pinion V to revolve six times to once of the sweep. To construct a \" pitch with the above leading-screw, the gears would require changing so as to cause the pinion to revolve twice to once of the sweep. For the construction of any fraction of the above pitches, the gears would, of course, require proportionally changing. The pitch for the leading-screw would, for general work, be better if \". The ^" pitch could, of course, be used, but such a coarse feed for fine pitches is objectionable. For grooves above 2" pitch, \" pitch leading-screw would be best. The spindle A is not necessarily secured in such a base as shown. The idea is simply that it must be firmly held in some- thing that will remain stationary. While the spindle is shown here self-supporting, it would be better for general work were the top supported by a brace. To do this, the spindle would require to be prolonged farther above the wheel than here shown. While there are no sizes given, any one requiring such a machine can ver}' roadil}', from the views and doseription, pro- portion and construct such a machine as the size of a job may require. 76 SWEEriNO GROOVED DRUMS IN LOAM. SWEEPING GROOVED DRUMS IN LOAM. Tin: two engravings, one on p. 77 and tlic other on p. 79, each reprcsentnig a different plan of sweeping a large grooved drum in loam, are not only instructive in so far as they repre- sent practical processes, but are interesting, in connection with that shown on p. 73, as showing that the trade of the moukk-r demands the exercise of talents of a high order. I am indebt^-'d to the courtesy of David Matlock, manager of the I. P. Morris Company's foundry, Philadelphia ; and Homer Hamilton, of the Hamilton AVorks, Youngstown, 0., — for being able to present to mj' readers these plans, which are well worth the consideration of practical men. In the plan ado[)ted by Mr. Matlock (Fig. 3H), a spiral groove, as shown, is cut for nearly the entire length of the si)indl('. The pitch of this gioove is made the same as the pitch of the grooved drum is intended to be ; and a set screw projects through the arm of the sweep, and enters tlie groove in the s[)indle. Of course it is plain, tiiat, in revolving the sweep, it will have a corresi)onding spiral movement. Mr. Hamilton's plan (Fig. 37) involves the use of a plate B, the working-face of which is turned up in a screw-cutting lathe to the desired pitch. At F is represented a i)iece about S" long, dowelled to the main plate so as to be readily removed. The roller at E permits the sweep to be easily* revolved. Tlie drums, for the making of which this plan was originated, were 14 feet in diameter, and 7 feet 4" in lengtii. They were stiff- ened b}- inside rilis and flanges, and had G" outsidi- flanges at ends. They were poured by dropi)ing the mdal from the top, and, when done, were said to l)e liist-chiss castings. SWEEPING GROOVED DRUMS IN LOAM. 77 I will not dwell upon forming the inside, of di'unis, as Uiat is a matter of secondary importance compared with sweeping the outside or groove portion of drums. Fig. 37. In the loaming or sweeping-up of a mould, a straight sweep is generally lirst used ; after which this is detached, and the 78 SWEEI'INf; fiUOOVKI) DIUMS IN LOAM. s\Vff)i for foiiiiiiii^ tlif grooves :itt;iclu'iti(»ii of the swrcp is shown in lilm-k. TliLs p'>rtion of the s\vt'('[) could lie nuuli" cf slu-i-t or hoilcr iruii phitcs, ms tliL' proji'itions iirc very ciisily broken if of \v()o«l ; or lliis por- tion could \xi wootl, faced with a thin sheet-iron plate, us represented at F (Fig. 3late is not so placed until after the mould is roughly swept up with the straight sweep, which is done by letting the sweep rest and revolve upon the collar A, which, as now seen, is dropi)ed out of contact, in order to show the position of the sweep when forming the grooves, the incbned plate B having been lowered to the bottom holding-step T. Then, after the straight sweep has done its work, the inclined plate is raised to its proi)er ix»si- tion, and held by the set screw shown in B, after which the collar A is dropped out of contact, so as to allow the sweep to travel in a spiral direction. The sweep starts at F; and when It has passed away from F this piece is removed, allowing the sweep to travel more than the whole circumference of the grooves. The sweep is then returned, F being replaced, and another revolution is made ; and so on to the end. This plan causes the sweep to be turned back over the same surface every revolution it makes, and is very objectionable, for it is apt to tear and rough up the surface of the mould. To avoid this, the dowelled piece F can be left out, and, when the sweep comes to the step, let it drop down upon the starting-point of the inclihe. To do this, there will of course be left a narrow strip the entire length of the swce[), that cannot have the grooves formed : as this strip needs to be but V\ or such a matter, wider than the SWEEPING GROOVED DRUMS IN LOAM. 79 sweep, it can very readily be filled up after the balance of the grooves are finished ; and then, after leaving the starting-point, the piece F can be set so that when the sweep gets arouud, it can be made to travel more than the whole circumference, thereby sweeping off that portion or strip of the grooves which was filled up. By this plan the sweep is always travelling in the same direction, and the little strip to be filled can be swept 80 S\VKKI'IN(; (iUOOVKl) DUTMS IN l.tlAM. with one rc'VtjluliDii if llii' jolj is iiiU'Hijicnlly |)Lrfunin,'tl. Of the two plans, llit- latter one is tlec-idcdly the best to adojtt. In loainin^ or sweeping up the irrooves, tlic method adojitcd should depend ui)on tiie size or piteh of the grooves, and also upon the nature of the loam. If the grooves are not over |" deep, and the loam a fair stirfening kind, the grooves may ])e swept up without much delay. ]>ut should the loam ])C a slow stiffening kinin tlu* assmiifd diaiiittiT to llic top of tlic |>itcli or aiijilf. In iiiakiii*^ :i tciiiph't to <)l)taiii tla- aiij^li,' for any (It'sircil l)la. in l)uil-k. 'I'lif cmU of llic scant liiij^s ri'sU'd upon tliu handles of llit- llask, \\lii
  • , Fig. 41, As a further precau- tion against the loss of a casting by the crushing of dry-sand moulds, it is often advisable to close them together, bolt or clamp them, then hoist off the cope, and examine the mould before casting them. In moulding cylinders in dr}' sand, this is especially to be commended. In some cases, I have the mould closed before any cores are set : then, with a lamp inside the mould, the joints can be plainly seen, and felt if uneven or overshot ; the thickness of fin can also be noted, and, if found too thin, can be cut out when the cope is raised up. Before leaving the subject of dry-sand joints, there is another point worthy of note. After the mould has been blacked, the joint should be washed over lightly with the same blacking, thiimed with beer or molasses-water to a degree that will just blacken the joint-surface. The same treatment should be given the core-prints ; as it not only makes a better-looking mould, but by forming a hard skin prevents such parts from crumbling away when being brushed or handled. After blacking the joint, the}' should be sleeked to level down any lumps that may be on the surface. Often the blacking of the joint is neglected, or it is blacked with thick blacking, thereby increasing the thickness, making it liable to be crushed. Again, the joint will be wet until it is little better than a bed of mud, therel)}' getting it out of all reasonable shape. Either of these last-named operations is likely to bring about bad results, A consideration of the joints of loam-moulds embraces more than can be covered with a single article. Some of tlie features connected with this subject have been referred to in previous loo ClUSlilNCJ AM) 1 INNIN*; rASIlNf;S. articles. :in(l a few points iiiiiy lie niriitioiud here. The joints of loMiii-nioiiltIs slioiihl !»(' liniMil ii|ioii the .same Lrtiicial priii- cipii' as those of div-saml inoulils, l»iit tlic tiniiinj^ is acconi- plislic'd ililTeiviitly. Instead of sleekitiij down tlie joints, they are y parting: tiie joint in this way, the fin can generalU' he chipped olT, and the surface smootiied with a file, so as to present scarcely any sif^ns of tile joint. 'I'his plan also provides for hiding overshotues.s, which will often occur in loam-castings. Tills bead is generally applicable to joints formed by a sweep, and for irregularly formeSometimes sections of loam-moulds arc swept independently of each other, and closed together in get- ting read}' to cast. P^or such moulds, the sweep can often Itc made so as to form the fin. In this way, the fin will be quite even in size, giving the easting a symmetrical appearance. Fins, at their best, are not an element of beauty, and the re- fined moulder will study to see that they mar the general ap- pearance of castings as little as possible. MAKING AND VENTING CORES. 101 MAKING AND VENTING CORES. The subject of cores is a very important one in its relation to tlie production of good castings. Some years back it was cus- tomary in many shops for the moulders to make their own cores ; l)ut at the present time core-making is coming to be a distinct branch, at least as much so as loam, drj'-sand, or green-sand work. To excel in core-making is a credit equal to that of excelling in any otlier branches of the moulder's trade, as the success of the moulder in i)roducing good castings is often largely dependent upon the skill of the core-maker. Take, for example, the casting of a steam-cylinder. As a rule, there is not much to fear with the outside. The main risk is in the cored parts. Cylinder cores, as a rule, with the exception of the centre core, ma^^ be classed as thin and crooked, and arc much more difficult to make than larger cores. The question is often asked, How thin is it practicable to make and use cores? To answer this, it would be necessary to know the general shape of the core, and its position in the mould. A much lighter core can be used if set vertically than if it must be set and cast horizontally. AVhen set vertically, there is a much better chance for the gas to escape ; as it is not so suddenly covered with iron in pouring, and there is no great lifting strain on a verticall}' set core. The greatest difficulty to be overcome in making thin cores is in rodding and venting them : in fact, here is where the greatest skill in core-making is required. The chief cause of gases in cores is in the materials of which thej' are made ; and the least gas there is in a core to be got rid of, the better, other things being equal. hlli MAKIN(J AND VKNTlNfi COKKS. A laifio amount of fins rosiilts from llio iiso of Hour. Tlic ^usc'S from slroii^ly-imule Hour corfs, in u siiuill Icjw-roofed shop, often rciiiU-r it uiitoiialilc. Kosiii evolves Itiit little jras ; and hence in many eases its nsc is desirahle, particularly as it vents and ignites easily. Kosin is comparatively hut little used ; one rciison being, no douht, that it requires to he pulver- ized, and another that it requires more care to use than most material used for the purpose. Kosin cores will rarely admit of handling when hot, and are not so relialile for heav>' castings ; also the}' cannot be so firmly anoc iiiailc. Imt rallicr to sliow dilTi-rciit i)l:vns that have Ik-cii and may I't' piai-tisfd under dilTereiit elr- cunistaiiccs. When the ihawinirs of !i cylinder come into the pattern-shop, the pattern-maker should ccjnsult with the moulder as to the l)est way to make the pattern and the core-hoxes, to the end that it may be safi-ly and expeditiously moulded. In the engravings are shown three plans for making jxirt- cores. The fust one can often be used in making small cylin- ders, to the saving of work by the moulder, and increasing the prol)ability of good castings. The ports, exhaust and steam- ehcst cores, cannot always l)e made together, as sh(jwn, but very often they can be. The second plan, in which part of the core is swept up, is very handy for the moulder, as well as simple for the jjattern- maker. It gives the moulder an opi)ortuuity to see what he is doing, and saves him time and labor. The third plan, of having a full box, is one often resorted to where the cores are quite crooked and irregular, but should seldom be resorted to where the second plan can be employed. In the engravings are shown four plans for rodding cylinder cores. The lower right-hand cut represents the making of cast- iron rods. This is done by taking half the core-box, and l>ed- ding its face into the sand, which is made solid for the purpose. The box is then withdrawn ; and, by using a gate cutter, a frame similar to the welded core-iron shown is made. For a cope or covering, heavy paper is used, laying it over the face of the joint. Sand is then packed on the paper, and boards and pig-iron i)laced to hold the sand down when the frame is poured, which is done through the i)ouring gate, as represented. Irons like this are readily made, and are often good for short, thick cores. MAKING AND VENTING CORES. 105 Ou the liolit a,rc roproscntiHl wrontilit-irou rods, rcatly to have cast ou them a uarrow phite of etist-iroii. This is for tlie pur- pose of hohhiig the rods in their proper position at tlie print end. Single eross-rods are used, when making the cores, for hohling tlic other end together. For getting off the vents, holes are drilled or cast in the plate, as shown at 1, "2, 3, and 4. At X, X, are shown two views of a wooden support, cut to the circle required, and having notches for the number of rods required. This is used for holding the rods in proper position while the plate is being cast on them. This kind of rod is often good for very large, thick cores. Another plan, HOOK. Fig. 44. ■which is often better than the one above described, is as seen by Fig. 44. Here, instead of the wrought-iron rods being cast in or held li}' the print end, they are all held by their centres ; which, besides making them stitif, presents a " core-iron " easy and simple to make. The welded core-iron shown is one commonly used. Some- times, instead of welding the wrought-iron rods, they are riveted together. Frames of either kind make reliable rods for thin cores, large or small. The objections to them are, that thejfe are costly to make, and that removing them from the casting is somewhat troublesome. At D and E are shown two forms tliat are used for fastening wire or bolt hooks to them, to securely hold the core in the mould's print. The one at E is made by simply flattening the end, and drilling a hole through it. lOG MAKiN'i AND \"i:NriNf; (•f)iii;s, Tlic liflli Mini l:is( |i|;iii rcpicsciils tlic uso of siiiLrlo uxls, sottiiiu t lii'iii in llic core ;is il i.-, i:iiiiiikm1 up. 'I'lic ciil shows the iMopcr innimcr of pl:icin^ llic rods, Wciv tlic lon«; rods l:iid so as to have tlio f-ross-rods *S', scon in soction A 11, on the other side froiii Ihat rcprosenli'd. when the core was fas- tcMicd hy tlic liook I), in the print, tliere woidd he daiiixer of cracking the core. AViun made as shown, pnlling on the hook draws the wliole core with it. This plan I have employed for the port cores of large ni:iiin(^-engine cylinders. In one shop, where the custom was to weld the rods together, I started to make a set of lai'ge cores in this way, and was told by the fore- man that he had never seen large cores so made, and that he did not think the plan a safe one. As I had made larger ones in the same way, I argued him into permitting a trial. From tliat time on, there were no more welded rods used ; and the new plan saved in the neighborhood of live dollars on every cylinder casting. Regarding the size of rods to be used for such cores, llie thickness of the cores and a consideration of getting olY tlie vents, etc., must govern each particular case. The larger the rods, the stiffer will he the core ; but they should he no larger than is necessary, for it is more dillicult to get laige rods out of the casting than it is small ones. If no vents wore needed, the making of port cores would be much simplified. ]\Iore, generalh', depends upon the vent than upon any other feature. The vent-rod and rope shown repre- resent the two plans commonly emploj'od for venting thin, ci-ooked cores. By using the rod, a cleaner vent is usually insured than by using the rope. The rope is reliable, but requires vindi more care in its xise ; which not always being received, the lial)ility of failure is increased. The use of the lods calls for the most work. When rope is used, nothing more is reijuirt'd after the core is diy ; but with rods there are connec- tions to be made, as at /i", to make the vent continuous. To MAKING AND VENTING CORES. 107 ca of the way to secure such cores is shown at II and F. Before permanently setting the core, it is set in its print to see if the side 11 will form a close joint along its entire length. Should it be found not to lit pro[)erly. it should be made to do so, either by liling or by building up the print with Fig. 44a. SECURING COKE-VENTS. 109 loam or thick Mackino;. "When built up, if the hnilt-up pnrt is more tbau y thick, if the moukl is not hot, it should be dried, cither by using a hot iron, or in the oven. When ready, paste may be applied to the side that cannot be readily seen, as B. The paste should not be more than ^" thick, as, if the core is properly fitted, this is ample. Sometimes it may be advisable to put the paste on the mould ; and, again, it may be better to put it on both the mould and core. If both mould and core are warm, it is better to apply the paste to only one of them ; as by dividing it between both it makes a thinner body, and is likely to become dried before the core gets set, so as not to form a proper joint. Where both mould and core are cold, by dividing the paste between them there is not so large a body from which the print end of the core will absorb moisture, thereby weakening it before it is set in its print. In my prac- tice, I always endeavor to have either the mould or core warm, to dry the paste, when the core is set. A good mixture for paste is flour wet with black wash, which may be cither thick or thin, according as thick or thin paste is wanted. Clay wash may be used instead of the black wash. Both are good to keep the metal from burning away the paste, and finding its wa}' into the vents. AVherc there is danger of the paste coming to the surface of the mould, in places where it cannot be seen or come at to scrape off, I prefer the black wash, as it is not so liable to blow or cold-shut the casting as the clay wash. For setting cold cores in green-sand moulds, I prefer to wet the flour with machinery oil, as there is not so much danger of chilling or generating steam as there is when the clay wash or black wash is used. Rye flour is the ])est for i)aste, as it is not so sticky or clammy as wheat-flour paste. As regards the thickness or consistency of the paste, it depends somewhat upon circumstances ; but, as a rule, it should be so thick that it will not flow. 110 SKCriUNfJ CORK VHNTS. AiiotliiT iiii|i(iiiMnt iioiiit in m.-iUiiiL: :in()iil twice that of tlic jtiiiit. Tiu' face (•(iuc (if llic plates can ln! i\ept }" away fioni tlic pattern. They ail' generally phiccd upon a little sand sifted on it, the plates iieinii ^vet with elay wash to make tlie sand stick. In the larirc enirravinj; of the section of a hcltcd cylinder inonld, two pinns of sccuiin sliow openings inadi" ill the llask opposite the exhanst and port-core prnits. ^\'hen the cores are all set, and the outside joints donhled np, the oiien space T is rainnied n[» with sand. After the sand is raimiu'il up to the lir>t row of vent-holes, they are cleaned out, and short vent-rods pnt in tlu'ni ; then more sand is raniini-d, and the second row of vent-rods placed, and so on till the joint is reached. (These rods can all l)e placed hefore commencing to ram up if we avoid hitting them.) When the cope is on, the vent-rods are placed, and the sand is rammed throngh the opening i)rovidcd at N. Sometimes circumstances will not admit of the vents being taken otT through the cope. In such cases it is often advisahle to connect the upper portion of the core-vents with the lower, as represented in the top and bottom parts of the belt-core. (The bottom half of the l»elt-core is shown all black, so as to prominentl}' show the line of vents.) Where the cores are not made in halves, this is somewhat difficult to accomplish ; Init by referring to the article "Making and Venting Cores" ([). lU(j), the plan of connecting and venting such cores will be understood. At J* is shown a iilan foi- taking off vents, which we niav lie obliged to use when tlie llask is not adapted to the job. or when the i)allern is not jiroperly made. A\Tu'ne\'er the joint of a mould must lie raised abo\-e the joint of the llask, as ;?hown at 7t, the more room tlii're is to mount a strong body of saiub the more likely it is to keep its form and to su[i[Miit the cores, etc. Tlii^ plan of moulding is SECURING CORE-VENTS. 113 not al^vaj's objccliouable ; but unless there is room suflicient for the vents to be cared for within the body of the mould itself, and then all led off through one opening, there should be open- ings in the flask, as shown at Y, 2, 3, and 6. Taking off vents through the joint of a dry-sand flask, as shown at P, is by no means reliable. It would be much better to drill holes through the sides of the flask for the purpose. When the face of a cylinder is moulded as here shown, it is often advisable to have the cores form their own prints, similar to 7, 8, and 9. This plan saves work in rodding, and labor in setting and securing the cores. The exhaust core-box shown is a handy one for cores made in halves. XX show the sweep being used for striking out the circle portion. (aiEEX SAXl) MOULDIXf;. CASTING FINISIIKI) WORK HORIZONTALLY. In ;i previous arlic-lo is dcsfiilud tlie casting of a hydraulic lioist iu dry sand. In this the manner of niaivin<^a lonfror one in green sand will be deserihed. It may l)e asked, wliy, if a casting about twenty-three feet long could be made in green sand, one a]»out fourteen and a half feet long could not be made in the same way. It would be uureasonablc to expect that siicli eastings will be as perfect cast horizontally in green sand as if they were cast vertically iu dry sand. The longest one would have been cast vertically, had there been a foundry near by that had the facilities for doing it. The casting had to Ite made; and, as no one would cast it vcrticalh', some one must do it horizontally. The Cuyahoga "Works was selected to d(^ the job. The castings, when finished up, presented a very creditable appearance for green-sand work. The upper cut shows the dimensions of the casting, the dotted lines representing stock allowed for holding dirt and for finish- ing. The length of casting as taken out of the foundry was twenty- four feet two inches ; eighteen inches of which, as sliown at the gate end, was added as stock for holding dirt, and was cut off in finishing. The thin rib (31) was cast on as a dirt riser, and was also cut off. No. 30 was a flange used to assist iu the moulding; and. as it was a good thing to attach the pouring gate to, it was left and cast as shown. The reason for pouring the casting entirely from the end tliat was nyt to be finished was that ftuch gated cfistiDf/s Kill be the dirtiest at the gate end; iu fact, such gates, as a general thing, do not distribute dirt only in Ihe section io whieh tliey are attached. If one were to 114 CARTING FINISHED WORK HORIZONTALLY. 5) 11; I H» CASTINti FINISIIKI) \V()l:K IK )K I/,()N T.M.I.V. cast ail (>iK'n sand Mock liasinii such an iiiulcr LTalc, he would uitot likely sih' ucaily all llic \t'i' the uaU'. The ai-liou of the litjuid nielal foi'uis :i wliiil- jiool. as it wen- over the Ljate, tlierchy prcveiitiiijj; the dirt from llow iii^ away wilii the inetiil. A.s a fjeitend thiiKj, the portions fartltcsl from (/dies trill he the cleanest parts of (i cdslinf/. Tlieir cK'auliiiess will (k'|)end much upon the style of trate usi-d, etc. To fuilher discuss the iiii[>(iitan1 (|iicsli()n of piopcilv uatiu<_j moulds, the small cut, Fi^. 1(1, is ;^iven. At E is shown iiii under uate, similar to the one in the large engraving. The airow represents the How of metal. The core shown has pre\enteil portions of the dirt fiom rising to the top of the co[)C. At // is shown a style of gtiting that will not conliue the dirt to the Fig. 46. gate portion of the casting ; and, in fact, to correctly foretell where the greater portion of the dirt will be collected, is often a dillicnlt task. Such gates are distributors of dirt ^ while such as the one at E confine it. This is a point that must be eonsiil- ereint ; for, if any metal sliould get in, you niJLilit ex[)ect a ''blow-up." In making the vents in tlie centre of the halves, as shown, instead of at the joint, as is generally done, there is less risk ; for, if iron does tind its way to the joint, it can do no harm, the parts where the connections are, of course, being excepted. - When the cores are butted together in the mould, two pieces of -J" gas tubes, T and K, are placed, the cavity for their in- sertion being cut out in making the cores. An end view of the cavity and tube is shown at section through S, K. After the tube is inserted about one inch in each vent hole, the rest of the cavity is carefully tilled up with new moulding-sand, whiih, if wet with beer, is all the better, as it will air-dry more solid than if the sand is dampened with water. The tubes are better for having a few \" holes drilled in tiieui, as tliis will allow any gas in the green sand used to escaiie. Before smoothing off CASTING FINISHED WOKK HORIZONTALLY. 110 the green sand, it is well to vent down to the tubes with a fine wire ; the holes at the surface being well closed, the green sand is then oiled over. The balance of the work is treated as is commonly done. The cut of pattern, skeleton and mould, is shown wider than its proportion to length. This is done to give a better chance for figures, etc. The cut of the casting is proportionately sho\YU. IJO IIKAVV AND LIGHT WUUK SKIMMI.NO-GATKS. iiKAYY AND LIGHT WORK SKr:\r:\iTxn-r;ATr.s. As a supplement to the previous chapter, '• Casting Finislied Work Horizontally," the following will be found an interesting and valuable addition, in whieh Figs. 47, 48, and oO are plans for skini-gating heavy work. When one has from ten up to thirty tons of iron to pour into a mould, conditions in gating will seldom permit the use of such skimming-gates as are generally used for ordinary work. In pouring ten tons or more of iron through a gate into a mould, there can be no dribbling process allowed. The iron generally requires to be got iuto the mould as quickly as practicable. Figs. 4 7 and 4'J re|)resent plans of gat(>s which I have used on heavy woik with much success. While they act as skimnu'rs, there is n(;tliiiig to prevent their letting in the iron about as fast as if there were one direct gate from the basin to the mould- entrance. In Fig. 47 the metal runs down A, passing through JB to D. From D it goes through E to the mould. Fig. 48 is a plan-view of this gate. It will be seen that the gate B is so placed that it sends the metal iuto D upon a whirl. The inlet- gate E, being higher than B, as shown, admits of a good whirl being generated before the metal rises up to E. The inlet-gate E, if desirable, could be on a level with or below B. The best Avhirl is created by B being below E ; as, when upon a level witli E, its opening destroys part of the circle, thereby not permitting of as good a whirl being created as if the circle in front of B were complete as shown. In Fig. 4'J the metal passes down // to K. and from K to F. if and i^ are siin[)ly one straight gate ; the portion between // HEAVY AND LIGHT WORK SKIMMING-GATES. 121 and R ho'iivr as deep again as at F, where it runs into the mould. A" being deeper than F, and having the riser R at its end. gives a chance for the dirt to be kept up above F; thereby allowing, Fig. 49. Fig. 47. Heal'!/ JTork Shimming Gates, after the start, clean iron to enter the mould. Tlie farther apart the uprights H and 7^ are, the deep part /l, of course, being extended also, the better the chances to catch and hold 122 III.AVY AM) I.KMIT WOItK SKI.MMlNf;-OATKS. the iliil. 'riii> jilMiiis iiol rccoiiiini'iKlfil as l>(iiiL( as v admitli-d hiU) till' iiidiiM. I'spi'cially iiiioii the start. The ^ate sizes j^iveii aie only to present some idea as to their niativr i)rop<)iti()ii ; fo> instanee, IJ heini^ 7" diameter, will admit of the '.)}/' diameter tiatc vl, ereating a good whirl, and also gives JJ plenty of room to hold dirt. For [)raetieal work- ing the monlder will, of course, have to use his judgment as to the size of the gates in applying them to the conditions t(I whirl is foniiod, it will drive :iii(l hold the dirt in tlic oontrc ui JI, tlu-rehy luvvi-iitiiiii; it from cntcriiij^ thi' oiith-ls which connect with the horn-gates. The re.'ison for nsiiiLj the horn- gates is, that by their nse there is not such a direct current caused as would be were the gate level from // to the mouhl, whieh can, of course, he used with this skim-gate if it is so (h'sired. The less currcnt-iiilluences side-gates exert from //, the more whirl there will he, which is the main success of such a style of skimming-gate, A, N, and G illustrate the pouring of several pieces from one horn- or branch-gate, while T shows only one piece being poured. The bowl // is best formed by having a pattern rammed up when making the mould. There might be several sizes of such patterns, so that one could use the size best adapted for the piece or pieces to l)e cast. Did one wish to further increase the utility of the skimming- gate, the whole thing could be formed by i)atterii. as per sketch Fig. .01, The holes seen at each end are simply for the purjjose of holding and guiding the upright gate-pins, P and 7;?, while ianin)iiig up the cope. Several different sizes of these patterns could be made, either of wood or iron. If of iron, they could be cored out so as to make then) light ; and not only could they be used for forming the skunming-gate in the uowel, but in copes as well. The branch lines at B show where the outlet from 7/ should l)e cut. These outlets could, did one desire, l)e made as part of the pattern. As there shown, the whirl will be better preserved. While in Fig. oO two outlets are shown, cut from H, it is not advisable to do so if it can possibly be avoided ; for the reason that the whirl in 77 will be greatly lessened therel)y. With re£»i-ence to the proper proportion of such gates, ideas aie given (on p. 101, vol. i.. and on p. 17 of this book) with which most readers are no doubt familiar. Any shop that has a line of small work which reciuires to be IJnislu'tl up should in some form or other have skimming-gate HEAVY AND LIGHT WORK SKIMMING-GATES. 125 patterns, not only for the purpose of saving labor in cutting the gates ; but, as ever}' practical foundrynian knows, to leave the cutting of skinHuing-gates to the judguieut of most mould- ers produces a gate which is far from being a cleauer or puri- fier of metal before it enters the mould. The author's attention was lately called to a good thing in the line of a light-work skinnning-gate, patented by Richard Cross of Cleveland, O. The principles of the gate embody several good features worthy of notice and study. The gate as seen shows a side and a plan view of the pattern. They are made of differ- ent sizes, ranging from one suited for pouring a five-pound casting up to one for a casting weighing a thousand pounds. For heavier work two or three gates could be attached to a mould if desirable. The gate pattern is made of cast-iron, the inside being cored out so as to make them light for ease of handling. In using the gate, set it upon the mould-board in such proximity to the pattern as may be desirable. There are right- and left-hand gates, so as to still increase their utility. In ramming up the cope, the pouring-runner gate is set at the end of B. The cope being lifted off, and the pattern and skim- ming-gate drawn, a connection from K lo the mould is cut; the cutting of w'liich, and the setting of the skimming-core, are all the moulder is required to do to give himself a good skimming- gate. "When pouring the mould, the flow of the metal is illustrated by the arrows shown. The metal going in at B causes a whirl which prevents, in a great measure^ at the start, any dirt from passing under the skimming-core, and thence up into the mould. This is something our ordinarily used skimming-gates accom- R. CROSS. PATENT GATE Fig. 53. iJti IIIIAVY AM) T.KIHT WOKK SKIMMINT.-fJATKS. plisli lull fffldy. In tliciii ;ill. tin- liisl IIkw of iron rrpiicnilly • ■.•lilies more or less dirt witli it into llii- mould. 'I'lit* idea wliicli -Mr. Cro.is lias cniltodii-tl in iiis .skiiniiiiiiLr-t^ate is indeed worth noticing;. At V\- are set forth soiiii' more ideas in iiit-rmi' F S. The lower jtavt of this long runner IVoiii which the inlet gates 7, -S, '.), and 10 arc eut, is made in the nowel, and is made the deepest at the skimming-gate end, so as to insure its being kept full at the end which admits the metal into the moulil. The gates 7, 8, 9, and 10 are supposed to lie cut thill, and of an area sufficiently small to insure their t;ikinu the metal no faster than the long runner, and gates P ami 7>, will admit of, keeping them full while pouring. This long runner might often have the blind risers, 1, 2, 3. 4, o, and G. omitti'd. Of course, by their use (if the gates P R are kept full) there is very little chance for any dirt that might escape from D (tr R finding its wa}' into the mould. In cases wher.- there arc many castings to make, did one desire to use such a runner having "blind risers," there often might be a pattern made ami rammed uj) with the mould. Also upon the toj) of these "blind risers" it might, in some cases, be beneficial to occasionally place risers which would extend up through the cope, as seen at 4, though as a general thing such would be of little practical value. In some cases the air passing up through risers (were it safe to leave them open) may make suHicient air- current to carry or float some dirt to the riser ; but, as a general thing, the dirt is more lir.ble to stay between or alongside of a liser, should it lie caught there through the upward rising of the metal. HEAVY AND LIGHT WORK SKIMMING-GATES. 127 There arc, no doubt, nianv who cannot see the reason why the gates 7, iS, 9, and lU could not have been cut nearer to the skimnihig-oate DllP, thereb}' saving the necessit}- of cut- ting such a long runner as shown. Tlie reason for cutting such a long runner is simply founded upon the fact, tliat^ the longer the distance throvgh lohich iron is made to travel before it can enter the moxdd^ the better the chances for catching and prevent- ing the dirt from getting into the mould. This long-gate or runner principle applies towards cleanliness, the same as gating a casting, as far as practical, from the parts required to be finished; which is set forth in the previous chapter, "Casting Finished Work Horizontally." Often, in small work, when a number of small pieces are made in the same flask, should some of them require to be fin- ished they could have no better skimming-gates than to let the metal run through the other pieces into them, thus gating from one piece to another ; the piece which receives the first iron from the pouring-gate will naturally contain the most dirt. In pouring any casting requiring to be finishecl, the hotter the metal can practicall3- be poured, the cleaner should be the casting. Pieces gated or run from others especially require to be poured with very fluid iron, not only for procuring cleanli- ness but to insure a good full-run casting. In the first volume, reference is made in several places to the dirt accumulated in ladles, and pouring-basins or runners, and commonly called "impurities." Treating this subject scientifically, the impurities so rapidh' gathered upon the sur- face of skimmed ladles are chiej'i/ due to the affinity iron has for the oxygen in the air. When a ladle is skimmed clean, it is not long before a scum is seen to gather upon the surface of the metal. This scum which occurs from the oxidation of the surface of the metal will, as long as the metal's surface is exposed to the atmosphere, whether in the ladle or on its passage to the inlet-gates, be created. This impurity, coupling lliS lll.WV AM) I.ICllI WORK SKI.M.MlNf;-(;.\'l KS. with the (lust aud tva.shed sand, of pouriiiff-hnsifis or runners, is the reason why we are often Kni|iii.se«l at the aiiiotiiit of dirt enatnl in pouring iiioiiMs with IVesh. ( Ir.in. hkimnu-il ladU'S. The i'liiiction of the skiiiiininL'-gate is U) ealeh and prevent tlii.s dirt I'loin passing into thi- nionld. Of course, good skiinniing- gates will not eonnteraet the evils of nionld-scabhing, ete. ; Imt with intetligence used in gating, in eoneert with a well-uuide niuiild, surprisingly clean castings can be made. II rung i tivnal I'Utc nfT'ucc l^lute. I'igurcs^l'iti ish Size. tinpi \nl(l lie ; l»iit it is astoiiisiiinii to note liuw small :i iiiatlcT will lU-stmy the jMtsitivc action of tlii' i^ati-. 1 will first try to show soino of the errors made in this respeet. The first one is in the bottom of a lonij Imsin, which, instead of having an incline from the pouring-end down to the gate, as seen from P to T", is made to incline exactly the reverse, as fiom R to Y in basin marked ^^ Wrong." 'I'his causes the iron tartls that will save hand jonil-making. Tlu'rc are used as common property four kinds of 1)()anls ; the first l)eing tiie wooden, the second the sand, the third the i)laster-Paris, and the fourth the match lioard or plate. Making these is with some shops a common affair. whiK' with others it is the reverse. There are many mouldeis, who. were they told to make a match plate, or plaster-of-Paiis Ijoard, could nt)t do so without instruction. At the left, in cut. Fig. 55, is illustrated the making of plaster board. At the right is a section of the ])oard as completed. In making this board, the pattern is rammed up, and the joint made the same as if a cope were to be rammed upon it. Instead of the cope, a sectional view of a wooden frame is seen, the inside of which is even with the inside of the nowel. The joint should be made tight, so as to prevent leakage. The plaster is poured in through holes, K, K ; and w'hen set, or hard, the board \a lifted off, and the sand washed off the face of the plaster with watei' and a brush. After the face is dry, it is given a coat of laiup-bhick shellac varnish ; and, when it is dry, the board is ready for use. In making plaster boards, there are a few details which it may be well to notice. Plaster-of- Paris is made by I toiling or burning gypsum, a mineral consisting essentially of sul- l)liate of lime and water, the proportions being : lime, ."52. 5(; ; sulphuric acid, 4G.51 ; water, "20.1)3. (Jypsuin (b'ltrived of its water by burning leaves a powder, tliat, when mixed with its Cope Small Flash Hitigt Constructing I'laatcr Hoard jflustvr Hoard Cotmtlvtvd Fig. 55. THE MOULD-BOARD AND FLASK-IIINGE. 135 own bulk of water, formes a ereani}' paste which ahnost im- mediately becomes solid. lu using plaster-of- Paris, the li(iuidity of the mixture should be regulated by the thickness of body required. For thin bodies, two parts of water to one of plaster may be satisfactory ; but for general work one of plaster to one of water will be nearly right. lu preparing to pour a plaster mould, the outside of joints should be either carefully stopped up with clay, or firmly banked up with sand, to prevent leakage. If nothing but the water comes out, it is, of course, all right ; for much of that is disposed of, and if it does not leak through the joints it is ab- sorbed by the sand in the flask. The holes for pouring in the plaster should be as large as practicable ; for, the quicker a mould is filled, the better for filling thin places or corners. If a mould has any body at all, it will shrink so as to require being filled up after it is poured. Before starting to pour a mould, one should have plenty of water and plaster ; for it does not work very well to have to run away from the job to pro- cure either after a mould has been poured. With practice one can guess very nearly the amount of mixture required to fill a mould ; and It should, especiall}' for light-body moulds, be all mixed before starting to pour. For thick bodies we may par- tially fill a mould, and then complete the job by a second pouring ; but for general work plaster-of-Paris requires prompt and active work. The patterns used should be oiled, in order to prevent the plaster from sticking to them. In forming the joints, special care should be taken to insure that the mould-board will form a joint that will not only lift clean, but one that will leave a finless and true jointed casting. At //, H, H^ are seen nails driven in for the purpose of assisting in holding the plaster in place. In some cases, nails are driven all over the bottom boards, as well as at the sides of the frame. Again, some will, where there are heavy bodies of lol) rilK MOULD IJOAUD AND rLASK-lIlNGK. jdastcr to liolfK put in bars nailod to the frame, or sornrc to it .strips or Mocks drivi-ii liill of nails. I'lastcr lioanls aic onlinanly used only wlioro, from the crooki-dnoss of llie pattern, other l)ourds cannot lie as cheaply made, as perfectly fitted, or kept as trne when being nsed. ^\'()oden boards, when for irregular joints and linely fittcil, are preferred l>y moulders ; as they arc generally light, will retain good edges, and can be moved with little risk of being broken. For irregular shaped patterns, there is prol)ably at tlie pn-seut time none more popular than what is called the '' sand board." The common way of making sand boards is simply to ram up the uowel hard and solid, and then, after making a good firm joint, ram up a false cope or frame. The kind of sand used for the boards has much to do with their life. Some take all new moulding-sand, mixed with aliout one to ten of flour; others will use no flour, but will wet their sand with thick clay wash. Samuel L. Robertson, a man of much experience as manager and journeyman upon light work, informed me of a receipt for the mixture of sand for mould-boards which he had nsed for making irregularly shaped patterns for Taylor & Boggis, Cleveland, O. The mixture is composed of fine sand, boiled linseed oil, and litharge. The sand should be very dry. To al)out twent}' parts sand add one of litharge, mix them thoroughly, and then sift the whole through a fine sieve. Wet Avith the oil to a temper of moulding-sand, such as would be nsed for moulding. This mixture is rammed the same as one would ram all moulding-sand. The board is left to dr}- for about twelve hours, and is then ready for use. The oil gives the sand firmness. The litharge is used as a dryer for the oil. It is not essential that all moulding-sand should be used : almost any sand of fine grain will do as well. Parting-sand, for in- stance, may sometimes be mixed with one-half moulding-sand to good advantage. Should there at any time be corners or edges broken, they can be mended by patching ou beeswax. THE MOULD-BOARD AND FLASK-HINGE. 137 In light work, the keeping of the joint edges of sand-mould boards sharp and unbroken, is of the utmost importance. A great many, to lielp preserve them, will nail all the joint edges : even then they will become ragged, and cause bad joint-work. The objection to plaster-board for fine work is about the same ; much working in and out of the pattern soon breaks the edges. The boards made with the oil and litharge keep their edges good and true surprisingly long, and it is on account of this that they are thought so well of ; and any who will give them a trial will, no doubt, be greatly pleased with the results. Alex. L. Faulkner, one of our Cleveland moulders, holds letters-patent upon an elastic follow-board composition, wliich I lately understand is being ranch used, and spoken very highly of. To some extent the above composition is like his ; but, from what I can learn, his manner of mixing and manipulating his composition makes a much superior " foUow-board" to that which the above will give. Any one doing a large business in light work will no doubt find it will pay them to investigate this matter. As an auxiliary to the fast production of small work, the match-plate is often used to good advantage ; the making of which, although a simple affair, is in the minds of some thought to be work requiring fine manipulations and measure- ments, the same as is required in the making of wooden match- boards. In Nos. 2, 3, and 4 (Fig. 55), is illustrated the manner of constructing match-plates ; two patterns being selected, in one of which the indentation comes below the joint line, and in the other above it. At No. 2 the nowel is rammed up and joint made, F and E being the patterns. The cope, having been rammed up, looks as seen at top cut shown. The process so far is simply what one would do, were he makuig a casting from each of the respective patterns. As, instead of doing this, we intend to 1:18 TiiK Mori,n-iu)Ain) and flask-iiinge, cfiiistitict a iii:il(li-|ilati', cxti'iKlcd iii:iiii|nil:it iods nrv n'(|uiiv(l. A^ llic iciUnii iiorliuii i.s iikhiIiU'iI, wlial is ii<»\v waiiti'il is U) mould tlic! plate pcirtioii. This is done l»y ltuilrinciples below THE MOULD-BOARD AND FLASK-HINGE. 139 set forth, I siinply give thinking the ideas maj^ prove of vaUie in some chisses of work. When tlie centre of the hinge is on a line with the centre or joint of flask, the lift, at the moment of starting, tends towards the hinge side, thereby clearing any indentations the soonest upon side opposite hinge. To more clearly illustrate this, the cuts "inward" and "outward" are given. At inward, the centre of hinge B is considerably below the joint. The moment this cope is started, the lift will be inward, as shown by the arcs SS. In the upper cut, on account of the centre of hinge being above the joint, the reverse would be true, as shown bj^ arcs BR. The distance of the hinges being so far below and above the joint, the arcs drawn from the centre of hinges show a true inward or out- ward movement, as the cope is raised or lowered. It is evi- dent from this illustration, that the matter of having a cope go from or towards the hinge side can be controlled, thereby assisting in getting good lifts when a movement in either direc- tion is desirable. Of course, the farther from the joint-centre the hinge is, the more rapid the outward or inward movement. The intersection of the line MN., with arcs cutting same, shows in w'hat ratio the given radius or outside of flask rises com- pared with the inside. This ratio increases proportionately as the radius, or width of flask, increases. The cut of flask hinges shows two styles that are handy for light work. The upper style is to be secured to the jpints of flask ; the lower one, to the sides. Either could be constructed so as to bring the centre of hinge below or above the joint, to cause inward or outward motion when first starting the cope, should it be desired. lll'KS, COKKS, AM) HOJ.LOW ril'K TAl'lKUNS. PIPES, GREEN-SAND CORES, AND HOLLOW PIPE PATTERNS. There nre few foundries that do not, in somo ff>mi. make more or less i)ii)es ; and it is astonisiiin^ to note how much faster the same class of pipe-wi^k will he made in some shops than in others. This is mainly due to the dififerenee in the facilities and rigging. In some shops, a man may have to work liardcr to make one i)ipe than he would in others to make four; and, as a general thing, the shop that could tin-n out the four would lequire the least skill. Shops that produce such castings the slowest are, as a general thing, the ones tiiat liave the fewest to make, and therefore cannot afford the expense of getting up labor-saving rigging. There are times wlien a little outlay in some shops would be the cause of procuring much work, that, in the end, might result in the manufacture of a good paying specialty. The general jobbing-shop way is to make solid, dried-sand pipe-cores. The extra expense made thereby is the requiring of flour, and sometimes beer or molasses, to mix with the sand. It also requires much labor to make them, fuel to dry them, and tlie loss of sand ; and after all the time, labor, and expense, we can seldom produce a perfect, round, even core. A plan practised in some shops that make a specialty of green-sand pii)e-castings is as illustrated in cut. Fig. 56, showing the sweeping of a green-sand core. This style of core is surd to i)roduce a round hole ; and. with rigging properly gotten up, one man can make a large number t)f cores in a day. The sizes of l)ipe generally made by this plan range from o' up to 12". Fig. 56. PIPES, CORES, AND HOLLOW PIPE PATTERNS. 141 In making the core-arbors, there are two plans usnally adopted. One is, to east arbors having prickers, and the other, ribs, npon their snrface, to assist in holding on the sand. To show what is meant by ribs, the sections F and S are given. At A", and in longitndinal section of core, prickers are illnstrated. As a general thing, the ribs are used for the smaller sizes of arbors, on account of their making the arbors stiff, thereby preventing the core from springing up and shutting off the thickness of metal when the mould is poured. The larger arbors are in diameter, the more resistance to ■ springing they generally have when moulds are being poured ; so that arbors over five inches in diameter can generally be made stout enough without the ribs. For holding the sand, prickers are to be preferred. The ribs separate, as it were, the sand into sections ; whereas the prickers keep it together more in one bod}'. The larger in diameter, the longer can pipes be made. A foot pipe could be some nine feet long ; a '6" pipe, four feet long ; and sizes between, in proportion. Of course, the stift'er arbors are made, the longer can the pipes be made. AVere chaplets used with this class of cores, as with dry-sand cores, they could be made much longer. There is a way whereby chaplets can be used with some green-sand cores : that is, to have a knob about one inch in area cast or riveted to the arlior, as above the chaplet X. This spot, being even witli the surface of core, rests upon the chaplet, thereby caus- ing iron and iron to come together. For the cope, a small body of sand is taken out of the core, and some small plates or washers inserted, the top surfaces of which had better be kept ^" or so below the core surface. The space around these wash- ers or nuts is tlien filled ni, and the core then made as suKJoth as the rest of its surface. Upon tlie top of the inserted pieces the cope chaplet rests. In so chapleting cores, care is requu-ed ; for, should the chaplets come otherwhere than intended, the core would be burst, and the casting, as well as the arbor, 1 \-2 riri'.s, (•(jKHs, and hcjllow rii-i-: rATiKUXS. most likely lost. "Witli such work, exact nic.isnrcincnt and (it- tiiiixs nn- ivciiiiriMl, With hiriic-diiuiictcr pipes, there iiiij^ht l>c (l;iii;^^r, l>y thus chapleting, of bursting the casting, on ar-coinit of the knob X, and the pieces above it, making a brace that would prevent contraction. .Sometimes there is no danger of the core springing downwards, but tlierc is a tendency to rise. When the lifting-strain of the fluid iron comes upon it in such cases as this, the bottom requiring nc^ chapht. the knob X could be upon the cope side, and the cores thereon be chaplett'd down, casting the pipe by having a chajilet only ou the cope side. The reason for using the washers or loose plates instead of a solid liody being secured to the arbors, as above X, is to alhnv the arbor to free itself. Were the top the same as liotlom, there would be immovable iron to iron. By having loose washers or plates, the jarring of arbor soon causes it to be free, thereby lettuig it come out. Core arbors should l)e well perforated with small holes, to allow the gases to escape. The thickness of sand upon arbors ranges from f" to 1". The more dry the sand can be practi- cally used, the better. In sweeping up a core, the process generally is to wet the arbor with clay wash or water, and after being set upon the horses the sweep board is set, sand is packed l^y hand upon the arbor, after which, with a man turning slowly, the sweep l)oard is lightly pressed forward until it strikes the gauge guide which gives the diameter wanted. The arbor ends, A II, can be used to give the diameter ; but having the core gauged independently of the arbor is to be preferred, as the friction of the turning will wear away the guide, and also more or less vibrate the arbor, thereby often causing the sand to droj). A i)oint that here might be mentioned is, that the less sleek- ing done to pipe cores, the better. In fact, it is l)est not to sleek them at all. leaving the suifaee as it is swept, as thereby the metal lies mure kindly to the core. PIPES, CORES, AND HOLLOW PIPE PATTERNS. 143 In casting the larger-sized pipes, it is essential that the arbors should have reliable bearings. 3" up to 4" pipes could be cast by having sand print bearings ; but above this last size the arbor ends, A II, would be better if turned up true, so as to exactly fit the flask iron ends, as shown at M 31. The cut T T shows the end without the core in. The arbors being true, the flask ends would of course requu-e to be the same. By haviug arbors set in such bearings, it is evident that the core will be kept central, and that its weight cannot sink it down, or the liquid ii'on raise it up ; that is, as far as the prints are concerned. It might be well to mention that the pattern prints must fit into the flask ends when moulding the pipes, in order to have the mould central with the flask ends. In making arbors having such iron-end bearings, one, if not both, should be made smaller than the niside of intended pipe, so they may be readily got out of the castings. The longitudinal section of mould shows a flange on one end and a socket on the other. This is only to illustrate the idea that either kind can be made. In the smaller sizes of pipe, it is not necessary that the arbors should be larger at socket end, as shown at //. If the arbors are straight their entire length, and the sand reasonably tough, the little extra thickness re- quired to form the socket will hang. The plug seen at R is inserted for the purpose of lifting the core. AVhere arbors are large enough to admit a trunnion l)eing riveted on, as seen opposite 72, it is advisable to do so, as they can be revolved easier. Revolving arbors, b}^ having their whole diameter turn in a bearing, as seen at A^ cause much friction. The cuts of elbow and branch pipes illustrate the making of pipes with hollow patterns, they being the same as the castings wanted. At jB ^ is shown a sectional view of the pattern. The nowel having been rammed up, the core arbor P is then set in and rammed up. The cope part of pattern is then set 144 rirF.s, corks, and hoi. low riri-; I'aii i:kns. oil, mill smikI tiii'kc(l ill. Tin- jninl Imviii;^ Iici'ii iiijuli-. tlir copo is rainiiifd up. and, after hciiiii lifted off, tiii' top pallfiii is drawn. iJy taking hold of tlii- ailior handles. Nos. 1, 2, and 3, the eore is lifti'd out ; the bottom pattern is then drawn, and the mould finished. The core is then set haek. and cope closed. The end of arbor at No. 3 is of a style different from Nos. 1 and 2. Arbor ends as at Nos. 1 and 2 are handy for small pi{)e. The arbor is set on the mould board, and the iKJwel half of the pattern over it ; then the nowel is rammed n]t and turned over, the arbor forming its own print. This style is not recom- meiidid for heavy cores, as it does not give print enough to hold uj) very much weight. The arbor as at No. 3 is of the same form outside the i)attern as it is iuside. To form prints for such arbors, with hollow patterns, there could be half-round blocks, as sliown in plan at B, rammed up with the nowel half of pattern, and tiu-n. wiu-n the nowel is rolled over, draw out the blocks. This would leave prints formed ready to set in tiie arl)ors. The quarter-turn pipe shows the plan of an ai'bor made so as to balance the core; the balancing wing i)rojecting beyond tlie mould prevents the back ll' from sinking down as it would were !)()th ends of arbor the same as at J\r, and the back not chapleted. This style of an arl)or can, of ct)urse, be operated as regards rolling-over and print-making, the same as the T arbor described. At yis shown a core rod, and core made upon it. The head D admits of the core being lifted vertically, and also is a sup- port to the core if rested upon its end. This class of green- sand cores can be used vertically or horizontally, and for pipes a1)out one foot long, 2" or '•>" in diameter, where their manufac- ture IS to be made a sjiecialty, they are worthy of consideration. The cores are rammed in a box I'udwise. and reciuire to be vi'iiti'd, fur wliicli, in some cases, it might l)e well to have two or tiiree vent holes drilled thiou'ih the head D. PIPES, CORES, AND HOLLOW PIPE PATTERNS. 145 Green-sand cores, as a general thing, require more or less rigging, which is one reason why more shops do not use them. The holes formed by green-sand cores, as a general thing, for smootJiness and being true, surpass those made by dry-sand cores ; and generally thinner castings can be made with green- sand than with dry-sand cores. The making of green-sand cores often requires much skill. There are many cores being made of dry sand that could be made of green sand ; but, like many other things in moulding, it often requires practical ex- perience and good judgment to decide the feasibility of making them. IK) lti:i)l)IN(MN AM) liOLLlNG-OVEU. BEDDIXO-TN AND ROLLTXG-OVER. BF.nmNc-iN and rolling-over ptitte'riis in monUling have eaoh tlR'ir special advantage. As a general thing, rolled-over nionlds are the simplest to constrnet ; the reverse being the case with bcdded-iu castings, A moulder that cannot successfaUi/ turn out a good general run of casti7igs by roHhig-over need never attempt it by bedding-in. The writer is well aware that there are castings that cannot be as reliably made by rolling-over as hy bcdding-in ; but this fact does not change the sense of the statement made. It will be acknowledged by all practical moulders who have had experience in both rolling-over and bedding-in, that to do general bedding-in requires higher skill thau rolling-over. Any shop that does most of its moulding by rolling-over can often get along with less-skilled mechanics than where the patterns, as a general thing, are bedded-in. "When a moulder is furnished with uice patterns and flasks, the requirements are often like those of machine labor : the physical, and not the mental powers, arc the ones most required. Were there more bedding-in practised, ice should have more and better-six illed tradesmen. A novice, in travelling through the foundries of the country, would l»e at a k)ss to reason why sliops, in making similar castings, do not adopt similar methods. He sees one bedding-in almost ever^' thing : another he finds rolling-over every thing. In many cases, this puzzles even practical men to reasonal)ly explain. One can go into many shops, and there see i)atterns being bedded-in, that, all points considered, could be better rolled-over: then, again, he will find the reverse, there being large, expensive flasks BEDDING-IN AND ROLLING-OVER* 147 used for moulds that could be made in less time and with far less risk by being bedded-iu. There is no doubt that upon this point there are sliops that are working in error. Almost every machinerj' foundry has some jobs, that, in point of economy and safety, would be better were they bedded-iu, and some that would be better rolled-over. Sometimes circumstances may be such as to call for a pattern being bedded-in when, properly, it should be rolled-over. This, however, is no excuse for the wide difference in shop practice. I have seen practical men, who, when questioned why they did not have certain jobs bedded-in, would say they knew it was the proper way to mould them ; but, having so little of that class of work to do, they did not like to have their shop floors all dug up. This is, no doubt, in many cases, a good reason for not bedding-in work. Shops in which, most of the work is bedded-in are, as a class, the dirtiest and ugliest to be found. It is practically impossible to keep them as clean and orderly as a shop that does all rolling-over. A foreman that loves order hates to sec his shop a jumble of holes, sand-heaps, and foundry- tools. He may, to some extent, control and keep order ; but to this there is a limit. It can be carried so far as to be a source of expense rather than of profit. My lot has been chiefly to be employed with the dirty class of shops. It has often made me feel envious of my brother tradesmen who work in the clean shops, to think witli what comfort they can work ; and I would long ago have been one of their number, were it not for the charm that bedded-iu and heavy work has for me. There is a fascination about beddiug-in, that many moulders enjoy. The advantage that bedding-in has over rolling-over is, in the first place, the saving of flask-making ; second, the rigiduess with which sides and bottoms of moulds can be supported against the strains of high and heavy heads of metal ; tliird, the assurance it often presents of making a casting the dupli- 148 •uKDDINf; IN AM) KOMJNO-OVEU. Cite of the pattern in shape. The twistinir iiiid wnnehinf; that are ein;>; turned over, often nmUcs it inipos- sihle to make a eastin<^ as true as its pattern. This point was ahl}' broucM.sive jiasks, that conld he saved liy l»e(Ulin<:f-in ; especially so where there was only one or two of a piece to make. He started at the jol» ; and at tlio end of abont two li(>urs he pnt on his coat, remarkin, ramming up the cut-out place, as at *S', with facing-sand, until the whole side is rammed up. This plan for heavy work, where sides or flanges cannot be gotten at to ram them solidly up with facing while the pattern is in place, is a good one to adopt, as it gives every chance to make a firnj surface when the pattern is withdrawn. There are many pat- terns where portions of level beds can be used to assist in bedding-in, the plain surfaces of the patterns resting upon the beds, and the irregular parts being tucked up. Wherever a levelled bed can be used, it should be, as there is no way that a mould's bottom can be controlled and made so reliable. Although levelling a bed is a simple affair, it is astonishing what a small per cent of our moulders know how to go about it ; yet to accomplish it requires no great skill, as will be seen by the following. In levelling a bed, one side, as F, should be first levelled up, after which set the opposite one, P. Then upon the top of each and at one end, as seen, set a parallel straight-edge (by parallel, I mean that it must be exactly the same width at each end, not 6" at one end and 5^" at the other) . The straight-edges F and P do not require to be par- allel, but N must be if a level bed is wanted. With the parallel straight-edge, level across from F to P, then try the level on P ; and if it should not be level, make it so by raising or lowering the end at P. Then test the straight-edges by going over them all two or three times if necessary. Another point to be watched is the level, which in a foundry soon gets IFA 1JK1)I)1.N(J IN AND lUJLLlNC; OVKU. out of triitli. The way to test a U'vcl is to turn it on .ioiniim; r;ui:i;N sank .Mori.Ds. sfiils a licltcr lioltiiMi ii|i<)ii wliicli lo set j^ajif^tTH than tlic part iiiaikiMl *• w loiiLr." To lift a \nn\\ ol'sauil, nr a joint, tlie less sainl tluie is uikUt (li(> piLTLjcrs, the Itetter. Hoiiu'times j(»iiit-<^aii:;4ers are set with IK) sand under tliein ; hut this is not ^i-nerally to lie approved of, as it does not make a neat joint, and niii:;ht, in case of straining at the joint, cause the mould to " kick." AVhen it heconu's necessary to patch a joint, from not get- liiiti a good lift, it is usually very tlillicult U) get it as perfect as it would otherwise have l>een. Sonu'tinies the pattern can he set on the cope to assist in getting the n-ijuired shape ; but even tlu'U it can seldom he accurately done. The best-jointed castings are those where no joint-patching was required. The word patched should generally be connected with hntched; although it is easier to botch a job than to patch it, and some moulders who will do a good job of patching are far from being botchers. At X, F, and N xa shown a plan of setting lifting-bars, that I, for two reasons, seldom approve of. The first reason is, that it compels the placing of the flat side of a bar parallel with the surface of the iiattern, thereby often necessitating ramming and holding a thin, flat body of sand in its place. In ramming sand in such narrow pockets, the best judgment must be used. Tf the sand is rammed too hard, the gases will not escape freel}', and scabbing or blowing will be likely to result. An- other objection is, that when it is necessary to roll the cope over, the thin, flat cake of sand is likely to drop off, unless securely " rodded." I always try to have bars for lifting out pockets, or carrying hubs or other projections, arranged so that there will be a con- siderable body of sand around them. This not only lessens the danger of bad results, but gives more room for ramming up and for seeing what is being done. The second objection to using bars in pockets as above shown COPING, VENTING, AND JOINTING GREEN-SAND MOULDS. 157 is, that setting the gaggers is inconveniently clone, and the danger of a "■drop-out " is increased. In this cut, three ph\ns for making deep-pocket joints are shown. At A is represented a plan that will give a free lift, but involves setting many gaggers, and unhandy ramming. At 6 and 9 is shown how this plan of barring causes gaggers to be set, which makes the ramming awkward, marks the joints, and does not securely hold the sand. At B the joint is made more nearly vertical, by which the above objections are to a great extent removed. In making a joint for such partings, the more nearly vertical it can be made, the better : 4" slant to a foot in height will generally work satisfactorily. At K. the irregular line, is represented a plan sometimes resorted to on the plea of lack of room, poor tools, etc. The plan shown is a very poor one. Numerals 1,6, and 9 upon the left represent lack of jndg- ment in trying to lift a body of sand. The gaggers seem to be set on the theory, that, if they are only gaggers, that is all that is required. The sand at 9 would be more likel}^ to be lifted if the gagger were not used, as its length is only about that of the body of sand to be lifted, and iron is heavier than sand. Nos. 1 and 6 represent conditions not much better. No. 1 shows how eas3' it is to put one clumsy gagger where it will do the least good, or where there should not be any. K I could not have bars as at JE" , and it were necessary to set a gagger at 1, I would keep it up about 3" higher, and turn the toe of 8 the reverse of what it now is, so as to bring the point of the gagger under the bar towards the hub. The hook shown on 8 is generally made only on wrought-iron gaggers. It is often serviceable for carrying heavy liodies of hanging sand. In some shops, wrought-iron gaggers are used almost exclusively, while in others cast-iron ones have the pref- loS roi'iNf;, vKN'riNO, and .loiNrixf; rjUKKX-SANi) mot'lhs. croiicc. AVliilo I prefer those of cast-iron, for {General use, as they will not spring, are eheaper to make, and can he readily broken olF to any desired iengtii, I also like to have some wronii;ht ga.if'];ers, as they can l»e heiit to set npon slanting snrfaces, etc. I am aware that some will object to breaking gaggers, and that in some shops the rule is that they shall not be broken ; but l)efore 1 would allow them to l)e left sticking out of a cope, as at () (wlu-re tliere are none short enough to be found), I would have them broken so as to come no higher than 5. Gaggers sticking up, as at C, are lialjlc to be hit, resulting probabh- in losing the casting. I never allow gaggers to be left standing above the cope, if it can l)c possibly avoided. Gaggers 4 and 5, in connection with the bars as at E ^ repre- sent good practice. Gagger 2 shows how gaggers are some- times badly set by the side of deep hubs and flanges. No. 3 represents a better plan ; and if the cope is to be rolled over, use more gaggers as the height of ramming increases. The points of gaggers against the flat surfaces of hubs, flanges, etc., cannot do the harm flat surfaces can when set as at 2 ; that is, by producing hard and soft spots in the mould. The ramming is also an important factor in getting good lifts. The ramming of a body of sand to be lifted should be firmly and evenly done. In the cut, at the point marked " Copes staked," may be seen the marks of the rammer impressed in what should l)e a level joint. In some cases this would pre- vent the sand from being lifted, even though well barred and gaggered. In making irregularly jointed snap-flask moulds, the joint is generally the point of particular importance. Fins on such castings often condemn them. With this class of work, a per- fect joint will, in most cases, provide for a perfect casting. A good bench-moulder pays especial attention to his flask-pins : he sees that they are uot loose or shaky, and that they fit true. COriNG, VENTING, AND JOINTING GREEN-SAND MOULDS. 159 Floor-moulders have so many other things that claim their at- tention and time, that the joint seldom gets the attention it deserves. It is apt to be thought, if the casting is all right with the exception of the joint, that a chisel and file will soon fix that. The quicker such ideas are got rid of, the better. A floor-moulder should take the same pride in the joints of his castings that the bench-moulder does. In small work, there are two objectionable joint features. One is the fin, and the other " overshotness." In heavy work, the fin can seldom be avoided, but overshotness should always be. The stake marked "Ring" shows how stakes are often driven, thereby providing for bad lifts and overshot castings. The stake on the opposite side is driven correctly. The ring on the stake is made by cutting off pieces of wrought-iron pipe of the proper diameter. They are good for protecting the stakes from the blows of the sledge-hammer. In staking flasks for ordinary work, at least two-thirds the length of the stake should be driven in the ground. Some- times, for greater surety, it is advisable to drive two stakes, one behind the other. • With good sand or plaster-of-Paris mould-boards, the skill and labor of making partings or joints are saved. It is where joints must be made by hand, that the skill of the moulder is tested. With some irregular light mould joints, it is often advisable to start up the pattern when the joint is nearly completed. This will show if all parts have been made so that the pattern will draw freely. The pattern is then to be lightl}' rapped into its bed, and the joint completed. Then the cope is rammed and lifted, and the pattern withdrawn. To still further insure getting a "good lift," it is often a good plan to airange for rapping the pattern before the cope is lifted off. This is done by having rapping-plates on the IGO COI'INO. VKNTINC, AM) JOINTINC fMtKEN-RANn MOULDS. pattoni, if of wood ; f»r luiviii'^ liolcs in the ii:ittcin, if of iron. Then, wln'ii r:iiiiiiiiiiij; np tlic cope, rain up i^ati-.s in IIil' holes; and then, with a pointed l»ar set in the; pattern holes, it can he lightly rai)i>ed in all directions. This is a i»lan adopted l)y most bench-nionlders, the onh' dilTerence being that their rap- ping is generally done through the same gate-hole as that by which the mould is poured. "With copes where two or more men are required to lift them, it is often a good plan to raise the cope an inch or two by slightly raising a corner at a time, inserting a wedge to hold it up. Again, it may be advisal)le to raise one end or side at a time ; but in either case the corner, end, or side should be raised only a small distance, — sometimes not more than ^^" at a time at first, — which distance can usually be increased at each succes- sive lifting. In order to assist in getting good lifts with a crane, iron starting-bars arc sometimes placed as shown. Usually the first starting of the cope is the most important. If it is started so as to jerk one side up before the other, the most careful gaggei'ing, ramming, etc., will have been of but little avail in giving a first-class lift. There are two more points upon which I will express an opinion, and which may be of interest to those moulding heavy work. At F, in the lower cut, is represented a plan of cutting fins, which may be new to many. Of course, fins are objec- tionable, and should be avoided upon light eastings, and upon heavy ones where the joint is on the casting-surface, as in col- umns and similar castings. But for heavy castings, where the cope surface ends at the joint, or the mould does not project up into the cope, as shown at P F, and also for bad or heav}' drawing-})atterns. cultiug fins is often advisable for two reasons. The first is, tliat in diawing heavy patterns the joint of the mould is to a greater or less degree started. This may 1)0 sleeked down, but to get the benefit of any doubt it is often COriNG, VENTING, AND JOINTING GREEN SAND MOULDS. 161 wise to cut for a fin. Of course the idea is, to be sure the cope does not touch the mould at tlie joint's edge. The fiu should ruu from the surface of the mould back from 2" to 4", ■wedge-shaped, as shown. The thickness of the fin at the mould should be determined by a consideration of the degree to which the mould is started in drawing the pattern, and to some extent by the temperature of the iron to be poured. For dull iron, the fin should be thicker than for hot iron. For safety, and to assist in getting good heavy castings, they are usually poured with dullish iron. In pouring dull iron, the upper edge or surface of the casting is likely to be wavy, presenting the appearance of cold shut. Cutting a fiu at the edge of the cast- ing is, to some extent, a remedy' for this ; as it assists in the escai>e of confined gases and dust, or permits them to be held in a space, which if the metal does not fill no harm will be done. Observmg moulders know that an open sand casting can, by pouring with dull iron, be made from ^" to \" thicker than the mould, for the simple reason that the top edge runs rounding, allowing the surface to run higher than the edge. Coped castings would run rounding in the same way, were it not for the fact that the head pressure forces the metal into filling the top edges ; but this head pressure is sometimes insufficient to fill the edges. In the case of a mould in which a fin is cut, the chance of the above occurrence is measurably lessened ; and, the thicker the fin, the greater the extent to which it is lessened. The sides and under portions of moulds are often vented direct from the sui'face of the joint, as on the side P. Vent- ing as at S is not always reliable for heavy castings, because there is a chance that the metal will get into the joint ; it also makes the management of the joint laborious. On the side F is represented a far more reliable way to vent such moulds. AVhen the pattern is rammed up to within from 4" to 6" of the joint, the side is then vented, and fine cinders placed as shown. 1(')2 roi'IXC. VI'.NIINf;, and .lOINIINf; (lUI'.KN SAXT) MOT'I.DS. 'I'lu' rciiiaiiHlrr tiC the (Ii'|illi is tlicii r:imiiicat(('in-maki'is work as if tlu'y wiMf liou.st'-joincrs, or were makini; tool-chests or children's toys, onl}' occasioniUly getting an idea that they are working for the foundry by sechig a dirty moulder i)ass them. AVhy pattern-makers will not give suf- ficient tai)er to patterns, when there is nothing to [)reveut it, is a question that has often puzzled many a moulder. The attain- ments of the pattern-maker in the way of draughting, and in working wood into various forms, count as nothing with the moulder if he constructs patterns that will not draic icell. The moulder's skill is proved by having a '■'■ gjod cast ;" the pat- tern-maker's (if he only knew it), by having a '■'■ good draiv." To have corners, edges, or portions of moulds started or broken through ill-drawing patterns, is not only very aggravating, but is often the cause of defective castings. Another point is the hammer al)use that patterns receive. ]\Ioulders are called destructive because patterns are pounded. If we are destructive, the pattern-makers are greatly to l)lame for it. Give us patterns properly provided with draw screics or irons and rapping-holcs, and of a good taper, and our acquired practice of unmercifully hammering every thing that comes along will very soon be lost. Before patterns can be drawn, they generally recjuire to be loosened. To accomplish this, the moulder must do some ham- mering. Some one may suggest the use of a pounding-block, to preserve the pattern. As a general thing, this is used when DRAWING AND MAKING PATTERNS. 165 practicable. The pounding-block cannot always be used to loosen a pattern, because it frequently only causes vibration. To loosen and to vibrate are different things. The loosening is required before starting to draw. The vibration is the second requirement, or that necessary to lessen surface friction or adhesion when drawing the pattern up. Arrangements for loosening patterns are seldom provided. Let one go through almost any machinery-pattern warehouse in the country, and he will find the patterns scarce having good provisions for preserving them from the effects of the "loosen- iug-bar " and hammer. "What rapping-holes are seen were most likely first made by a moulder with an auger or a pointed fo o °^ o O o f \, o 0/ r"' — , ^ \'i/, ,:\f. \ Joint 1'' Fig. 59. Fig. 60. iron bar. In a short time the holes become so large, you can hardly see any pattern. It is proper for pattern makers and owners to take some of the blame for abuses of patterns, and to provide for increasing their durability. The expense of inserting iron rapping-plates in wooden patterns is but little ; and, were the custom once established, the benefit derived would soon be seen. Rapping-plates should be placed so as to jar the whole pat- tern. This will often necessitate the building-in of more lumber than is necossarj' for making such shells as are often turned out and called "patterns." Rapping-plates can be either cast- or IGO DRAWING AND MAKING PATTFJINS. wrouglit-iron, niul made in wliatcviT fsliape the form of pattern may rcMiuiie. For soiik' iialtcnis, the idea given of a ea.st-irou l)laU', ill FIlj. ;")'.), will work well. The size shown is that which would lie bnitable for hiri^e patterns. For small ones, the size coukl, of course, be decreased. Some pattern-makers go so far as to make a practice of inserting draw irons or jtiates. This is an excellent plan. But, if the pattern is one to ie(piire much rapping, there should also be a raiiping-plate. Jn some cases, the draw and rapping holes may both be in one plate. For such patterns as small gear-wheels, and others where there is no room except for a small plate ou the hub, the draw- Fig. 61. Fig. 62. plate should be the one. Then the moulder can both rap and draw the pattern with the same draw-screw. Patterns arc as likely to be destroyed from the lack of draw-plates, as they are from the lack of rapi)ing-plales. It is sometimes said, if the pattern has a weak spot, the moulder is sure to drive his draw-spike or screw there. lu some cases this may be true. The moulder generall}' tries to insert his spike or screw where it will best balance the pattern : therefore, should the weakest spot be there, in there, of course, goes the spike. Too many patterns are made without provision for drawing them. 1 have used patterns that, before they could DRAWING AND MAKING PATTERNS. 167 be got out of the sand, would be — and the mould as well — literally torn to pieces. I will admit, that sometimes moulders are thoughtless regard- ing where they drive draw-spikes, or place screws ; but what class of tradesmen would not be, under the same circumstances? Our patience is often sorely tried b}^ labor and grief caused by the negligence of pattern-makers. Could we cause them as much extra laboi' and trouble as they often cause us, I think they would try to accommodate, and study more to assist the moulder. In small work, the facilities for expeditiously drawing pat- terns are much better than for large work. This, in a great measure, is due more to the moulder than to the pattern-maker. Small-work patterns are generally under the moulder's super- vision, because of their being chiefly made of iron or brass. All parts are given sufficient taper to insure their drawing well ; and, where assistance can be given in steadying the drawing, it is generally done. In Figs. GO and 61, two ways of doing this are shown. Fig. 60 shows the common way of using steadying- bars ; while Fig. 61 will, to many, present a new idea, which, although only adaptable to a narrow range of work, is, never- theless, worth notice and thought. The principle was first brought to my notice through the kindness of Samuel E. Hilles, of S. C. Tatum & Co., Cincinnati, 0. As seen at K^ the runner part is made round. In the plan-view, at SSS, are shown the brazed gates, which unite the runner and pattern. In drawing the pattern, the end, as seen at H, is simply lifted until tlie whole surface is clear of the mould. The runner /i", being round, acts as a fulcrum for the pattern to roll upon. For such work as sewing-machine legs, etc., this device could often be used to advantage. In heavy or large patterns, the moulder does not have the facilities for conveniently drawing his pattern, the same as in light work, not because it is not possible, but because the foundrymau, in heavy work, does not 1G8 DUAAVIXn AM) MAKINO PATTERNS. have llio iiiniin^fiiiciil of liis )ia11t'ni-in:ikiiig lo ro |Trpat an exti-nt as in Ii;j:lit work. E J'J, Fit;. een pulled apart, moulds Iti-okcii, and moulders enraged, through the lack of a few simple draw-irons. Missionaries desirous of suppressing evil thoughts and swearing could accomplish much by placard- ing pattern-shops with the words "Taper" and "Draw-irons." Now, I do not wish to be understood as thinking moulders arc all perfection. I have seen much " smart Aleck " business among moulders, regarding the df\^ign of 2'>o.tt€rns. There are man}' who can show great moutli-wisdom, and find fault with details when patterns are completed, which, in reality, are so far above their conceptions, that they have not the least idea of the dillicultics attending, or the thought and skill required in making them. One never hears such men approving any thing. They 0|)en their mouths only to find fault. Instead of finding fault with patterns, we often ought to feel thankful they are as handj' as they are. It is no trifling matter for a man to take the general run of drawings, and therefrom conceive, and fully plan in his mind to perfection, all the details of a pattern. The pattern-maker often does well, when we take into consideration what he knows al)out moulding, and the con- ditions under which he does his work. If they would only give us "good drawing," and patterns readily "rapped," I would never say a word. SKIN-DRYING GREEN-SAND MOULDS. 169 SKIN-DRYING GREEN-SAND MOULDS. Skin-drying moulds is a term applied to green-sand work, where the surface of the mould is blackened over similar to the process in dry-sand work, and then surface-dried. Skin-drying is generally done for the purpose of giving stability to the surface of the mould, and for assisting in the peeling of solid castings, as anvil blocks, etc. There are a few shops that practise it with lighter work, solely for the purpose of giving their green-sand castings a " diy-sand skin." Then, again, some shops find it necessary to skin-dry much of their work, because of the nature of their sand, which has but little body to withstand the heat and wash of metal, or contains too much clay. In skin-drying moulds, much judgment is required ; for a plan that will answer for one mould will seldom do for another. Not only have ways and means to be devised for drying, but the nature of the sand has to be considered as well. A sand, to work well, should, when dried, present a firm porous crust. Some sands, on account of their weakness, must be mixed with some substance that will give them a body. For such purposes, flour, beer, molasses- water, or clay-wash may be used. "When flour is used, it is mixed in the proportion of one to twenty up to one to thirty, according to the quality of the sand. The beer, molasses-water, or claj'^-wash may be used in connection with the flour in place of water for wetting the sand ; or the flour may be often omitted, and the sand be sufficiently strength- ened by aid of the above washes. Sometimes sand, because of its closeness, requires some 170 SKIN l>ItV!N(J (iUKKN-SAND MOULDS. sharp sand mixed with it in order to make it work wrll. AVhile some sections ])ossess mouIdin^-sand natnrall}' adapted for skin- drying, others do not; and tlierefore more or less ''doctoring" will he reciuired to make it work properly. In using the above mixtures, it nnist l)e understood, they are used as a facing : all required is to have it face the pattern from 1" to 2" in thickness. For a backing, the " heap-sand " is used. "Where copes are skin-dried, tliey sliould, as a general thing, be well ''gaggered," and sometimes nailed, as the drying of their surface forms a crust tiiat may easily drop off unless held up by the gagger support. Some moulders practise nailing the sides of moulds that are over G" deep ; and this practice is not one to be condemned, as it will often result in obtaining a good casting. The gates and sections of the mould where the metal first enters are generally the points that should at least be sur- face nailed ; for in skin-dried moulds, if the surface once gets broken, the under crust soon washes away, for it offers but little more resistance tlian so much dry dust would. Skin-dried moulds demand that all joints be well finned, for the least touch may readily cause a crush. There is no class of moulds that require more delicacy in handling, for its surface is a crust that has but little union with the body of the mould. Some moulders will not even trust to the nails for holding the portion at the gates : instead, they have cores made the shape of the mould, and ram them up with the jjattern. This is the most relia))le plan to adopt to prevent the moulds from cutting at the gates when there is a quantity of iron to be run through them. The facing for skin-dried moulds is, as a general thing, worked or used a little damper than facing would be for com^ mon green-sand work. After a pattern has been drawn, and the mould finished up by using beer or molasses-water for swabbing purposes, the next process is that of blackening. In blackening a mould, two plans may be adopted. One is to blacken the mould in a way similar to that used in a dry- SKIN-DRYING GREEN-SAND MOULDS. 171 sand mould : the other is to rub the blackening on dry, and then after sleeking to go over the surface with molasses- water or beer, the molasses-water being the better of the two. Rubbing on the blacking is of course only necessary upon the sides, etc., of the moulds, where a sulHeiently thick amount will not adhere if shaken out of a bag. The blacking can be put on with camel's-hair brushes. The two plans of blackening may often be advantageously used upon the same mould. The plan of rubing the blacking on dry, and going over it with the molasses- water or beer, does not dampen a mould's surface as much as blackening the mould with all wet blacking, similar to the blackening of a dry-sand mould. The reason why these two plans of blackening will sometimes work together is because, in drying the mould b}' pan or sheet plates, etc., there are some parts which will naturall}' receive more heat than others : by using judgment in dampening the facing, in connection with the adoption of the modes of blackening, all parts of the moulds are more apt to become dry at about the same time ; while, if the fire acts much upon some one part after all others are dr}-, there is danger of some places becoming bui-nt, which is avoided if all the parts become dry at about the same time. After a mould has been blackened after the plan of a dry- sand mould, some make a practice of sleeking them. This is not the safest plan to adopt in every case. By sleeking the wet blacking, a smoother casting may be produced ; but unless very carefully done, there is more or less danger of the sleeking causing scabs. If blacking is used thin enough to not clabber, and the coats are put on with fine camel's-hair brushes so as to show no streaks, castings will result about as smooth as if tlie moulds were sleeked, and the danger of scabs caused by sleek- ing is avoided. In skin-drying moulds, methods must be adopted best suit- ing the work in hand. For instance, some moulds, such as anvil blocks, etc., may be dried by setting in them a square 172 SKIN-DUYTXO fJREEN-SAND MOULDS. or round kettle; then, nijain, some moulds may ho dried by means of Jhit, oblong, ov scjiiare puns. Often there are nioidds where neither of the two plans will answer, because they are so shaped that the kettles or i)ans cniniot be well used. Thin sheet-iron plates perforated with small holes are often used by laying them over the mould. This plan is, as a general thing, seldom used when kettles or pans can be utilized. 'i'hc fuel general!}' used for drying is charcoal. In firing with it, the heat thrown off should be mild and steady, espe- cially upon the start, since too strong a lire is apt to blister or buin the mould. Sometimes the cope and nowel may be dried together, by having the cope propped up clear of the nowel, and the the between them. Then, again, the mould may be such as to admit of its being closed while being dried, the riser, etc., being left open to let out the steam. Greeu-sand cores are most advantageously skin-dried by placing them in an oven ; and, as in drying the moulds, the heat should be kept mild and uniform. To ascertain if a mould or core is dried deep enough, either cut a small hole into the surface with some shaip tool, or press the surface with the fingers. The hardest places to diT by pans, etc., are the corners. The sides of some moulds might be burnt to pieces before the corners could be dried. To get the corners dry, it is often necessary, after pan fires have been taken out, to place some hot coals around or in the corners, or to dry them with hot irons. Ik'fore a novice undertakes a difficult job, he should have practised upon minor jobs, which, if spuiU^d, entail but small loss. Experience, coupled with judgment, is necessary, to be successful iu drying moulds so as to turn out good castings. SETTING AND CENTllING CORES. 173 SETTING AND CENTRING CORES. If there is an}' one thing that a macliinist dislikes, it is bor- ing out holes that are not cored centrally. And Avhy a moulder cannot always set cores ceutrall}', is something of a conundrum to him. Moulders can sometimes make excuses, and are gen- erally ready to take a cast-iron oath that the core was set right, which the machinist, of course, cannot dispute. Still it will, I believe, be acknowledged by nearly all moulders, that excuses for^ores out of centre are about the worst kind of excuses we have to make. When 30U get a moulder so that he cannot say any thing, it is a sin to torment him further ; but it isn't often that you will get him there. Why it is that all cores cannot be set centrally, is something that cannot be fully explained. To set a core centrally or straight, does not generally call for any great mechanical skill. What is more demanded is care and thought. The accompanying engraving may assist in showing why some holes do not come central, and perhaps afford some help in setting cores. The cause of holes being out of the centre is not always on account of the cores not having been set centrally. There are many things, such as uneven closing of flasks, bad-fitting flasks, ill-lilting cores, etc., any of which may result iu crooked holes. Some readers of "The American Machinist" will remember about four years back discussions upon core prints. To my mind, the elaborate systems that some writers advocated had but little to do with what prints are practically used for. Prints are for no other purpose than giving bearings and holding cores. 174 SKTTINf; AND fKNTKIXO PORKS. Gate otrt rcitt Rod Jivttuin linting Fig. 63. And whether bottom core prints are tapering, as seen iu SETTING AND CENTRING CORES. 175 "Third core," or straight as shown in all the others, has but little to do with the holes being out of the centre, which 1 think is a vital part of the question. The top prints are the essential ones, and even they often have but very little to do with poor holes. As a general thing, long top taper prints afford a better chance for the cores to accommodate themselves to their centre. For long cores, where there is uncertainty about meas- uring, such prints as are shown with "First core" are reason- ably certain of providing for a true hole. This form of a print is the most reliable one that can be used, as with it the moulder can see whether the core is iu its right place or not. In fact, by the use of such a print, it would be a hard matter to get the core out of its centre. Some may think that such prints should always be used ; but they are objectionable from the extra labor and time which their use involves. It may be said that there is no gain, if, through the quicker plan, the casting is lost. But all castings are not lost : it is only the ones that are wanted in a hurry, that we generally lose. The cause of man}- crooked holes where the popular common top-tapering print is used, is as illustrated at "Third core." As a general thmg, nearly all pulleys, gear wheels, etc., have either feeders or pouring-gates on the hub. These holes break away, or weaken a portion of what should be a firm, true, sound, tapering print ; and the core leans to the weak side, with the result shown. There are a few moulders who always close such moulds as the above, with the gates in place, as shown at "Second core." By this plan it is evident that much of the risk is lessened. The gaggermg and nailing around the second core is another safeguard against the core crowding the print to one side. Another point that will bear looking at is that of the taper ends of cores. A large numlDcr of foundries make all their common sizes of round cores, without having the taper end formed. Then, again, a large number of shops make their cores having the taper end on them. 17li .Sl'.IlINd AND CliNIKlNf; COKMS. I'ilinir :i 1:i|kt on cores is often very olijc* lionalilo, cspo- ciiillv wlitii iltHic l>v a <-:iri'lrss nionldiT. A fair illiistiulion of cuivli'ss work of lliis kind is slmwn on coif />. // ijood inamuje- ment. A careful moulder, wlieii filinif a taper print, will compare his print and core together, as .seen at I<\ thereby making a core that is likely to fit and fill the female print S. Cores made with taper prints arc often one-sided from being laid down upon plates to iby ; allhouuh .some. \>\ nailing the heads, or packing sand under to hold them up. will get a very fair tapered head. In some shops, in ordi-r to get good round straight cores, they are dried in half-round iron lioxes of the same diameter as the cores. This is the only reliable way of making true taper-end round straight cores. The plan is an old one, and the cause of its uupopularity is the expense of making the cast-iron l)0xes. At £", L), 11, and at the square, are shown the methods generally employed in setting and centring cores. With the exception of £", the plans are popular, and call for no couunent. i^ is a plan whereby cores in such moulds as green-sand pro- peller-wheels, and others having no points to measure from satisfactorily, can be centred. The plan is simply the driving of three or four stakes outside or away from the mould, before the pattern is drawn ; the stick E being placed against the pattern print, and all the stakes driven the length of the stick away from the print. The core, when set by the same stick, will of course occupy the same position in the mould as the print occupied on the pattern, and therefore, as far as centring is concerned, must be right. The plan is applicable in other ways than shown ; and, while the idea may be old to some, it will be new to many. The carrying-off of such core-vents thiough the cope ia SETTING AND CENTRING CORES. 177 often the cause of weakening top-prints, and also the cause of blow-ups, from metal getting into the vent. To avoid this danger, it is often the better plan to carr^' off the vent through the bottom board, or b}' running a long vent down into the moulding-floor, as in the case of bedded-in moulds. Some will say, That is all right, providing you have a cinder- bed under the mould. I have carried off the vent of cores as large as one foot in diameter, and more in length, by simply driving a ^" vent rod down three or four feet in the sand below the mould. Green sand, where you have a large body, is capa- ble of carrying off and holding more gas than is generally thought ; and to such as have never tried this plan, I would say : " Do so," for I know they will find it an easy and good way of carrying off ordinary-sized vertically set core-vents. If the cores are small in diameter, and long, — for instance, say 2" diameter by 24" long, — it would then be best to take the vent up through the cope in concert with what would pass downwards. Very long, small-diameter cores cannot be too well connected with outlet vents, as the vents from such, espe- cially if quickly surrounded with metal, require to have a fast delivery. In the upper part of the cut are represented ideas that to some may be of value. It represents the finning or chamfer- ing of core-prints, in order to prevent the crushing of flanges, etc. At P and V there is not the chamfer which is seen at A and R. Many moulders seldom think of chamfering a print, and to the credit of such may be placed many bad castings. Chamfering core-prints should be performed upon the same principles as finning joints ; and whenever the print is short, or the core too heavy, there should be bearings to assist the prints, or chaplets, in holding the core, placed as represented at TT. The greatest cause of flange-crushiug is probably due to the irregularity and over-size of cores. Should any of my mould- 178 SETTINfi AND fKNTIUNO rOIlKS. ers lose n casting throu^'h the altove cause, I should hold them rcsponsihle, allliDUj^h 1 lui.ulit reprimand the eort--inak»'r for making the core too lar<;ce. In our shop it is the custom for all pipes or moulds having flanges upon them to lie tried off ixnd on in concert with calii)ering the i)rint and core. This gives the moulder a chance to see if there is any liability of his mould or flanges being crushed ; and, if there is, he has time to tlien remedy any evils tliat might result in a l>ad cast- ing. The practice of thus trying off and on all such moulds has been the means of saving many a casting from going to the scrai»-pile. I think it is a safe assertion to make, that in not one of fifty shops is this the rule. 1 know that it takes more time ; but will say that in our shop I have yet to see a casting lost thiough having a crushed flange, — a thing that but few foundrymen can say. Having all castings good, far more than balances, in dollars and cents, the little extra time taken up in trying off and on all such copes. IMPROPER SETTING AND WEDGING OF CHAPLETS. 179 IMPROPER SETTING AND WEDGING OF CHAPLETS. It is not uncommon to see castings lost from improper setting or wedging of chaplcts. The work lost may be liglit or licav3% The little error that will cause the loss of a casting worth one dollar would cause the loss of one worth hundreds of dollars. In selecting this subject, there was no thought of pre- senting improved plans or ideas ; but, if possible, to show how castings may be and are lost through want of care or judgment, — not practice, for moulders that have worked a lifetime at the trade can be found who are no more expert in this respect than apprentices. The mould chosen to illustrate this subject is a large piston. The number of cores in it is eight ; for each core there are three cope chaplets required ; so that, altogether, we have twenty- fou- obaplets to be set and wedged ; and, should any one of thib number be wrong, the result would be a bad casting. .Often there are moulds where the vent of some core can only be taken off through the bottom of the mould ; and the core may be of such a form as to have only a small bearing on the sand and the rest on chaplets. This core may be the only one in a mould, to lose which would involve hundreds of dollars. When setting this core, the moulder is very careful that the vent portion has a solid air-tight joint ; for, if the liquid iron should find its way between the joint of the core and mould, all would be lost. After the core has been carefully set, the next important tiling to be done, to insure safety, is, after the cope is on, to wedge down the chaplets, so that the head of iron ISO I.Ml'KOI'KU SKTllNf; AM) \Vi;i)(;iNf; OV CIIAI'LHTS. oamiol r.'iisc up tlio ooio, and :illow tlio iiictnl to {X^'t into tlic voiil. Al i>, on the ligliL-liaiul hide; of tla- fill, is uu illuslra- tion of bow a great many cores arc dungeroubly \vcil;j;cd down. As this chaplct is shown, with its wedge and blocking, there IMPUOPER SETTING AND WEDGING OF CIIAPLETS. ISl are three things that could liappcn which would allow the core to rise up so that the iron could get into a bottom vent, as shown at D. It may be well to remark here, that, whenever it is possible, vents in cores, similar to the one shown, should be arranged to be let off through the cope, as shown on the oppo- site side at H. There is always more or less danger in taking off the vent through the bottom. These remarks are intended more particularly for the draughtsman and pattern-maker, who should always reineml^cr that in moulding there is generally more or less risk, and that they can very often greatly lessen this risk by having a little thought for the moulder's interests, as well as for their own. Going back to the sul)ject : Suppose the mould shown is being poured ; the liquid metal is rushing through the runner or gate ^1, and the head soon becomes high enough to exert a pressure up and against the chaplets. Now look at chaplet B, and then at chaplet K. It will require but little observation to decide which one is liable to let the core rise sufficiently for the iron to run into the vent D. (The chaplet W, between B and /r, would not in actual practice be used. It is only shown there for the purpose of illustrating the ideas.) Now, the chaplet B is by no means an exaggerated illustration, but a true sketch, representing the wa}' a great many chai)lets are secured, sometimes on jobs upon which a great deal of money and labor have been expended. The lirst noticeable weak point is at 2. Here we have onl}- one point touching or resting on the core. This chaplet-hoad may be stiff enough to stand as it now is, but its chances are very slim indeed : if it does bend when the liquid iron makes it hot, up comes the core sufficiently to let the iron into the vent. Again, suppose the head will not bend : it is plain to see, that, in the way the wedge is placed in relation to the head which rests on the core, it would not require a very great strain to tii) up the rail, so as to liecome loose with the wedge 3, thereby letting the core rise up. There is still 182 iMrrjirr.K sr/niNf; and wr.DniNr; or f'li.M'LKTS. aiiotluT |iossiliility of this t-oic rising np. We will suppose tiial lu-illuT of llic nitovr ri'siilts should occur, and tliiit the chaplct will stay ill the position shown. Over the top (jf this chaplet and wedi^e is a railroad-har. This bar is held dcnvn hy having a wedire, I, placed between it and the cast-iron beam X. Now, the way this chaplet is placed to the outer edge has been known to allow a rail, or similar bar, to partially roll over. Another feature frecpieiitly seen in wedging down the bars of a cope as well as chaplets is, that, instead of putting about an equal number of wedges on each side of a bar, they will all be placed on one side, and that most likely the weakest side, as at 4. At *S' and T may be seen some very fair illustrations of the wa}' many unaccountable bad results are accomi)lished. A chaplet wedged, as shown at S, will often cause bad results. When a heavy pressure comes uix)n a chaplet thus wedged with cast-iron wedges (which arc generally used), they will frequently lircak, on account of the two faces of the wedges not coming well together, as is shown, and thus allow the core to rise up and make the casting thinner than it should be, or allow the iron to run into bottom vents (should there be any), and cause a blow-up. The chaplet K is wedged in a way to be relied upon. After l)la(ing the wedges by hand, they need to be tightened. For this puri)()se a hammer should seldom be used. The hammer- ing to tighten them should be done by some lighter article than the common run of shop hammers. I once had some words with a moulder about losing a casting through bad clmpleting. lie was certain he had tightened his chaplets, and went so far as to call upon his helper to testify to his using his hammer. There is no (piestion but what he did tighten his chaplets, and the sketch of the hammer seen no doul)t shows how he used it. Some moulders will say the ''blocking " and chaplets cannot alwaj^s be arranged so as to use two wedges. It is admitted IMmOrEK SETTING AND "\^"RDGING OF CITAri>ETS. 183 that there arc often such cases ; but, as a general thing, there is just about as much mechanical skill, or the lack of it, shown in placing the upper blocking, as there is in the general hand- ling of chaplets and wedges. Some moulders are just as liable to place blocking bars six inches above the chaplets as thej' are to have the bars clear only a quarter of an inch. The first blocks the}' can lay their hands on will be used to rest the bars on, instead of making a point to study the j^^'opei' relation for careful wedging. About the proper distance to allow for wedging space between bars and chaplets is ^". The less blocking and feioer pieces that are used between bars and zvedges, the better it is for the wedging and for the safety of a casting. At W is shown a chaplet wedged in a reliable manner where the use of blocks is required. Careless moulders often lose castings by the way in which they set bottom chaplets. R and M show wooden blocks, hav- ing sharp-pointed chaplets driven in them. The block li is apt to split, caused by using too loffg a chaplet. The block M shows the other extreme. Both of these blocks are liable to cause a bad casting through settling of the cores. Often, in setting a heav}- core upon such chaplets, the weight will force the sharp points deeper into the blocks, thereby causing the core to settle and make the casting too thin. Perhaps, if the core's own weight don't do it, the wedgiug-down of the chap- lets will. "When wooden blocks are used, they are better if made of hard wood sot with the* grain up ; and the moulder will have to use his judgment as regards the proper distance to drive in a point, as the size of the chaplet, the weight of core, and nature of the block must be considered. As a general rule, ^" is a good distance. The two inner bottom chaplets shown are placed in cast-iron stands. The chaplet on the right is reliably set. The one on the left illustrates why cores are sometimes broken, or the casting •■' comes thin," by not having a solid bearing. 1S4 IMPUOrKU SKTTINf; AND WEDCiING OF (11 AI'LKTS. The wedges sliown, liaviiig (limeiiKions fjivcn, are of goo«l proporlioiis for cusl-iron wedges for geiieriil foumlry use. Many shops use wrought-iron wedges. Altogether, they are decidedly the best ; but, ou account of their cost, the cast-iron wedge is used in the majority of shops. It would be better to keei) a few wrought wedges, as there are often jobs where the cast wedges are not safe. More solid wedging can be done by (he size shown than l)y thicker ones. In wedging iron and iron, there is a tendency to slip ; and the more tapering the wedge is, the more liable it is to slip. When securing a cope, or a number of chaplets, with iron wedges, they should generally be gone over two or three times; the tightening of one wedge will often loosen others, making it necessary to go over them all once or twice after the first wedging, each time rapping lighter. The cat shown of a chai)let-stem represents one got up l»y an acquaintance, A. M. McGee, who is employed in a bolt-works, Cleveland, O. He has devised a machine for putting on the heads, and also claims originating the stem shown. Be that as it may, the plan is a good one. For slantnig cores, similar to those shown in the piston, chaplets with forged heads are not the best, on account of there being no chance for the head to adapt itself to the shape of the core. "With a stem having a shoulder and riveting-tip like the one shown, as large a plate head as desired can be readily riveted on in such a manner as to be loose or tight. Castings are often lost on account of chaplet stems not having sufficient shoulders on to hold down the riveted head wlien the pressure comes upon them. The advantage of having such a shoulder as shown is apparent. There are often cores used that require the chaplet heads to be bent, to correspond with their irregular surfaces or slanting faces. Such cores call for extra-careful work in placing and holding the chaplets. In some cases it is best, if condition IMPROPER SETTING AND WEDGINO OF CIIAPLETS. 185 will allow, to file away a portion of the core, so that straight- headed chaplets can be used, or when makuig the cores provide for this. In some foundries, round-column cores, etc., are often made flat where the chaplets are to rest, so as to allow the use of flat-headed chaplets, and give the chaplets a solid bearing. The triangle shown illustrates a plan that can often be adopted when a core has three chaplets in order to make the securing of chaplets more easy and reliable ; by the use of this, the chaplets are sure to have a good bearing, and also are not jarred by the use of wedges. To show the manner of thus securing a cope and its chaplets, suppose that we are going to get ready such a mould as shown. After all the cores are set, and clay balls placed where the chaplets are wanted, the cope is lowered down to receive the impression of the balls. The cope is now hoisted up, and the chaplet holes in the cope made ; being sure that the holes are all reamed out at the face of the mould, as shown at A". Castings are often lost through neglect of this. TFand B are illustrations of how such losses can occur. In W, the chaplet is all incased firmly in the sand. In B, the hole is ill-made at the stem-end of the chaplet ; and the one is nearly as dangerous as the other, as the chances are ten to one that the sand around the chaplets will be found dropped when the casting comes out. This may be caused either by having to push down the chaplet so as to rest on the core, or by the jarring of the chaplet when being wedged. When the chaplets are all placed in the cope as they should be, by having the hole " just easy" enough to permit the stem to work up and down in, and also made larger at the face of the mould, so that there can be no danger of breaking down the face, the cope is then ready to be closed. For uneven or slanting core surfaces, as here shown, it is a good plan to place a little flour on the cores where the chaplets are to come, which is known by the clay liS'i iMruorr.u skttinc; and wKixnNf; or fiiAi'LF.rs. niarlxs ; then when llu> coiic is lowered down, niul all of tlio eli:i[ikl.s rappL'd d(jwu soliil, the eopo is a;j;ain hoisted, when, by the impressions upon the flour, it can be known whether all of the ehaplets have a solid bearing. If all are now found to be right, the cope is lowered down to stay. In the piston mould shown, there are eight cores ; and to hold down each core, there arc three ehaplets used. The toj) ones arc made of ^" iron, and the bottom ones of f" iion. The heads on all the ehaplets are 2" square. The thickness of the metal in the casting is aI)out one inch. There being eight cores, tlieie are also eight triangle plates. These plates are now set, one u\)on every three chai)lets. The round dots rep- resent where they rest upon the ehaplets. After these triangles are all placed, there are then two plate rings placed over them, as seen at 1' and F ; these rings being kept up high enough to admit of a wedge being placed between tliem and the triangle plates. Over the top of these two rings are now placed four railroad-bars, they also being kept high enough to admit of wedging. Set at right angles to and on top of these four bars are placed two heavy cast-iron beams, as shown at X. The rings, rail bars, and cast beams are all held up by blocking on the outer edges of the cope, similar to that as shown at U. The inner and outer rings Y and F are connected by three arms, which also extend from the outer ring to reach the outside of the cope. These rings, licing made purposely for this job, were made in one casting. On top of the beams are placed all the weights needed to hold down the coi)e, if there is no chance to bolt it down. After tlie weights are all on, then carefully wedge the rails, rings, and triangle plates. It is not intended that this article should cover all the ways that castings may be lost through improper wedging or setting of ehaplets. The field for blunders in this line is too broad to attempt any such ta^k. RULES FOR WEIGHTING COPES AND CORES. 187 MOMENTUM AND RULES FOR WEIGHTING COPES AND CORES. In the article on weighting down copes (vol. i. p. 113), I wrote that it was absurd, to my view, for one to say he can figure the exact weight required to liold down a cope. I think the following examples will fully prove that there is a momen- tum, and that it is absurd for one to say he can figure the exact weight that is just sufficient to hold down all copes. To say the " statical- head " is all the pressure copes are subjected to, is to maintain that there is no momentum, and no difference in moulds or forms of gates in producing a pressure. Some argue that there are no conditions to be considered in the weighting-down of copes ; that it is simply a question of hydrostatics. To prove that there arc other conditions to be considered, and also that the momentum of the iron at the moment many moulds become full has an influence, I would simi)ly ask any one to take a pattern l^gV' square b}' about 1^" thick, common rule measurement, and make four moulds, two of them to be poured as shown at K, and two as shown at //, Fig. 65. Any one will admit that these modes of pouring cover the common manner of making pouring-runners, as first receivers of the iron, as it is poured out of the ladle. The mould K, if brought up quickly, would require more weight to hold it down than it would were it brought up quickly and poured as shown at //. Either one of them, if brought up easily, can be held down by having the cope and the weights weigh two hundred and thirty [)Ounds ; but they cannot be brought up as fast as possible, and be held down by that 1R8 iM'i.i'.s von \vi;u;irnNfi coi'Ks and cduks. >v('i<2[lit. Fuitlicr, I will iillow llic iiso of Uiiily-fivc pounds liKJif weiglil on K lli:iii llit' licrid c-:ills for; and (^\i:u tlieu the t^ pi? ■< '.!, copo will lift if the mi'lnl is liroiiulit up fnst. Both of these moulds huvf tiie full bcnclit of a v'lscv. Should cac-h bo poured Fig. 70. Fig. 71. '■i^ liifjircnt Styles of Gatiny Moulds. RULES FOR WEIGHTING COPES AND CORES. 189 fast without a riser, there will be an iucroasccl momentum, and more weight will be required to hold them down. Fifty pounds extra, added to the weight the statical head calls for, would not hold down K if brought up fast. These moulds, compara- tively, present but a very small lifting area to that which some others do. Therefore, if such small lifting areas will give the momentum shown, w^hat must w^e not expect in large lifting areas ? As supplementary to the two above examples, the engrav- ings. Figs. GG-71, are given opposite, in which is further illustrated how figuring for cope weights is but an approxima- tion, and that for practical safe working one cannot figure the exact weight required. The forms of gates here shown are those also commonly used. In Fig. 67 we have the pouring-gate higher than the riser. The result is, that neither the height of the pouring-basin nor the riser can be figured from to obtain a weight which would be the nearest to the cope's lifting capacity. The mould in Fig. G7 is supposed to have 144 square inches of lifting-sur- face, and, as shown, has a basin or pouring-head 12" above the lifting or cope's surface. This mould, were the cope's lifting pressure theoretically figured for the 12" head, would require a weight, including cope, of 450 pounds. With the style of pouring-basin and gate shown, and having a riser 6" lower than the pouring-basin, however quickly, at the last, the gates or heads were "brought up," the 450 pounds could not be raised. On the other hand, were one to figure for the weight, taking the height of the riser for the lifting or statical head, he would require half of 450 pounds, as the riser's height is G", or half of 12" ; therefore, to weight for a 6" head, we should, theoretically figured, require a weight, including cope, of 225 pounds. With such a weight, even being " brought up slow," the cope would lift. The mean of these two weights — 337|^ pounds — is about as near as one could theoretically figure for I'M) KULKS FOR WKKJIITING COl'IvS AND COKES. a safe wcijrht. By pouring so as to l»ring up llie pressure slowly, :i wt'i;j;ht of 2.J0 pounds would iiold the lifl of the (J" lifiid. In all those experiments, after the ha.sins and riser were furini'd, the copes were weighed, and weights added until the copes and all weighed as per ligures giveu. "NVhen one has a jilain surface, it is simple enough to figure the head pressure; but, when one comes to apply hydrostatics to every thing that comes along, it is different. It may, in some jobs, be safe I'MOugh to take the mean of riser and pouiing-biusin as the lifliug-head, or height to figure from. But, unless there is over six inches difference between the height of riser and pouring- heads, I would not advise, in any of the styles of gates shown, to ligure tlie pressure from the mean of the head's heights. He that will make it a practice to figure from the highest point of the pouring-l)asin as the lifting-head, and then allow extra weight in proportion as the style of mould and gates are pro- ductive of momentum, will work the most securely. If it were always practicable to pour with very hot iron, and have enough area of riser to carry off the metal as fast as it could be poured into the mould, and also were one always sure of having as hot iron as he made calculations for, the height of risers or flow-off gates would then, as a general thing, uot allow any head pressure much higher than its own to exist. The duller the iron is, the more apt, in moulds having risers, is the statical pressure to approach that of the pouring-basin's height. Often the metal will freeze at the risei-s' entrance, and then, again, it will come up the risers so sluggishly as to retard the flow. While the statical head's pressure may be that of the basin's height, it does not always follow that the mould is being strained the full height of the pouring-head ; for in some cases, if the metal is dull enough to freeze in tlie risers, its dulness is very a[)t to exert less lifting inessure ui)<)n the lift- ing surfaces of the mould. The thinner the metal in lifting portions, the less lift is dull iron liable to exert. Where pro- RULES FOR WEIGHTING COPES AND CORES. 191 portions are thick, then should the risers, through anj' cause, "flow sluggish," or "freeze up," we are more sure a lifting pressure, the full height of the pouring-basin's head, is being exerted. There are many moulds, that, were they poured direct, similar to that shown at A" and in Fig. GG, even were the risers lower than the pouring-gate, would not have their copes held down by the weight obtained from the pouring-gate's height of head. Such styles of direct-poured moulds are productive of more momentum than any style of gated moulds generally made. The amount of weight required over and above what the pour- ing-head calls for, to hold down the momentum force, depends upon how many ladles are used, how fast the mould is poured, the square inches of area that the metal will suddenly rise up against, and the distance of risers below pouring-basins. It might also be added, that these three momentum factors, to an extent, enlarge in ratio as the height of head increases above the lifting surface. A class of moulds that generally will admit of the closest figuring, or that have the least momentum lifting prcbsure upon them, are those similar to the one shown in Fig. GS. Here we have the metal entering the mould, as represented by the arrow at N. Moulds thus poured or run from the bottom take the metal the fastest upon the start, and the rapidity of filling graduaUy diminishes till the end, thereby greatly lessening the force of momentum or strain upon the mould. If it were practicable to pour liJce moulds, having the same sectional area of gates and heads of like height, some to be run underneath, similar to Fig. 68 ; the others to have joint gates, as at Fig. G9 ; the pouring-basins to be large enough in both cases to admit of keeping them and the gates full, — the joint gated moulds could be filled up the fostest, and would require the most weight to hold the copes. The reason I would assign for this is that the joint gates admit of the greatest velocity in flow. 102 IIULES roil WKKJHTINC rol'KS AND CORES. Mct.ll will more roadil}' flow into air-space than into a body of metal. The hii^'her the heads, and the nearer a level the metal in hasin and mould approach each other, the more ap- parent this becomes. In pouring higli, vertically cast moulils entirely from the bottom, we can often see the top portion till up so slowly as to cause fears of its not running. Some might here say, the reason for the mould's filling so much slower at the last w^as simply caused by there being less bead pressure at the cud than at the beginning. "While this, of coui'se, is cur- Illwytiutionof Iltad Weight iiL Jit 'tt t It 1 1 114/ a 1 low Fig. 72. rcct, there are two other factors which help to retard the flow. One is the decreased fluidity of the metal, the other its weight. As an example to illustrate how the metal's weight will retard its flowing, the annexed cut is given. The example is simple, and the experiment readily tried. As seen, the metal's highest head is at F, and it escapes through the outlet W. By i)0ur- iug a steady stream into the basin (which, by the way, is far enough from the upright Fto prevent any effect such as direct pouring into would cause in adding to head force) , and keeping it full, we will, with a head of 12", as seen, throw up a stream about 8" high. Now, were it not for the resistance of the air, RULES FOR WEIGHTING COPES AND CORES. 193 friction of flow through the gate, and the weight of metal en- deavoring to descend, the head F would throw the stream as high as itself. In about the same ratio as here seen, docs this element retard the head's influence in rapidly filling bottom-poured moulds. After the metal has risen to that height in a mould to which the mould's gate or runner head could throw a stream into air-space when first started, then the further filling-up of the mould is more due to that non-momentum element in equilib- rium of liquids that is exercised in low heads settling to a dead level. The decreased fluidity of metal mentioned has also much to do in preventing heads suddenly finding their level, and causing momentum. The duller iron is, the more cohesive it is. It can become so cohesive in a mould, while being poured, as to entirely stop the flow before the mould is filled. As it is true these elements in a greater or lesser degree decrease momentum, it cannot but be seen, that with pouring- basins, and underneath gates similar to that seen attached to Fig. GS, the force of momentum would be greatly decreased. In fact, when it is practicable to use such underneath gates combined with pouring-basins, copes will, as a general thing, have but little momentum force exerted upward against them ; and if the weight obtained from figuring the gate's statical head and mould's lifting area (allowing a cubic inch of iron to weigh .26 of a pound) be placed upon an}' ordinary weight of cope, there will be no danger of its lifting, even were there no risers to indicate when the mould was full. A point which it may be well to draw further attention to here is the effect of directly pouring into runners, instead of first having the metal enter a basin from which it then flows to the runner, as seen at //, Fig. 65, also Figs. 68 and 69. The only difference between these pourers and those of lu Fig. 65, also Figs. 66 and 70, is, one has basins, and the other has none except the end of the runner, as at li, is enlarged. Pouring V.il Kri,r.s yon wi;u;htin(; cori-s and couks. diroot into ninnor-ijMtcs. to :in extent, often ^ivos the momen- tUMi lu;el-i>ivs.sure e(iiuil tu :i1mh).sI wluit it w/ the ireiyht of a cubic inch of iron. To obtain the statical pressure, for instance to the cope of the plate mould, 12" x 12" referred to above, the following example is for a twulve-inch head : Length of lifting surface I'J" Width of lifting surface 12" Lifting area 144" Ileiglit of gate 12" Cubic contents 1728" Weight of a cubic inch of iron .26 Statical pressure 449.28 lbs. In figuring up the pressure necessary to resist in holding clown copes, there arc often certain cores which have to 1)8 taken into account. The amount of pressure partly immerged cores will exert upwards dei)ends upon their de2)th of lifling area in the lirpiid iron beloio the toj) of pouring-gate or head; and after a core becomes wholly submerged, its lifting force cannot be increased. The difference in weight it would require to hold down a core two feet below tlie surface of a body of metal, and that required to hold it down if just \" or so below the surface, is practically nothing. Any rise of pressure is only attainable while the core remains parti}' immerged. To illustrate these points, there is shown a submerged core, as seen in Fig. 73 ; also a core uot submerged, as seen in Fig. 74. Supposing the RULES FOR AVEIGIITING COPES AND CORES. 199 section (Fig. 74) to be eight feet long, it would require 6289^%^ pounds weight to hold down its statical head-pressure. Adding t t : 13 1 5_ l' 2> 6:. f~ 1 r %' ^v' N • Actxtal Section of Average Section of Head Pressxcrc. Head Pressure. Fig. 73. the cope's and core's weight to the 6289y%2_ pounds, would allow such a poured mould plent}' of margin to overcome any Fig. 74. momentum of the lifting force. The following two examples show how the 6289^^0 pounds was obtained : — Length of lifting surface "Width of lifting surface Area of lifting surface Height from bottom of core to top of basin . Cubic contents Weight of a cubic inch of iron Statical pressure 96" 12" 1152" 18" 20736" .26 5391.36 lbs. 2<>(l KULKS l-Olt WKICllTING COI'KS AM) CORKS. As this only ^ivcs tlu; core's liftiiij; force, the copes must }»e added to olit.iiii the total liftiiii: power. The metal has a lift- Wi'j, surface of ,'{" upon t-aeli side of tln' coie ; and tlie deplli of tlie cope heing G", we havi', therefore, the folhnviu;^ for the total lifting force of the cope : — Lengtli of lifting surface 00" Width of lifting surface 0" Area of lifting surface 570" Height from bottom of cope to top of liasiu . . 0" Cubic contents 3450" "Weight of a cubic inch of iron .20 Statical pressure of cope 898 56 lbs. Statical pressm-e of core 5391.30 " Total statical pressure ' . . C2.'S9.92 " The above being illustrative of a core not submerged, \\c will next notice the conditions of a submerged core, illustrated by Fig. 73, which is a section of a pipe or column which we will suppose to be 12" outside diameter, the thickness of metal 1", length of casting 8 feet. In such a mould, it may in a sense be said, that there are two heads to exert a lifting pressure, one being that of the cope, and the other that of the core. From the joint up to the top of the gate, is the lifting pressure of cope. The lifting pressure of the core, as it is submerged, is the number of pounds of metal its body di>vill no (l<)iil)t appear very crude, it will, with inactiee, enable one to hecoine very proficient in tiie art foiiKS. plate Jiiid hinders s1iowo perfect; and in many cases there would lie no time lost. Of course brushing (ilT the icittern Hist allows it more time to dry. Nevertheless, if a jTood print is desii'etl, liy havinir a (lr\- jialtern, and folhjw- in^ the above rule, it will nut be the moulder's fault if it is not obtained. A blacking that will not print well can often be sleeked ; and, in many cases, charcoal is as benelicial in help- ing to finish a mould which is sleeketl as one which is printed. Charcoal is valuable in either case as long as the mould can be linislied before the charcoal becomes dam[), but after that more or less trouble may be experienced. "When blacking sticks, not only does it cause vexation and loss of time, but is often the cause of rough or scabbed skinned castings. In the Cuyahoga Foundry, we often have large green-sand moulds, which take a man half a day to sleek the blacking ou them : were the blacking stick}', much trouble would be ex- perienced, no matter how much wc might try to "doctor" it with charcoal. Where it takes a long time to sleek a blacked green-sand mould, and the blacking becomes sticky, we find the dust of silver lead an excellent thing to use, and wc often use it over ordinary blackings whether it is sticky or not. With a few foundries it is becoming quite a practice to coat their moulds entirely wnth silver lead ; and these moulds, when done, will shine like a mirror. The lead is chiefly used on ac- count of its peeling qualities. In putting it on a mould, many use camcl's-hair brushes ; and, again, others will shake it out of a bag, or throw it on by hand. Of course lead is expensive, therefore it is not apt to Ijc very popular. Not only is charcoal good to assist bag-dust sleeking, lint it also is an excellent article to have on hand for mixing with wet blacking. It frequently happens that blacking contains sub- stances of a very close or non-porous nature. These will often cause " blacking scabs." The introduction of a small jiro^tor- ELEMENTS AND MANUFACTURE OF FOUNDRY FACINGS. 211 tion of charcoal will often remed}' this ; as the particles, being very light and porous, open u[) the pores of the mixture so as to cause the metal to lie more kindly to it. The use of blacking is simply to coat the surfaces of the mould with graphite or carbon, to prevent the heat of licpiid iron from fusing or eating into the sand. Moulding-sands are composed more or less of silica, together with smaller quanti- ties of potassa, lime, magnesia, oxide of iron, etc. The po- tassa, lime, magnesia, and oxide of iron, are the parts that fuse. They combine with the silica to form silicates, or a kind of glass, which, upon heavy eastings, may form a scale from -^y to \" thick, where the sand is not thoroughly protected with a coat of carbon, or, connnonly speaking, ])lacking. All black leads consist chiefly of carbon ; the other ingredients being alumina, silica, lime, iron, etc. The freer leads are of these latter ingredients, the more intense heat will they stand before they will fuse. There are some leads, it is said, that no heat will fuse. As all good blackings are composed more or less of graphite, or, commonly speaking, leads, the reader will readily perceive the cause of their preventing liquid iron from eating into the surface sand of moulds, and why the}' provide for smooth-skinned castings. Of course, it is to be understood most 1)lackings are but partly composed of leads. The more lead blackings contain, the better they are for peeling. This applies to loam as well as to green-sand moulds. Consequent- ly the larger per cent of graphite or lead blackings contain, the better. But as these blackings are expensive in proportion to the amount of graphite they contain, and as many foundry- men overlook quality to buy cheaply, it offers a premium to the manufacturer to use cheap materials in order to make clujap prices. The cheaper blackings are composed principally of Lehigh, coke, or gas-house carbons, with additions of various minerals, and contain little or no leads. Lehigh, coke, and carbons are seldom giouud pure. The particles, as generally 212 i:i,i;mknts and MANri'ACTUiiK of i-olnduy facings. li<)\v(lL'riht it will float or wash. ^1(1 KI.KMKNTS AM) MANriACriKK OF FOl'NDUY FAf'INf;S. A imxliict iiol yet iiiciilinind, uiiil one soiiu'tinu's ustd in foimdiiis and niixtmvs of blackings, is soapsloUL'. Tliis is found in many of the Slalvs, and is I)y some foundiii-s nst-d (Hiitc extensively, wliile others condenni its use on account of the light color or skin it gives to castings. It is a well-known fact, that, although moulders use Macking every day, l>ut very few have the least idea of its manufacture, or properties which cause it to peel castings. This article may be effective in drawing attention and study to a material which, before it can be intelligently purchased or used, requires some knowledge of its coustitucuts and mauufacture. WELDING STEEL TO CAST-IRON. 217 WELDING STEEL TO CAST-IRON, AND MEND- ING CRACKED CASTINGS. Can wrought-iron or steel be united to cast-irou? is a ques- tion that is sometimes asked. Either can be so united, but in the case of wrought-iron tlie union is so weak that for an}' purpose requiring strength it is useless. With steel, the point of union will be stronger than the cast-iron : at least, I obtained such results in experiments upon different brands of steel. Uniting steel or wrought-iron to cast-iron, l)y the process here set forth, is, as far as I know, original. I have made many inquiries from well-informed parties, and all say they never heard of its being done before. The principle here involved, of uniting steel to cast-iron, is similar to that which foundrymen call "burning;" and there- fore the strength of the union will depend- greatly upon the shape to be united, and on the plan adopted for uniting the pieces. In "The American Machinist" of Jan. 15, 1881, is the writer's first attempt at mechanical literature, in the shape of an article upon "Burning Heavy Castings." This article, also seen vol. i. p. 267, sets forth the proper principles to adopt in mending, or burning, when its adoption is practicable. In the cuts shown with this, Fig. 82 illustrates the old- fashioned style of burning, and the one which, even at the present day, is quite generally employed. This, in some cases, is excusable ; but it is a poor i)lan to use it for every job that comes along. In the first place, there is much more metal re- quired to burn or make a union ; and, in the second place, the burning or union is seldom so thorough as by the plan illustrated 218 WKLDINfJ STK1:L to CAST-IKON. ill Fiij. ftO. At K tli(> ra!an (Kig. H2), this eatin<^ process is lost, from tiie fact that tlie falhn<^ metal can strike but a very small i)or- tion of that to he hnrncil. If u union is made, it has to be caused eutin-ly by the nietal's heat and llow ; as its falling force Viutlng Steel to Cast Iron Fig. 80. counts for bnt little. E represents the metal flowing into the cavity F, and // its flowing out. The inlet-gate being higher than the outlet, there of course is a current, upou which mainl}- depends tlie success of the operation. In jobs of tliis kind, the heat of the metal, length of How, and nature of the work, must be carefully considered in deciding as to the strengtli of the union. .Sometimes, by testing with a hammer the solidity of the parts may be determined ; but, as a geiu'ial tiling, o])servance of the points named is nlied u^ton. WELDING STEEL TO CAST-IRON. 219 In the burning of castings, invention and judgment are called for^ as the exact o2)erations auitable for one job ivill seldom do for another. In the engravings are represented four plans suggestive of ideas iu the art of burning. Having noticed two figures, Figs. 80 and 81 will be referred to. The left of Fig. 80 represents the burning of large flat surfaces. The reason the mould is inclined is to insure keeping bare the material to be burned or united to the cast-iron. In this cut are embodied two plans of gating. At S is an inlet-gate, placed Mending a C'racheJ I'lange. Fig, 81, Old Hashioned Style of H.urning. Fig. 82. SO as to be opposite the outlet P, thereby causing a flow over the entire surface. Were the surface broader than 2", there should be branch runners cut from the main gate S, to assist the spreading of the metal over the entire surface. The basin and gates, Nos. 1, 2, 3, 4, 5, and 6, give ideas to show how the metal may be made to drop directly on the surface. The inchnation given the mould should not be more than enough to insure an easy flow. In the burning or uniting of large surfaces, it IS best, when practicable, to have the plate or body which is to be united to the cast-iron made as hot as possible ; for, the 220 WKI.DINf; S'n.Kl, TO CAST-IKON. Iiottrr tlic iilrilf, llic less lliiicl iiicImI will Ik- rc'iuircd to make a wcl.l. In sDiiie casi's tlic sti'd or wnnifilit-iroii inijilil l)o hcntt'd in a, foij^e or furnace until the face to he united was al)out at a fu.sin<};-point ; then liy (|uickly jihicing the hod\' in a. ready- formed mould (which would, of course, require to he made of some such a material as loam or dry sand), and covering it with tlic C()\)e, the liquid cast-iron would n-quire little if any "■ llowinii; throuuh " in ordi-r to make a weld. Such a jirocess wonld also he commendahlc in view of its diminishing the contraction strains. The strains that must exist in large surface l)odies that are welded with such a difference in tem- perature as the " cold-l)uruii)g " process demands, cannot Init he serious, and often the cause of fractures or cracks in the welded hody. The cut of the "mending a cracked flange" shows a plan that may be applied in various forms. The section represents the actual burning of a cracked flange by Robert Watsou, in Todds & Co.'s foundry, Leithwalk, Scotland. The casting was a cylinder, one flange of which was cracked half wa}' around. The casting being placed in about the position shown, the parts around the crack were covered with cores, to prevent the metal from striking or burning other than the portion in- tended. The cores were made in short sections, and, of course, expressly for the job. In mending the cracks, it was done in sections of from 6" to 8" \n circumference at a burning. The stream of metal would fall directly into the crack, until it was seen to have cut about f" of an opening. After each burning became solid, the cylinder, having a chain around it hitched to the crane, would be rotated into place for the next operation. The arrangement at one end of the portions to be l)nrned was such as to allow the metal to run off. thereby preventing the gathering of a head that would prevent the falling stream from forcibly entering the crack. The point of '•• run-off" would be WELDING STEEL TO CAST-IRON. 221 about on a level with the highest point of the circumference. As iron runs level, the surface of the mended portion would present something like the section H. This unevenness could, to some extent, be chipped off, so as to leave a circle as near as .could be without doing injury by jarring. As soundness, not symmetr}', was the point sought, a little roughness was not objectionable. In burning such a crack, there is a liability that the hot iron will eat its way through, and leave the inside, at M, rough. To prevent this, cores could be made to the proper circle, and- firmly rammed up against it, or a loamy facing could be used at this place. As to the success of the operation, Mr. "Watson says that the cylinder, about two months after the operation, was returned to have the balance of the flange mended ; it having cracked, while the mended portion remained as sound as any portion of the cylinder. Some may argue that the first burning was the cause of the second crack : it may have been so. And this is a point that should be often considered ; for, no doubt, such burning will often cause more or less strains elsewhere. To assist in pre- venting this, the casting to be burned should be made as hot as practicable, either by placing it hi an oven, or by surround- ing it with fire or hot irons. In this case the casting was not only heated in the oven, but hot scraps of iron were placed upon it. The crack was mended during the time of one heat ; and, as Mr. Watson puts it, he never worked more lively nor sweat so much in his life. As already stated, steel and cast-iron can be united. I will further explain the cut at right of Fig. 80, as it may present ideas in arranging for the union of many differently shaped articles. This mould represents the uniting of a piece of steel 1" square b}' 12" long to the same-sized piece of cast-iron. The process was as follov/s : A pattern 1" square by 24" long is rammed up in a flask. In the bottom board, as shown, is a 222 W r,Ll)IN(l STHKL TO CAST-IRON. hole :ili(»iil 2" in (liiumtcr. This admits of A" thicknofs of sand to pii'Vciit tlic hot iron whidi iiiiis liirotigh the 1" iiole iVoin ltiiniiii;j; the hoard, and also causes more surety in piug- !j;ing-u[) ; for, were the hole surrounded with wowl, the hot iron would burn itself through, and make the i)lugging-u[) a blun- dering job. Il would l)e better in all cases if iron plates were used instead of the wooden l»ottoni l)oards. In faet, if all the llask were iron, it would be l)est. Tiie mould having been UKuU', the piece of steel is hxid in, an-ended in a pit, as shown; and then, with about a hundred pounds of hot iron, start pouring a steady small stream. Wiien within about twenty-live [lounds of the end, slacken the pouring, so as to have but a small stream, at which time let the outlet A Ije (fuicklj/ and reUabbj stopped. After this, quicken the iK>uriug until the mould is lilled up. Properly sloppiii'j the outlet is very essential; for, if not done successfully the first time, it is as inju- rious to the buraing as is slow, blundering welding at the black- smith-forge. A good plan for stopping is to securely place a l)ointcd ball of clay ou the butt end of a rammer or iron stop- per, make sure aim, and (irmly hold the hole closed until the metal is set. The gate B illustrates the use of side inlets, should it be desirable to burn more than one piece in the same flask. Tiic lurning of steel is by no means restricted to straight 1)nrs. A variety of forms might l)c united l)y simply making the moulds the shape cf the article wante 1, and then placing the steel in its proper place. The gates could be often fonned, and the mould so placed as to have the hot cast-iron flow over most of the exposed surface, to an outlet. In some cases, there might be a necessity for moie than one outlet, as well as inlet, to which there would be no ol)jectiou. The point to be aimed at is, to have a deadfall upon all possible exposed parts, and an outlet from same if practicable. With many moulds, it WELDING STEEL TO CAST-IRON. 223 would be better were they of dry sand, or loain ; a,s the drop- ping and washing of much running iron would often cause a green-sand mould to produce a rough-skinned job, and cause dirt in the mould, which would of course be injurious to the work. Cast-iron used for burning purposes should he very soft, and the temperature of the melted iron as high as possible. The reason for giving soft iron the preference is simply because soft iron does not chill as quickly as hard iron, and will retain its life or fluidity longer, and also will form a stronger union. The amount of iron to use for burning purposes will be regu- lated by the class of burning to be do)te, and condition of the iron xised, etc. For direct falling upon plain sqtuire or round surfaces, the following tal)le might in some cases be used. The tal)le at its best is but a rough approximation ; for in some cases it might be in excess, and in others it might prove deli- cient. In burning any work that needs a stead}- stream, we should have plenty of iron ; then, in the methods which have been described, and by the use of good judgment, we should be able to decide when sufficient metal has been poured in ; then the pouring can b(? stopped, and the remaining metal used for pouring other work. Tlie amount of iron that may suc- cessfully !)uru a like piece of work to-daj^ may to-morrow be insufficient. There are many things which cannot always be controlled in giving to any calculation a certainty of assured success for burning or mending castings. For a surface of 2", square or round, use 250 pounds ; 3", 400 pounds ; 4", 550 jwunds ; 5", 700 pounds ; G", 850 pounds ; 7", 1,000 pounds. Above 7", for every additional inch added to the square or diameter, add 200 pounds. This, if con- tinned up to a surface of 20" square of round, would call for 3,600 pounds of hot iron to accomplish the burning (for the square surfaces, it might be well to add about ten per cent to the given weights) . A point that might be well to mention is 22\ w i:li)IN(; stkkl to cast-iuon. lli:iL tlif jiarls to bo hiinicd should 1)0 cliippi-d, or tlic sonlo roinovod, so as to givo tho lluid iron tlio bost possible clianoo to oat into its surface. Any who are interested in this subject will lind additional infonnaticMi in the article '' Hurning Heavy Castings," vol. i. p. 2G7. Before closing it might be well to state that the softer the metiil is in the cast-iron castiug to be burnt, tlio lu'tter the chances are for making a successful weld, and also tlu- less is the risk of cracking the castnig during the operation or afterward. After the burning, the slower the cooling, the loss the danger of checks or cracking. In many cases it is well to keep the ca-sting warm as long as practical, l)y surrounding it with hot scraps of the iron used for the liurning. Tile subject of mending or burning is one well worth stud}'- iug, i\m\ one tiiat generally calls for good judgment and expe- rience. Before a novice undertakes a difficult job of this Jiind, it looxdd he better for him to experiment zcith unimportant pieces. Burning is a job that seldom can be done twice, on account of the surface losing much of its life or texture. Should the second burning be required, the fractured surface should be cut down until good inetal is again seeli ; but in all ca.ses use ViW precaution towards making thejirst burning a success. FOUNDRY ADDITION. — OVENS AND TITS. 225 FOUNDRY ADDITION. — OVENS AND TITS. Tlierc having been recently built an addiliou to the foundry in which the author has averaged ten hours per day as foreman for the last three years (the Cuyahoga Works, Cleveland, C), it has occurred to him that a description of the laying-out of the ovens, moulding-pits, and cranes might interest othere, and give ideas which would be applicable in other instances. The black border represents the outline of the foundry. The new shop, as bliown, is partly divided from the old shop by the par- tition-walls A B. In the old shop, there are four cranes, two cupolas, brass furnaces, moulding-pits, etc. As there is noth- ing except their antique histor}' that could be set forth to inter- est the reader, I have omitted showing a plan of the old shops, cranes, pits, etc., and devoted the space to showing section- views of the loam-pit, ovens, etc., of the new shop. Credit is due the president of the works, J. F. Ilollowa}', for providing for comfort, and furnishing handy facilities for the new shop ; no expense being spared to provide every thing requisite in that direction. "We have abundance of light and ventilation, steam- cranes that rapidly do the moxing of heavy loads, excellent moulding-pits, and ovens that surpass any I know of for prop- erly drying moulds or cores. Although we use slack or soft coal for the fires, a mould or core will, when dry, be almost as clean as when first put into the oven. Another important feature is, that the ovens will dry rapidl}', and still not burn, a mould or core. The three ovens, as will l)e seen, are fired from one pit. The draught flues being at extreme ends of the oven, and the channel for beat to travel being diverted from side to side, there is but 22«> ror.NDKY addition. -ovkns and rns. a small cliaiu'c for iicnt to cscapo cntcrinjr tlirotijih tlii' joints aiitl tliickiu'ss of the hoilc r-|ilatrs up into the ovi-ii before it can eiitiT tlif IliU' at J\ IL and A'. Tlif Mnuw-likc lines represent the heat passini; from the lires to tiie Hue. Tiie partitions, x , divert the direetion of the heat, anil also support the covering plates and earriage-track. A ckarer idea of |)artitions, etc., may be had from the elevation of the oven at K. The covering plates, 2, 3, 4, 5, G, and 7, arc boiler iron \" thick, cut into sections the width of the flue's partitions. As neither of the oveais is partitioned like another, the sizes of plates all differ. The core-oven plan is shown having the plates and track lairl. As will be seen, the plates upon the oiitside of the track, which are shown I)lack, arc free at any time to be lifted, in order to clean out the soot. Were the i latcs in one continuous piece, the width of ovon, then to clean out the ovens under Hues would necessitate the lifting up of the carriage-track. Where t!ie fire enters the Ih-st flue or paitition, the bojlir-plates are left out, and in their place a cast iron plate \" thick, having prickers "2" long, and daubed up with fire-clay, is used. This is to prevent the direct flainc from luckling and burning out the plates. There arc no holes whatever in any of the plates ; the heat pa'^sing through them and their joints, which, of course, are not air-tight, heats up the oven. Were there holes in the plates, they would seriously injure tlie draught of the under flues, and also let much smoke into the ovens, thereby destroying essential points to be overcome in using slack for firing. To be able to fire with slack or soft coal, and still keep moulds and cores free from soot, is something that will be ajipreciated by all moulders and core-makers that work aroimd ovens. Not only does soot make every thing look dirty, but it is more oi' less productive of rough eastiiiga. Another arrangement which I doubt I)eing found in any other foundry oven is that for [JreveiiLing smoke. L [)ou each siclc of / y^r-i — : — . . . "^\ J'/'<^\\\liuSj Floors ^N^'^'Vi Co^>^»om I- iimh 1. \Vi ■ O Fit/. «,7. TT \-\J Plan and Scctiot of Bulling Binders Section A. B. ~ )\j- "/ um 1 - ^-'-f=^ is^^j -3r-=-5j — 3- Channel ^ 7? "^ Hlvviiliuu of Loam I'it Elevation of Oven Ciii/assed for them to think '• any thing is good enough for a foundry." u ;i" LADLE AND CASTING CARRIAGE COMBINED. 231 LADLE AND CASTING CARRIAGE COMBINED. TiiK engravings seen are perepectivc and plan views of a carriage nsed in " onr foundry." Wishing to " kill two Ijirds with one stone," I devised it so as to be safe for conveying large ladles of metal as well as heavy castings. With the ladle set in the car, should any thing break, it could not fall more than two inches, nor is there any danger of the ladle sliding or falling off the car. To see a heavy ladle full of fluid iron awaj' up off the ground, does not inspire one with feelings of security or confidence, although it may be per- fectly safe. Many shops that are obliged to truck their metal and castings have two cars, one for ladles, and the other for castings. I think many of them would prefer to have one, could it be made to answer the purpose of both. The construction of the car is very simple, and it costs but little labor to make. The car was cast "open sand ; " the ladle- box was formed with a dry-sand core. The holes as B, E, JC, and F, made a good bearing for the box-core to rest upon. These holes were cast in for the purpose of lightening the casting. The pockets, as at //// (Fig. 88), are simply for the purpose of placing in arms, should a wider carriage be desired. The carriage-wheels are IG" diameter, and are cast solid. The axles are wrought iron, 3" diameter, and were cast in the wheels. Before the axles were cast in the wheels, they were used as chills to form the carriage's axle bearings. The axles at this time were S^^" diameter. After being cast in the wheels, they were turned down to 3" diameter, in order to make tliem ex- I*lan and Elevation View of Car. Fig. 88. I.ADI.i; AMI CASTINfi (WItlilAflK ff )M I;IN KD. ju'tlv rtMitrMl with the wliccl rims. :inil :iL llic smiiic time to leave a lltllc |il:i\ ill I lie lic.-irillL;-:. The caiN as si-cu al Ry is made hollow, so as to form nti oil- Fig. 87. hox in which waste saturated with oil can be kept. The leathers over the axles, shown in view (Fig- t>7), are for the purpose of keeping out dirt. LADLE AND CASTING CARRIAGE COMBINED, 238 I think the practical man will sec tiiat the plan described is one which slioiild cause a heav^- carriage to run easily. Al- though this carriage weighs about forty-five hundred pounds, two men can readily move it. For pulling heavy loads, we have a wire cable arrangement which is operated by power. The carriage-wheels might by special arrangement be made with their axles run in "anti-friction" bearings similar to the method set for thin "Travelling Crane" (p. 414), if one wished to im- prove upon the car shown. In making such anti-friction bear- ings, it must l)c remembered, verj' " fine fits " are necessary to make the bearings a success. The pci-spectivc view of car shows it loaded with one of our crane screw-ladles, which for simplicity in design and good working is worthy of notice. The hand-wheel is detachable, ftnd the rim of same is made of wood. This prevents it from becoming hot from the ladle's heat, thereby leaving it free to handle. While the carriage shown is applicable to but few shops, it may give ideas that some time will come in play in others. To state the most weight such a car should carr}', is at its best but guess-work : however, I would sa}', that, if squarely loaded over the axles, the car should carry thirty tons. 231 CUIH.Kl) HULLS, KOLL 1-1-ASKS, IILNNKKS, AND (;ATKS. MAKING CHILLED ROLLS, AND IJOLL FLASKS, UUNNLHS, AM) CATKS. In tlio oi)p;r:\viiijT will be foiiiul iihistruh-d dilTcrcnt ways of oonstiiictiiiu- lla-sks. inniu'is, etc., for making fliilli'(l njlls. For a viTV valunlilf rcaliiri' ol" this artic-lc, I am iiidrldi-d lo .lolm K. Parker of lii'loit, Wis., who, a short time since, sent nu' ;i sketch and description of a simple and novel plan, and one that may be vahial)lc for other purposes, than roll-making. I\Ir. I'arker was led to devise the rigging shown, to overcome the dillicnlty of the upper neck cracking, caused by the vertical contraction of the body of tlie roll in cooling. For this pnr- jiosc he makes what he terms a sleeve, as shown in the engrav- ing (Fig. SD). It is made of cast-iron, and is about i,'" thick when (inished. It is turned on the outside so i\s to fit eiusily in the chill. Tiic distance this sleeve sits into the roll varies from G" to 20", according to the length of roll. The upper neck of this roll is moulded the same as in an ordinary lla.sk. In clos- ing, or getting the mould ready to cast, the height of the neck is regulated by placing three scantling, or screw nuts and blocks, as rei)resented at 7), ^I, and A". The blocks A' can l)e either iron or wood, and ditferent sizes and numbers of pieces can be usi'd as reijuireiL The rolls thus made are used for paper macliinery, and var}' in sizes from 6" up to \l" in diameter, and from -10" to SO" in length. The chills are made in lengths of 20" to 30", and set one upon the other, as shown at /'. To luuuUe these chills, tlie trunnions shown at R are used. DilTerent thicknesses of chills are recpiired for dilTerent CHILLED ROLLS. ROLL FLASKS, RUNNERS, AND GATES 235 diameters of rolls. As a rule, about |" of chill for cverj' 1" diameter of roll is about rigbt. lu the instance of a roll 14" Fig. 89. diameter, the chill will be 5i" ; for one 30", Hi". This thick body of iron is not for the purpose of resisting the pressure 2.''li CIIII.F.I'.I) ItOM.S. ItOI.I, ri.ASKS, IirNNKUS, AM) f;ArKS. (liic to the Ik'.kI, lull t^) circff .'i ticcp chill frcMii tlio siirfnrn of till' casting, mill to |tir\iiit ilic cliill liom cnicUiiig, ivMiltiiii; from lliL' siiifuci' bring stidfU'iily hcjitcd. The foUowiiiLT i^ :i Uihie giving the thickness of chill for rolls ranging from 1" diamctAT to iJO", and varying in length from oiu" foot u[) to that ix'(|uiix'd for the common lengths of rolls made. I>IAMETKK TlllCKNKSS DiAMKTKK TlllCKNKSS ! DiAMKTKK TlIK KNEMU OF KOLI,. OF Chili,. OP KOLL. or Chill. OF Roll. OF CUILL. 4" 2" 13' 41' 22" sy 5" 2r 14" 5}" 23" H' ('." 3' ir>" 5^ 24" 9' 7" sr IC G" 2.r 9J' 8' 3^" 17' 2(;" 9J" 0' 31" 18" Of 27" lOJ" 10" 19" 'i' 28" lOi" w 41" 20" 7i" 29" lor iL'" ■\r 21" 71" 30' Hi' The diameters 4", 5", 6", 7", and 8" are not, as will be seen, flgnred upon the basis of §" per 1" given, for the reason the body of the chill wonld then be a little too light for safety ; or, as stated al)ove, it is the sudden heating-ni) of the chill's sur- face, and not the pressure of the metal, that we have, in great measure, to contend with. The smaller rolls have nearly the same inlluence in suddenly heating the surface of the chills as the larger rolls. Suddenly lu-ating the surface of course ex- pands it : therefore more or less strain nmst be exert«^'d upon the cold iron back of the surface. From tliKs cause I have seen car-wheel chills fly in two pieces before the mould was lialf full of metal. I think the moulder will now see why the author did not adhere to his basis of ^" to the 1" for the small sizes of chills mentioned, and the advisability of making the CHILLED ROLLS, ROLL FLASKS, RUNNERS, AND GATES. 237 smaller-sized chills thicker in proportion than those above the 9" in diameter shown. In making chills, the best of iron should be used, or they will uot last long enough to pay for the making. The surface of a chill becomes rough from use, then checks, and eventually is useless. Often, in breaking them up, from the surface to an inch in depth the iron is found to be burned. "When chills are made in sections, to make different lengths, or for convenience in handling, the joints must be made true and tight. For clamping together, flanges can be cast on, as shown at Y. Mr, Parker, for securing his flask, uses two long bolts and a top ring binder, shown at N. This binder, being placed on top of the sleeve , is bolted to the bottom plate by bolts E E. Should it be desirable to use such sleeves independently of the lower part of the flask, lugs or handles could be cast on the chills, and the sleeve held down and operated by means of the bolts shown at F F. The lower portion, or neck of the rolls, is moulded as shown at the right-hand side of the cut. The flask parts at Fto allow for making a whirl-gate, as shown in the plan of '^ joint ^^ " of thesmall flask (Fig. 90). For ramming-up the pouring-runner X, Mr. Parker uses a cast-iron pipe, the arrangement of the nowel being similar to that shown in the details of the small roll flask. Black-lead is rubbed on the chills to prevent the iron from sticking, and the rolls are poured with hot non. Some men, after the chills have been taken out of the oven, where the}' were placed to be heated for casting in, wash the face of the chill over with a thin coat of blacking, composed of ordinary blacking wet with molasses-water. In order to economize space, I have shown, attached to the cut of Parker's flask, another device sometimes used in pouring such jobs. W W are plan and section of a basin which can be connected 2.'i8 riiii.i.i:i> uom.s. uoi.i, ii.asks, inNNKUs, and catks. to or cast on tlio end of oitlicr a square or roiiml niiincr-pipe. li reitresents a i]nart('r-lnrn pipe uv box, jointed to tiie runiu-r- pipe and tlasU al S S. Tliis anan;4cMicnt saves llie work of parting llir tla>l< to ^ate the mould. At 1, ll, ."). and I, is shown the manner of eonstnieting the elI»o\v in halves and liolliiii^ 'together. Tiiis permits of its Iteini; taken apart, shoidd there be any trouble in getting out the c:isling, or from the breaking of gates. It would be safer to have the lower joint S secured by bolts; the upper joint S can be secured with clamps if desired. The small chill flask represented in Fig. DO is a very conven- ient one, and its construction embodies ideas that are ai)i»licable to other jobs. At K K is shown a sectional view of the guide-iings made in chills. These are all turned out exactly the sanie diameter. R R represent grooves turned in a citst-ircMi '' mould-board." This is used to ram u^) the cope and nowel on. There is also turned in it a recess to centre and hold the neck pattern. By the use of this rigging, there is no possiljility of the ueck getting out of the centre iu closing. With the exception of the single-whirl gate, "joint EE'' shows the plan or top view of the nowel section J/. "Joint B B" shows the bottom view of section M. Numerals 2, 3, 4, and 5, on the plan view of "joint S S," represent lugs by which to clamp the bottom-plate. At joint B /J, the guide-i)ins are made to serve the purpose of bolts or clamps. Holes XX are for clamping and lifting the chills. The runner-l>ox has a loose plate made in halves. To hold this plate, two straps. 7' T. with tlireads, are used. Gating ciiillcd rolls is always a point of i)rominent consider- ation. As a rule, the hotter, the faster, and with the more uliirl, the iron fills the mould, the cleaner the chilled face. The temperature of the iron must, liowever, be regulated by a Fig. 90. CHILLED ROLLS, ROLL FLASKS, RUNNERS, AND GATES. 230 consideration of preserving the chilled roll from checking or cracking. I })refer getting the gate as near as practicable to the body of the roll, as by so doing the whirl of the iron is increased. The iron should be poured as rapidly as is possible, without any stopping. In the cut (Fig. 89), as shown, one basin is represented higher than the other. Some prefer the lower one, — so as to make sure of not filling the feeding-head too full, thereby leaving room for hot feeding-iron. Others prefer the higher basin W<, as giving more force to the metal entering the mould. By careful watching of the moulds, large basins of iron, which are uot conveniently melted in the absence of an air- furnace, arc avoided. It is only the chilled portion of the roll that rcHpiires rapid pouring or filling ; so, witli long-necked rolls, the pouring can be slower toward the last, giving a better opportunity to " watch up " the rise of the iron. V>y using the double wliirl-gate, shown at " joint BB," nearly double the amount of whirl is given the iron. With this double gate, I have seen dirt gathered to the centre in a ball nearly 3" m diameter. This would rise up through the neck into the feeding-head in a solid body so as to admit of being taken out, leaving nothing but clean metal in the head and casting. Had the mould been poured without the whirl-gate, this dirt would have been great!}' scattered, and lodged against the surface of the roll or under the upper neck. This whirl-gate is useful not only in making rolls, but often for other classes of castings, especially those cylindrical iu form. 240 MULLDLNU MACHINES. MOULDING-MACHINES. AViTH somo classes of woik the use (;f :i macliino for assisting the moulder in inakin( |iMltcnis. Ill ii>iii;4 this m:i(liiiir. (hr in<*iili|ci' siiiii»lv iidjllsls tlic iiiocket-rule8 commonly used. •2[u lUillVAl-l-.M AULAS \()\i I'orUINfJ fiATF.S. TABLE FOR K(^ri VALKNT Ai:K\S I\ SC^'AKK AND ItKCT- AXGULAK GATES To THAI' OF KUl\\I> <;ArF:s (sec note on ]). JJ.")). UoiM) <}ATE8. Sqcarb Gates. KECTi.NCill-AU Gates 1" Thick. Uk.CTA.NOLLAll Gates 14'' Thick. IvELTANULLAIt Gates 2" TuicK. ICbCTANOlLAK Gates •2\" Thick. r = V iY' = H" n" = lA" ir = ii^' = 1"X 2^" If = 1"X OJ" = ij"x ^^" ■2\" = 2" = l"x 4" = U"x 2|i" 01" = •2^" = 1"X ;j" = 1 1 " V •>'>'/ 2J" = 2t^" = 1"X 6" = li"x 4" = 2''X.r 3' = 2H" = l"x 7^15" = l^"x 4 J" = 2"X.'3j''5'' S\" = 2r = l"x S^V' = U"x .0^ = 2"X4t%" = 2rx3^" sv = w = l"x 0|" = U"x 6135"= 2''X4t» = iV'xSj" 3f " = 3^" = 1''X11J5" = li"x 7f = 2" X .-;}/' = 2i"X4-iV'' 4" = 3iV' = l"xi2-r*5" = Ij'-x S|» = 2''x{;\" = 2i"x.5" 4}" = 31" = l"X14f5" = irx 9^" = 2''XTV' = 2.V'xr,5» 4\"^ 4" = Vxim" = IV'XIOI" = 1)11 y^o" ^:: 2\"xf)J" 41" = 4A" = l"xiT|" = n"xn\i"= 2"x8t'' = 2i"X7r 5" '= 4f." = 1"X19^" - 11"xi;3tV"= 2"xon" = 2r'x7r The term "equivalent" used in this cliapter does not iiniily that two or more small gates having a combined area equal to one large gate, all having like " head pressure," will deliver the same amount of metal per second. The flow of metal is retarded by friction, in ratio to the surface area it comes in contact with. Now, although four 2i" round gates are of equal area to one 5" round gate, we find the frictional resistance to the flow of a like "head pressure" through four 2l" round gates to be double that generated in one 5" round gate, simply because the combined circum- ferences of four 21" round gates are 3I.41G0 inches, whereas the circum- f< ronce of one 5" round gate is 1.5.70SO inches. As gates are generally combined under varying complicated conditions, the tables as given can be lietter practically used than wIkti' tlu y are lumbered with the question of frictional resistance. ERRORS IN FIGURING WEIGHTS OF CASTINGS. 247 ERRORS IN FIGURING WEIGHTS OF CASTINGS. Some of our industrial papcre having lately given much prominence to the rule of dividing the cubic inches contained in a casting by 4 in order to find its weight, the author thought it proper to state in this volume his reason for not having given this old rule among the tables eoutained in vol. i. The reason for not adopting this rule is simply because its use will give a result which is too light for practice. Before adopting the factors laid down in vol. i., the author had given the subject numerous tests, not only in carefully noting the weight of specially made castings and different grades of iron, but also in having pieces planed up to " fine measurements," and carefully* weighed. To show the " shortage " of the product obtained by dividing the cubic inches contained in a casting by 4, we will take for an example a block measuring one cubic foot. In such a block there are 1,728 cubic inches : this, divided by 4, gives a weight of 432 |X)unds. Now, the actual weight of such a l)lock (when fed solid, of coui-sc), made from ordinary gray iron, is about 450 pounds. So we find, by figuring with the divisor 4, a shortage in weight of 18 pounds for ever}- 450 pounds ; or, for every 100 pounds, a shortage of 4 pounds. The above shortage is certainly quite a serious item in figuring for heav}- castings. For example, take a casting weighing 10 tons: we find the divisor 4 would give a shortage of yOO i)ounds. The author's main reason for here referring to this old rule is simply to show its error, and prevent any one from being deceived thereby. The factors, as laid down for figuring the weights of castings in vol. i., will, if followed, be found to give answers as near accurate as it is practical to obtain. CONTRIBUTED CIIAPTEIIS. TilF. folloAvitip; fivo rhaptoi's all oiiLrinally appcanMl in " The AiiKTicaii Machinist ; " with the exception of Mv. ^lallett's, which appeared in '^ Iron Trade Heview and Western Machin- ist " of C'U'vclaiid, (). The anthor's attention wxs attiacli-d to those aiticles Ity their novelty and piactical ideas, and. think- ing they would be of much value to the readers of this book, he dccidi'd to insert them ; and would here tender his thanks to the resiK>ctive writers, especially to Messrs. Masters and Ilairi- sou for their kind dedication to him. MELTING SMALL QUANTITIES OF IRON. Bv Ruhert E. Masters, CoLiMnvs, Ga. The following plan for melting one hundred to throe hundred pounds of iron in a common ladle, I respectfully dedicate to Thomas D. West (as one of the odd methods of melting iron) for his second volume of "American Foundry Practice." I imagine I can see a smile illuminating the features of the mouldei's in some of the finely ctpiipped foundries whore they melt from twenty to fifty tons of iron per day, at the idea of melting a coujjle of hundred pounds ; still there are hundrcils of suKill shops where the knowledge of a method for doing so would be a source of eonsideral)le profit, besides sometimes retaining a customer. For instance, ]Mr. E has a small sh(tp, nud only casts once or twice a week : a short distiincc from him (perhaps in the same town), 1) & C'o. ha\e a large shop, and cast every day. E has just taken off a heat, and will not 248 MELTING SMALL QUANTITIES OF IRON. 249 cast again for several days, when in walks a customer with a broken-down job that will require from a hundred to two hun- Fig. 92. dred and fifty pounds of iron to ix)ur off, and he must have it immediately. E doesn't want to lose the job, or run the risk li."»0 MAKi.\(; A (Tia i;i) nn: i iiom a siKAKiiir i-aiikkn. of losiiiiT ;i cnstonuT. Now for tin- plan for doini; tlic joli, and iflainiuLX tin- cnslonicr : 'I'aki' a coinnion tlin'o-hiintlivd |ioini(l lac lit' J. l'"i^. '.12, (laulicd in tlic ordiiiaiv w ay. and •• II ic it ii|i " until yon iiavt- a solid coke or coal liiv. Then take a plain cylinder li. made of lip;lit l)oil('r-iroi), 'U'>" loni^, and of tin- ri<|lit diameter to lit the top of ladle. This cylinder slionld have a 2" hole at one end for tnyeie i)ipe, and should he danlxd np same as a ladle, ani|K', 'I'lu' li'iiylli of till' Jiipc w.Ms llitii iii:iii;('(l on tlic stalvc :ii>o\c, tlio pattei'M |>I:icim1 upon tlic coics, f( iir ronnd stick.s placi'tl around the slukf to help luing olT tin; vt-nt oC tin- coif. anefort' (hawing it out; Imt. in most cases, we used cores like those shown in Fig. Of), and marked c and c res|)ec- tively, as they could be adajited to either end by simply revers- ing their position. They were niadi' in segments one-sixth of the total circumference I'equired, that size being found the most convenient. When sockets were cast on any of the pipes, the cores to form them were made on the same general principle, and, for (.»bvious reasons, i)laced at the bottom of the mould. MOULDING PIPES ON END IN GREEN SAND. 255 The advantajxos attondino; this iiietliod of moulding thin pipes nre too api>ait'nt to any om- :ic(inainted vvitli the trade to require more th:in a passing notice. Ortliiiarih', by the old method, consideral)le expense and delay would Ije incurred in making a pattern and core-box, not to mention the provision of large and substantial tlasks in which to do the moulding subsequently. By careful ramming, a mould made in this way is safer than l)y the horizontal pltin. as tliere is no danger of a run-out, of hav- ing the core rise or sink, or of '"cold-shut" if the iron be a little dull. By this plan, also, two lengths of pipe can be made in tiie time taken to mould one V)y the old method, and the moulds take up less room. Of course, the moulds cannot be lilaekcd and sleeked ; Init by using fine sand, and ramming regulaily, a good surface may be obtained, if desired. In this case, the castings were not required to i>e smooth : so long as they were light and solid, they answered the purpose. A good plan to form the [jouring-gate is to take a puUej'- ling about five inches larger than the ring used for the i)attern, and when the covering cores, with the gates filed in them, ai-e in place, to put the larger ring on them, and make up the sand ail around the outside as high as required ; then cut awa}' two places in which to pour the iron from the bull ladles, draw out the ring, and the mould is ready to pour. This kind of gate has the advantage of I>eing quickly made, besides being cleaner and more easily choked than a gate cut out w^ith a trowel as ordinarily. This plan of moulding thin pipes has been adopted by other firms ; but to many, the idea will be perfectl}' new. Of course, the deeper the pattern is, the better, as there is less danger of ramming the mould in or the core out than with a shallow pattern ; besides, the pattern can be drawn more each time than the other, and leave a more even surface both inside and out. Fig. 94 represents a plan of the mould when ready to cast ; 2.')G TIIUKK WAYS OF MAKING AN AIR VKSSKI,. Fii;. !»."» ;i ciiitiMl vcitifril section of the s;iiiif. Winn tlic pi|H'S arc to lir lr>\ to ii>c xiiiir loiinil iron ll.i>ks or linf.rs in wliicli to r:ini up tin- lower end of the mould, as tin- strain is vcrv ^r»'at, and will canse the ciusting to Ik' imicL JK-aviiT tluiii riMiuired unless properly secured. THREE WAYS OK MAKING AX AIIi-VESSEL. Bv Rop.KRT Watson, Clevelaxp, O. In making a easting like the one shown in the engraving, three things suggest themselves to the moulder : Fii-st, to make it; second, to make it well ; and, third, to make it at the least expense, and at the same time have a good j(jl) of it. There arc tlirec plans represented in the engraving for making this air-eliamlu'r ; wliich, it may be remaiked, is of a size not often reciuired, the dimensions being 00" x 48" and 2" thick. The moulder wlio made this particular casting made it in loam, bj' the fii'st plan represented. This is a plan considered by some old-fashioned, out of date ; while others maintain it is the safest plan, although a rather slow one. I will explain the three plans, and leave it for the reader to judge which is the best. In making this casting by the " first plan," we build up to A and D, and after loaming and sweeping smooth it is neces- sary to wait till the loam is stilT enough to 1)ear the wi'ight of tlic core. liy drying it with a fire-basket, a little time c:fn l»c saved. Then it is l)lacked with a mixture of charcoal-blacking and water, for the purpose of making it part clean. The sweep is then changed so as to sweep the required thickness, which, in smaller castings, is often done with green sand dampened with clay water; ])Ut I doultt if this material would be strouii enouiili to sustain a core uf this size. To be ou the THREE WAYS OF MAKING AN AIR-VESSEL. 257 safe side, it is hotter in this case to use loam and hrick splinters, and to thoroughly dry with the fire-basket. Then a coat of parting blacking is put on, and it will part cleaner if a coat of parting-sand is sprinkled on top of it. 2r)S THUKK WAYS OF MAKINfi AN AIU VKSSF.L. 'I'd luiiM the core, :i |)l:ilc willi prickers <»n it is iisod to form the liuttoiii, MS sliowii ;il A'. Thciv arc «lilTi'rciit ways of Itcd- (liiili tills plate. I have seen tlieiii heilded cold in a IkmIv of loam, Imt this re(jtiii'cs a long time to diy hard cnouiih to lift clean when the eorc is taken out of its Ix-d. A Wetter plan is to lu'at the plate, and have hoU'S in it in which to pack pieces of l)ricks an joint ^1 and B; space must be allowed in the centre for the lifting and blocking gear, as shown at //. After loam- ing and sweeping this part of the core, the sweep must be changed for another to make the proper thickness. I have seen the same operations gone through with as were with the bottom, — (drying the core, putting on parting-blacking, loam, brick splinters, etc. ; but all this in the case of the core is quite unnecessary. Instead of loam for the thickness, use green sand dampened slightly with clay water. Press this on firmly with the hands, and sleek a little with the trowel after the sweep has properly shaped it ; no parting-blacking is re- quired. Before sleeking it, if parting-sand is sprinkled on, it will assist in getting a clean [)arting. You can now start without delay to Imild up against it to form the outside, after putting on the parting-ring. "When at the top, if 3'ou have no sweep, it is necessary to have a ring to form the llange : the job is then so far complete. Aftt-r marking in several places, great care is required to part this olT, TTTF.EE WATS OF MAKING AN AITl-VESRF.L. 259 as tlic mould is green and easily damaged. This part should be dressed and blackened first, as it must be dried ; when this is drying, the upper half of the core can be dressed and black- ened, then put the cope part back in place. The loamed top-plate A" is placed on the top for the purpose of lifting the core out of its bod. I have seen two bolts used for lifting the core, the bolts being screwed tight on the top of the plate. The position of these bolts is shown at the right of //. This plan was not satisfactory^, and far from being safe, as it is imj)ossible to screw the bolts so as to have equal strain on them : therefore the core is liable to move, when free from its bed, by the effort to come to an equilibrium. If it does move, there is a poor chance of adjusting it with two bolts. A better way would be to use three bolts, then it can readily be adjusted. By having a strong piece of iron alongside each bolt, extending from the core-plate and tightly wedged, the bolts could be tightened to suit, with confidence that the core will not move from its proper place. This is shown to the left of ^. If the top-plate is not strong enough, it would be a good plan to use a three-legged cross, as represented at L. This, by bear- ing on the points of the legs as well as at the centre, would strengthen the plate. Two waj's of making joints are shown at A and B. Some make a bevelled joint, as at 13, the bevelled part serving as a guide in lowering. This is generally satisfactory when there is a good foundation. There is at 5 a chance of getting a iX)or flat joint from the prickers not lifting the loam ; also, when closing, there is danger of crushing the bevel part, if not closed entirely fair, which will spoil or disfigure the casting. The level joint at A is far better. This is made with two plates, which makes the joint iron and iron. It can be guided together by outside marks. A better way of guiding would be to have pins, as shown at X. To make these plates, have a bed with ■2r»0 TIIUKI', WAYS OF MAKIXr; AN AIK VI.SSKl,. llic size siiid form inaikcfl on it, and Iiiiili ciioiij^di to oast two jilatrs. Urfoii! (.-astiiiLj tlic liisl |ilalc, si't tin- ^iiidopiiis so the l)lalo will liavc a i^ood hold of tliiiii : the mipcr portions of these pins shonld i)e (jih-d, and a good coat of parting-sand put on them. After the lirst plate is east, put on ii go prevent the phites uniting. There can be three or lour lugs cast on the upi>er plate, as shown at -I, for the purpose of wedging ciiapU-ts. In this plan, tlu ii! are slunvn three ways of ruiniing the cast- ing, as at *!)', Y, and 11'. The runner at S is almost sure to cut and scab the core and mould. The runner at F is not so l)ad, but is open in a less degree to the same objection. I can w ith confidence recommend the runner W. Looking at the "second plan" in tlio engraving, the core and the mould are made separate. The bottom of the core is formed with a sweep iV. "When this is dry and turned ovi-r, it is laid on a bed prepared for it; care being taken to have the plate level, and placed centrally with the sweep. To insure its proper location, a nick may be made in the sweep that forms the bottom, to correspond with the top sweep, as shown at 2, 2, 2. For supj^orting the mould, a plate shonld be built in its npper portion to bolt to the bottom plate, as shown at 4. In the "third plan," the mould is made in three parts; the bottom when finished is divided into two sections, one of which is shown at R. The four lugs are to clamp the sections together b}'. Tlie top part can be made I)}' having a plate with prickci-s on it, OS shown at M. For closing by, the sweep siiould be made to make an outside mark to correspond with the under part, as shown at XX. Although the third plan seems to be the easiest and simplest, it is seldom adopted, for the reason that the bottom being the weakest part, or the part most likely to give way from over- pressure, it is essential to [)iovide for its being sound and solid ; A METHOD OF MOULDING GEAR WHEELS. 261 and the only way to do tins is by casting the bottom down as shown ia the first and second plans. A METHOD OF MOULDING GEAR WHEELS. Br "William H. Harrison, Braintree, Mass. As a sort of supplement to the most excellent series of articles which JNIr. West has been writing on the subjects of Moulding and Casting, I venture to present the following method of moulding heavy gear wheels, which I believe was original with myself, and which I have found exceedingly use- ful in a great many instances. It is really a rough substitute for a moulding-machine, and like a moulding-machine possesses the merit of making wheels which are tolerable approximations to truth. The method of making wheels by using short cores on which the teeth are moulded, and spacing them around in a pit, is one not to be tolerated ; for, although a thing may be made tolerably satisfactory to the moulder, the application of the machinist's calipers will show that the teeth and spaces var^' in thickness from the difficulty in setting the cores, while the cores themselves change their shape from the shape of the core-box in handling and drying. There are mill-owners who imagine they have accomplished a good work when they insist upon having the gears turned, thus truing the points of the teeth ; forgetting that the points of the teeth, even in the most perfect work, are not intended to touch any thing. It is, however, a somewhat melancholy sight to the man who bears the expense, to observe one of the old-fashioned boring-mills, or lathes of light weight, nibbling off a little cut, and the machine jumping from tooth to tooth, as though trying to make time between the cuts. 2(;2 A MKTIIOI) OK Mon.DINf; (UlAU WllKKI^S. Fi^. '.17 n proscnts :i section llir()ii<;h the s.-iiid of tlic foiimlry floor. A A is :i vi'iticul sj)iiulU', tapi'iid :it tlit- lower end, ami fitted t(j a tapered hole in the 1)a.se i)late. C is a eastiiiji, having l)()ri-d holi-s earefiilly fitted, so as to slide frei-ly upon tbo spindle. A boanl is bolted to C, whieli levels the lluor ou Fig. 97. the line a ?>, and leaves the mound c, if required, to form the boss. The cope is then placed, and rammed up as usual with a piece of tubing or gas-pipe slipped over the spindle to allow the cope to be lifted without disturbing the saml. The cope bein<' lifted and swung to one side, another board is used, A METHOD OF MAKING GEAR WHEELS. 2G3 which sweeps a pit in the green sand of the floor to the shape dcfg. The part / is for the boss at the lower side, and gr is the core print. The casting C is now hfted from tlie spindle, and the index plate D D placed and secnred b}' the set screw. This index plate is smoothly turned, and while in the lathe a number of circles are struck with a fine-pointed tool. These circles should be graduated, and the holes drilled on a gear- cutter, or, as the English say, a " dividing-engine ;" but in my case the dividing engine was a sharp-eyed apprentice, armed with a pair of compasses, a hammer, and a centre punch, in preference to the pattern-maker with his glasses and lead-pencil. The board F, having the pattern G attached, is now bolted to the casting C, and slipped down upon the spindle, and the point / adjusted so as to drop into the centre punch marks t, I, I, etc., and allow the lower end of the pattern to come down upon the bottom of the pit on the level c. The green sand forming the space between two teeth is then rammed, and the boai'd //, Fig. 98, laid on with a ten-pound weight on top of the teeth to hold the sand down, when the pattern is being drawn, after which the arm is shifted to the next hole in the index plate. It is well to give this pattern some draught ; not to make it lift easier, but because the straining of the lower part of the casting, particularly when the face of the gear is wide, tends to make that part larger. It is also well not to allow too much for contraction ; in case of these heav}' wheels, ■^" per foot I have found ample. After the teeth are formed, the spindle and attachments may be removed, of course leaving the plate B in the sand until after the casting is made. The arm cores and centre core may be placed in the ordinary manner, being made of dry sand ; or in some cases where the gear is large, and the arms plain, the core box may be laid in the mould, and rammed up with green sand, in the exact location where it is required to be. The cope may now be put in place, and weighted as usual in 2r,4 A Mi:riini) ok makinc; f;i;Aii wiikkls. work of tliis clianu-ttT. It will !»(» observed that in Fig. 98 the teeth are siiowii of tlie iiivoiiite f(»nii, which 1 adopted some years ago as liie best form for roiigii wheels. They certainly Fig. 98. are the strongest as to form, theorclioall}' ; and for smooth running, some of these wheels made with this rough apparatus as coarse as 1\" pitch on the pitch circle, I have never seen equalled by any gears moulded from a pattern. CUPOLAS AND MELTING IRON. SMALL CUPOLAS. "When trade is brisk, nearly all machinery shops cast every day ; when dull, man}' are more likely to cast once a week. Whether trade is dull or brisk, castings are wanted in a hurry ; often, the duller the trade, the greater the hurry. Some want them even before they are ordered : they think a casting should be had the same as a piece of forging or carpenter- work. Waiting for a small casting in dull times, is often caused through waiting for a decreased force to get up enough work to pay for running off a heat. The expense of running off light heats in some shoi)s is very heavy, the cost being regulated by the size of cupola : the smaller the cupola, the less the ex- pense. Small cupolas are not only good for running light heats, but are valuable for testing our modern brands of pig-iron. Pig- iron is something of a mystery, and to find its qualities it gen- erally requn-es to be worked. To melt a sample of pig-iron in a large cupola, is not always practicable, from the fact that castings are made of mixtures ; and, even would circumstances allow the first charge to be all of one brand of pig, there is little assurance of its being entirely free of upper mixtures. With a small cupola, and thirt}' to fifty cents' worth of fuel, three or four hundred-weight of pig can be melted, loith an assurance of the casting being all the jn'oduct of that special x)ig- Small cupolas are often as useful in large shops as in small ones. In the whole country, there might be found a dozen large shops having small cupolas, and out of the dozen there might 205 2f>t> SMAI.I- (TPOLAS. 1)0 four that linvc Ijocn useil over a dozen limes. It is ver}' easy to Imild small eiqiolas, but sonielhing of a job to success- fully run them. Notwithstaiidiug, the princiijle of melting is the sauie in a small cupola as in a huge one. For mauii)ulation in handling, there is not the room in small cupolas that there is in large ones; and on this account the small cupola has not been very successfully used. There are two styles of small cupolas in use. The first is upon the same plan as the common round, straight cupola ; and the second is made so as to be turned upside down, for con- venience in cleansing and dumping. Knowing the disadvan- tages attending the successful running of small cupolas, ranging from 12" to IS", I have designed, as shown, an original plan that I think will fully meet the requirements. The cupola here shown will occupy a space about four feet square. The working portion is hung by two cast-iron trun- nions, having a wrought-irou ly pin cast in each. The trun- nions work in a sliding rest, one of which, a face view, is seen at jB -4, in back view of cupola (Fig. 99). An end view of the slides is shown in the side view. The plan of the slides is seen in small cut at the top. Shown in back and side views, inider the sliding rests, are friction wheels. These slides are held in place by the standards SaS, shown bolted to the columns. By a slight push, the workmg-portion of cupola can be brought out from below the upi^er portion or stack. A pin inserted tlirough P to K prevents any further sliding of the rests. After this the steadying bars ////, shown in the plan as well as in back view, are removed. The cupola can now be turned to a horizontal position. To prevent the slides from running out of their roller bearings when mo^ing the cupola, the slides A B should have a projection on each end, as seen below the end at K. As the working-portion of the cupola is only four feet long, by the means of the drop bottom, a man can reach and see all parts of the inside, there- I-L- IT-^. Fig. 99. SMALL CUPOLAS. 267 by giving him a good opportunity to thoronghly pick out and cleanly daub it up. This is almost an impossibility iu the in- stance of mauy small cupolas. AYith this jjcirt of the zvorJCy rightly and handily 2'>erfonned, lies the main secret of successful melling in small cupolas. In picking out and daubing up small cupolas, care and clean- liness micst be exercised. The lining should be kept as smooth and even as possible : any roughness has a tendency to make the charges hang up. It is an easy matter for iron to become wedged in such small cupolas. The daubing would better stand a long heat if it were dried, all cracks filled, and then given a coat of good blacking, thereby making it as smooth and clean as the linings of ladles ; but for ordinary heats this extra work is not necessary. Not oul^- is it not essential that the cupola should be clean, but the iron and fuel should be clean as well. Dirt creates slag, and slag could soon bung up any cupola. The slag-hole, if properly managed, greatly mitigates the disastrous effect of slag. Dirt in any form is detrimental to successful melting. With large cupolas one may be somewhat careless and unclean, but with small ones attention to these points must be given. The thickness of lining for small cupola can range from 1|-" up to 4^". The 1^" lining is obtained by daubing the shell with three-fourths of good fire-clay mixed with one-fourth of sharp sand. To mix them well, they should be boiled together in a kettle. Common clay could be used, but in the end the fire- clay would be cheapest. A 2^" or 4|" lining is made with fire- brick. For small cupolas, intended for frequent use, the 2^" thickness of lining is about as thin as should be put in. For daily use the 4^" lining would be preferable, as this thickness would last longer than a thinner one. To use a 4|^" linnig to make the 12''' cupola, the shell of cupola would, of course, require to be larger than shown. The working-portion of cupola shown has, in the length of 208 fiMAI.I- riTOLAS. four feet, a tapor of 2". This is a pciiiit I am nware is not iiiuch practised, wiiidi is aiiotlK-r reason fur ill workings. t^oKill cit])ukis are heifer for liaviiuj a tajjer, as it assists in jtre- ventiiKj the stock from becoming wedrjod or hnng up. In constructing the shell for small cnpolas, there arc several ways in which it may l)c clone. One is to make it out of all l)()ik'r-iron ; and another, by placing cast-iron rings on top of each other, tying them together witli l)olts. A third plan is to l)ind vertically placed cast-iron slabs or staves with wronght- iron rings; and a fourth, to make a square shell by bolting together cast-iron plates. The fifth, a '^ crank's " plan, is to line up a flour-barrel. In the cupola shown, the bottom is made of cast-iron ^" thick. From the tuyeres up, boiler-iron is used. The slides A B arc of wrought-iron, and the platform plate of cast-iron. The plan of tuyeres shown is one that ■u'ill evenly distnbute the blast. At 1, 2, 3, and 4, are peep barring-holes, which may be plugged, as shown, with wooden stoppers, or they can be closed with swinging slides. Numbere 6 and 7 are nozzles to attach leather, rubber, or sheet^iron blast-pipe to. The pipes must be made adjustable, so as to allow the cupola to lie removed. The tuyere boxes, seen at 72 72 in back view of cupola, are independent of the outer shell, and are set in when lining up. These tuyeres, to work well m the three sizes men- tioned, should have an area of from twenty to twcntj'-five per cent of that contained in the cupola. P'or the cupola shown, use four tuyeres l^"x4". The milder, with proper volume, the blast can be admitted into small cupolas, the longer can they be made to run ; and this is especially so where all coke is the fuel used. The construction of the tuyeres as shown is. of coui-se, more expensive than were nothing but two round tuyeres used. Some small cupolas have the blast thus directly admitted. It is a cheap and ready plan ; but I think tlie plan given is the SMATJ, CUPOLAS. 269 best, as admitting the air as shown breaks its direct force, and admits it in a much more even and a milder manner, so that it does not have such a bunging effect as it does when passing directly from the blast-pipe into the cupola. The blast press- ure for cupolas ranging from 12" up to 18", using all coke for fuel, should be from two up to four ounces ; with coal and coke, four to six ounces ; using all coal, from five to seven ounces. The stacks for small cupolas need not be continuous, as for large ones. After a foot or two above the top of clmrgmg door, they may be led into the stack of a larger cupola, or into a chimney. Between the cupola cuts, is shown the mauner of charging. In charging the working-portion of the cupola, it would be better to have it slide out to come iu under the platform hole X. This would give a good chance to properly and conven- iently charge. After charging, it can be pushed back, locked, and the portion of cupola above platform charged. The half- inch of space between the platform and underneath portion of the cupola could be stopped up with cla3% to prevent the blaze from coming out. The platfonn as seen is but a plain plate. As shown it would be too weak to carry much of a load, and also there is nothing to stop stock from rolling off. To meet both those requirements, it would be a good plan to have a nb say l^"x6" cast all around the plate; and where it crosses the hole X it could be given an arch shnpe, so as -to allow the cupola to turn over. Still more to strengthen it at X, there might be a complete ring cast on the plate, just large enough for the cupola to fill. If this were not thought sufficient, still another rib could be cast on the plate on its under side, below the place where the pig pile is seen ; and to add support, which might be needed should a very heavy stack be used, the plate could be cast thicker than shown, and brackets carried from the columns up to it. 270 SMAT.T, rrroi,AS. Tlu* licd's wcinflit of fuel, pivcii in cut. i'^ iiifondon to \>\i\co tlu' hod :ilK)iit lo" aliovt' tlic U>\> of tiivti'*'- W'itli trials made! of coal and coke in the shop, the " to licighl i»f bed. If all coke is used, have bed 1h" above tuyeies ; and for a heat of 1,000 pounds or over, add from 2" to 0" to height. L'sing all coke Ix'twecn charges, continue as shown. Should all coal be used, donljlc the weight of fuel and iron in charging, which would be 20 pounds of fuel instead of 10, and 250 pounds instead of 125 of iron. The fuel should be small size, and the pigs broken into four or five pieces, and scrap in like proportion. The charges for a 15" cupola could be made as follows : On a coal bed charge 350 pounds of iron, after which, with coke for fuel, have charges, 17 pounds of coke and 200 pounds of iron. With all coal, double the charges. The charges for 18" cupola : on coal bod, 500 pounds of iron ; coke between charges of 300 pounds of iron, 25 pounds. For coal charges, double those above. If, in any of the three sizes, coke is used in place of conl for the bed, then make the first charge of iron no heavier than those given for tlio upper charges. Should the iron come too dull for very light castings, add to height of bed from 2" to G", and between charges two, four, to six pounds of fuel. By using coal for the bed, the cupola will molt more iron than if coke is used, as the -coal will stiind the effects of the blast better than coke. By slagging the cupola, it can often be made to melt near as much again iron as where no attention is paid to slagging out. The capacity of a 12" cupola, when slagged out, is about 1,500 pounds; that SMALL CUrOLAS. 271 of a 15" cnp<^la, 2.000 ; nud au 18" cupola, 2,500 pounds. With exc'olk'ut mauagemout the above figures might be ex- ceeded. I would here state, that, although I have shown the cut of a 12" cupola, for daily practical working I would not recommend the use of one less than 15". The reason for placing the few pounds of coke below the coal shown in the bed is more for the purpose of assisting in kindling the coal, than for saving expense. Coal is harder to kindle than coke, and in small cupolas the ditliculty is greater than in larger ones. The construction shown is in principle applicable to any of the three sizes mentioned. For a 15" cupola, the tuyeres should be increased from 4"x 1-J", to 3^"x3|", and for an 18" cupola 4"x4". Also for a 15" or 18" cupola, the slide bars and plat- ^ form should be stronger than shown for the 12" cupola. The tuyeres could be 4" lower, were all coal used ; but for coke the height given is required. A cheaper cupola could be con- structed, but for cheapness in the end I think the one here represented would be satisfactory. An idea which it might be well to express for one who was willing to forego the convenience allowed by having the cupola slide out under X to be charged up, is simply to dispense with this arrangement, and, in order to turn the cupola over and back, to let a part of the body — which is here shown to be above the platform — project below sufficiently to be cut so as to form a slanting joint, instead of being parallel as now shown. 2t2 SMAl.I, (II'OLAS. If this were dono, llio Iwo ]):irfK would fonn n joint snTnolIiinj; similar to tlial sciii at K, Fijj;. lOO. In tlitis all<)\viii;f tlii' upper liody lo I'loji'd tliroii<^h tlu' iilatfoim A. it woiiM UMiiiiic to lie lu'ld up by im-ans of hrackt'ts A', auti liy llii.s plan tlio hole X would not l>c rc(juired. "While ui)on this subject, it might be well to suggest an idea with reference to running large cupolas for constant light heats. Ill many cases, were the cupola lined up so as to make it smaller, much expense in fuel would be saved. For example, a 48" cupola could, at a small outlay, be lined up to 3U" ; thea when business warranted it, the false lining could be taken out, and most of the fire-brick saved for periodical American busi- ness depressious. COKE AND COAL IN MELTING IRON. 273 COKE AND COAL IN MELTING IRON. There having been recently many encomiums upon the merits of coke for raeltmg iron, and none for coal, it seems to me that some, through short acquaintance with coke, are a little too enthusiastic to show ui) one good fuel at the expense of another. I do not deny that coke is a good fuel to melt with : nevertheless, coal is also good, and in some ways superior, for which I would not like to see its use abandoned. I hope to here show wherein the merits of each fuel lie, and to present a few ideas that may assist those wishing to change from coal to coke. The merits claimed for coke are as follows : First, that it ivill melt faster than coal; second, that it requires less blast pres- sure; third, that it is a cheaper fuel than coal; and, fourth, that it contains less impurities^ and will malce softer castings. The first three are certainly true ; but regarding the fourth, I have doubts. Either through design, or lack of observation, there seem to be three important points in the use of coke and coal that have never been brought out. One is regarding the life and heat of the metal ; another, the length of heats ; the third, qualities required in melting heavy iron. The foundr3'men in my sec- tion of the country have had experience with coke for a long time ; and I have yet to hear any of them say that coke, on an average, is better than coal for making hot metal, for length of heats, or for soft castings. To run long heats, and have metal keep its life, is a very important factor with many foundries ; such, for instance, as those doing heavy work, where the first 274 COKK AND rOAI, I.N .Mi;i,TI.Nf; lltON. five or ten tons nifltcd liavc to stand in a ladli- from one to two hours, waitinii for more iron to lio ini'ltcd or anotlii-r ladle to lie lillcd. 'I'iii'rc is a nolalilc fcatiiri' — that ol" tlic life of liciniil iron — thai many shops ma}' not notice, as with them the metal may be said to he no sooner out of the cupola than it is pouretl into the moulds. I am a firm believer in meltini^ iron '* hot," as I know it to be a fact that stronger castings can be made by so doing. The length of heats has in my practice been increased by using coal with coke ; and in this section many foundries mix coal with coke, in order to do clean cupola work, and i)roduce hot iron. That a cupola will run longer with a mixture of Lehigh coal and coke, is admitted by many fouudrymeu to be a fact. In order to make my subject plain, and to show waj's of charging, the accompanying cuts (Figs. 101, 102) are inserted. The cupolas, as shown, are charged for ordinary heats. To run at their full capacity, about ten pounds more fuel should be added to each charge. To commence with, T will state that the dcscri))tion of the various modes of melting here given are not of test heats got up to show how fast melting can be done, or to present the two- sided question of economy in fuel. The heats described are the average practical workings of a few common, plain, round cupolas in Cleveland. The Cuyahoga, Viaduct, IColipse, and Globe Works have kindly allowed me to publish their ways of melting. The Cu3*ahoga and Globe Works make lieavy steam-engine and machinery castings ; the Eclipse does a large business iu house work and general jobbing castings ; while the ^'ia(luct Foundry makes a specialty of vapor oil stoves and light jobbing castings. These four specialties cover about all ordinary foundry eastings, so that nearly all cau apply one or the other to their own class of castings made. COKE AND COAL IN MELTING IRON. 275 The Cuyahoga and Globe Works each has two cupolas ; and, their smallest ones being of about the same size, I have chosen them to show their practice of using coke and coal. The Globe Works' cupola is charged with all coke ; the Cuyahoga, with coke and coal. The charges of iron, as shown, are con- tinued to the end of the cupola's capacity. The Globe Works' blast pressure is five ounces, obtained from a Sturtevant No. 8 fan. Time of melting, when using all coke, three and a half tons per hour. The Cuyahoga's blast pressure is seven ounces, obtained from a Root rotary-blower No. 5. Time of melting, with coal and coke, three tons per hour. The Eclipse Works' mode of charging, with all coke, for a heat of seven tons, is, 700 pounds of coke for the bed and 1,200 pounds of iron for the first charge, the balance of iron charges being all 800 pounds. Between the charges, 95 pounds of coke. The cupola is 35" inside diameter, having four round h" tuyeres, about 18" from the sand-bed to the centre. The height of charging doors, bottom to foundation plate, is nine feet ; blast pressure, seven ounces, obtained from a No. 7 Sturtevant fan. Time of melting, 6,500 pounds per hour. The Viaduct Foundry's mode of charging, with coal and coke, for a heat of six tons, is, 738 pounds of coke and 400 pounds of coal for the bed; first charge of iron, 1,800 pounds. The balance of iron charges, 1,200 pounds; fuel between them, 123 pounds of coke and 25 pounds of coal. The cupola is 38" inside diameter, and has four oblong tuj'cres of the dimensions shown at right of cupola (Fig. 102), their height from sand- bottom being about 16". The height of charging-door from plate is eleven feet ; blast pressure, ten ounces, No. 5 Sturtevant fan. Time of melting, 6,500 pounds per hour. As a general thing, in the charging of this cupola, there is not an}' fuel used be- tween the last charges. For a flux, the Cuyahoga Works use fluor spar. In using 27<> ("OKH AM) CdAI, IN Mi:i;ilNf; IIU)N. tliis flux, wo shovel al)()iit Iwi-lve iK)nii»ls on tlio top of each cliaiLTi'. with the cxcoptioii of the I'lrst two or tliive eharj^es. Ill iiirltiiiif witli cdUc, till' lire docs not icipiiro to he ntarted as early, simply heeanse coke does not n'(piiie a.s lon<; a time to kindle as eoal. The idi-a of time for kindling shonld l)e to allow sullleient to have the fuel all on fire before iron is char\l>out 7" bi-low the tuyeres a slag-hole was inserted. With lliLse changes the cupula would successfully melt twenty tons. 1~] stage "Floor Unv ObloiKj Tuyere 7 dia. 04 c J 'Flat Tuyere p.rcp nrfl Plan of Wind Box anil Tuyeres COKE AND COAL IN MELTING IRON. 277 In melting for machine or heavy castings, the iron is gener- ally allowed to accumulate before tai)piug-out. This accumula- tion causes the raising and lowering of fuel (that is, if tuyeres are higli enough to permit such action), thereby not leaving any inside body of fuel long at a time exposed to the cooling effects of the cold blast. The benefit of this cannot but be seen if connected with the reason for slackening the blast and barring a cupola, as noted in the following. A Lehigh-coal fire has more of a body than a coke fire. The blast, as it goes into a cupola, will more readily cool off coke than coal ; and the cooled bod}' of fuel, which more or less sticks to the front of tuyeres, if not attended to, gradually increases until it reaches nearly to the cupola's centre, which results in scaffolding or bunging up the cupola. To assist in preventing such results, the l)last should occasionally be slackened, the tuyere peep-holes opened, and then, with a bar, the cooled body of fuel, and frozen droppings of metal, shotdd be driven in towards the centre of the fire. This will greatly cause the cooled body to be burned up, the frozen droppings re-melted, and give a clean hot body of fuel for the cold blast to play upon. A point that has much to do with ill success in changing from coke to coal is using too strong a blast. As a general thing, about one-third less pressure should be used for coke than for coal. I know it is nice to see a cupola melt fast ; but not so enchanting to have to re-line it about every month, which will often result from too strong a blast. It is impossible to obtain good clean iron, or have a cupola run very long heats, where a cupola is being cut to pieces with the blast. The cupola on the right (Fig. 102) ran for about one year, almost daily, without being re-lined ; which will, I think, be acknowledged as a good showing. I do not credit all to the merits of a mild blast. There is another feature that undoubtedly had much to do with it : that is, the daubing-up of the cupola with fire-clay. 1^ I0« M«. •/ /roita^ I V -J J ^^ ■ Jfe*^ ; > '>^ -^ _j» .too lb,uft„.,l j^- Btaff« rioiur Lint Ot.luH.j luyrr* "-r din. r.rrv jvn* fif . 101. Fig. 102. 27ft POKE AM) COW. IN MF.I/riNf; lltDN. In l>otli of oiir fouiwlrv fn|i<)l:is, we constantly use coal and C'onnc'llsv ilk- i-oUc, as hlu)\vu, and. Iiy piopcr attcidion to slat^- giiig (which 1 am sorry to .say lias to be done thn»ned. and what was left of the block could be charged in another heat. Such heavy pieces arc l)est melted when one can arrange to have work that does not rccpiire the hottest of iron, lli-avy scrap when it is melted is superior to light for making strong COKE AND COAL IN MELTING IRON. 281 mixtures ; and, although it takes more fuel to melt it, it may ofteu pay iu the eud to do so. There is au adage that "it is a poor foundry that cannot make its own scrap." The way some lose heavy castings, one would think they were trjung to supply their neighbors. The loss of heavy castings makes heavy scrap ; and for some shops the above may suggest ideas to help them get it out of sight, and rid themselves of unpleasant memories. 282 INTKLLIGKNCK AND ECONOMY IN MKLTINO. INTELLIGENCE AND ECONO:\IY IN MELTING. TinntK has at no time been the scientific thought given the subject of melting that is given it at the present time. A few years back, there were more superstitious melters than intelli- gent ones. In fact, there can at the present day be found men who look upon the cupola as something more supernatural than mechanical. If any thing goes wrong, they give an in- quirer a look as much as to say, Question the gods. Dull iron one day, hot iron the next, and a bunged cupola the following day, may be excusable in some shops ; but, to the intelligent founder of to-day, such workings are connected with cause and effect, and a want of knowledge. The cupola can easily be the master if one docs not strive to master it. To master the cupola, is simply to have it do as one may wish. Hot iron one day, dull another, three or four dif- ferent grades got out witliout being mixed, heavy sci'ap run down, and fast or slow melting, are points that can be and are mastered by intelligence. It may be a broad assertion, but nevertheless the writer would say, that in no part of foundry practice is there a better chance to control results than in melt- ing. The chances arc far more in favor of a first-class moulder having bad results with his work than he would were he a first- class meltcr. Any mechanic well versed in l)()th l)raiK-hes, I think, will verify this statement. Having, just before completing this second volume, made a tour through many States, I was much i»leased to see with what interest many foundrymen — who, liy the way, were readers of " foundry literature " — luid taken up the subject of INTELLIGENCE AND ECONOMY IN MELTING. 283 melting. Kioht here I would like to say, that, although there are those who sueer at foiiiuby literature, a travel througli the countr}^ will prove that the most intelligent and progressive moulders are those who read it. It is not alwaj's the informa- tion we get from reading, that measures its value, but the thinking it often induces us to do. Among the most intelligent cupola managers, the question of economy in fuel is the all-important one ; so important with some, that it reminds one of the man who tried to teach his mule to live without eating. They keep striving until they find themselves sadly the losers. This question of economy in fuel is a misleading one. With intelligent management, and conditions alike, two distinct foundries may melt with a like low percentage of fuel, but what may be economy for one shop may be quite the reverse for another. One shop may have a class of work which will admit of being poured when the iron is in a less fluid condition than another. Then, again, work may be of the same class, but one has ar- rangements for taking care of the iron which will not admit of carrying it to the moulds as quickly as the other, such as the distance it may have to be carried by hand or power, etc. I have worked in shops, where, on account of their poor arrange- ment and that of their cranes, the hottest kind of iron would often be too dull to properly pour into the mould by the time the ladle reached it. Such a shop, if arranged so that the metal could be poured into the mould before it began to lose its life to any extent, might often with safety melt with much less fuel, from the simple fact that they would not require the iron to l)e in as fluid a state. There are many things to be considered with reference to what is true economy in melting ; and it is not right for one to insist that because some other shop may be melting one to eight, nine or ten, every other shop should do likewise. The 2^4 INTMI.I.Kir.Nf'K AND KPONriMY IN Ml-I.TINO size of cupola, luiLjlil of liiyt'ics, wciirlit of licals, nnd llip ques- tion of nimiiiiir tilt' luals imifoniily in \vcier cent, the iron comes down so dull that castings are lost, and ladles " bunged vp." It don't reipiirc the loss of mau}' castings to balance the cost of the few extra pounds of fuel it would have taken to make the iron fluid enough to fully insure the running of the castings lost because the metal was dull. Some shops can admit, in practice, melting done with a less per cent than others that do the same class of work, from the sinii)le fact that they have excellent facilities for handling the metal quickl}'. In cases where work is such as not to require very fluid metal, the low jjcrcentage that some may with suc- cess use certainly would not be advisable for others to practise. It may be thought tliat the iron is very hot ; l)ut if it had to be carried as far as it must be in some shops, and tiien poured into castings about as thick as paper, it would be found that there was a difference in " hot iron." Of course the writer has no intention to disparage economy in the use of fuel; but the only icay to rightly judge of true economy is to see the facilities of the shop, and class of tvork to be made. For my part, I would not question one to five as being extravagant until I knew all the conditions. As a matter of fact, when all the circumstances are consid- ered, iron is being economically nu-lted from one to five up to one to eleven. To melt lower than one to eight, is no doubt creditable, and a saving in the cost of melting ; that is, if by so doing the welfare of the cupola, ladles, and castings is not sacrificed; but were the facts known, more cupolas would be INTELLIGENCE AND ECONOMY IN MELTING, 285 found melting one to five than to seven, eight, or nine. I am well aware that melting one to eight, nine, or ten, sounds very economical when classed against one to five, six, or seven. To concisely give his experience and obseiTation on this point, the author would assert that in any cupola, running to its medium capacity, iron cannot be melted as hot or in as fluid a condition with fuel one to nine, ten, or eleven, as with one to five, six, or seven. Where intelligence is coupled with experience in melting, a good judge of fluid iron can easily detect the de- crease in the metal's fluidity caused by melting with less than one to eight. With the best possible management and condi- tions, I think almost all experts will agree with the author in saying that any less fuel than one to eight in medium-sized heats will show its results by giving a fluid iron with less life. Of course there are many cases where much hotter iron with one to eight can be obtained than others would give with one to five ; but what the author wishes understood by the foregoing is, that where one can " melt hot" with one to eight, he will notice a decrease in the metal's life and fluidity, should melting be done with less fuel. To properly charge and take care of a cupola, involves a knowledge many are not willing to concede. It is often sur- prising, how hot some melters can bring down their iron with comparatively less fuel than others use. The management of a cupola is every thing : some study to make it a science, while others act as if the cupola were only a hole iuto which the iron and fuel are to be thrown, and, if it does not come down right, lay the blame to a poor blast or cupola, etc. Some in melting do not even weigh their stock. In such case, there cannot be a uniformity in melting. If one wishes to master melting, he must at least weigh the fuel and iron, so as to have data from which to work. He can then regulate his heats, and have a 'uniformity that it is im'possihle to obtain by guess-work. When cupolas arc charged at random, one may see the first of 28G INTKLLIUKNCK AND KCONOMY IN MKI.TINO. the heat l)riii{]; down hot iron ; the middle, dull iron ; and the end, atjiiin, hot iron. Tlu're may be a half-dozen ehanfj^i-s in the lluidity of the iron, every charge seeming to make an altera- tion in this respect. Uniformity in melting requires the em- ployment of iiUcUifjence and system, "With this, one can have as hot or as dull iron as he may desire. With system, we know how high to have a bed, pressure of blast, and the per- centage of fuel to use, etc., to assist the obtaining whatever fluidity of iron we require ; and the cupola is as easily regulated as a clock. CONSTRUCTION OF CUPOLAS. 287 ODDITY AND SCIENCE IN THE CONSTRUC- TION OF CUPOLAS. EcoNO^rY in the use of fuel, and fast melting, are points sought for in coustructing cupolas. To this end, many odd features have been introduced. The noticeable oddity of some cupolas is in their outline, while with others it is all in the tuj-eres ; then, again, we see the two combined. / have often thought that oddity was devised to bewilder and blind, more than to attain ivijyroveynents in practical residts. At least, the attain- ment of oddity is sometimes the oul}' success. By oddity in cupola construction, is meant a departure or break from the plain round cupola having one row of either round or flat tuyeres. The different oddities, if shown, might fill a fair-sized book. Out of them all, but very few have any advantage over the common tuyere straight cupola. .Europe, no doubt, is far ahead of our country in the origination of new designs for cupolas ; but whether she has accomplished any thing more than America in true economy and sjjeed, is a question. Should any foreigners wish to compare notes with us, I would be pleased to have them mainly confine their tests to the two following points : first, the fluidity of the iron ; second, the greatest amount of iron the cupola can cleanly and successfully melt. These two points are generallj^ ignored in all newspaper accounts of cupola working. What is one to know of any benefit accruing by the footings showing one to eight or ten, if he is not in- formed of the fluidit}^ of the iron melted? Any cu2iola can be made to melt one to eight or ten ; btit ivhether the Dietul is only good for pouring or running solid blocks, or can run thin stove- 288 coNSTiii'tTioN OF crroLAs. jjlate aistinga^ in the point loc should Icnou: of to jmhjp as to the merits of the economy in fuel. Kt'ganliiig tlu; Icnixtli <>f time a cupola may lie iiin without buni;in<:;-ui), is auotlKT point of iinpoitancc. If one can daily melt ten tons in a 30" cniK)la. wlii-re others can hardly do it iu a 40" cupola, there surely is some advantage gained. The running of cupolas is somewhat like foot-racing. Sovie can do excellent icork in a short run, but cjice them a long one, and they soon become '■'' pUiyed out." With referciico to where there is a failure in the length of time a cupola will melt satisfactorily, 1 will venture the asser- tion that the fault is more often due to vii^mcuiogment, than to the design of the cupola. The great fault with cupolas is that of having so nuich heat escape up and out of the stack. Could the heat from two cupolas thus lost he conceutrated into a third cupola, a person would not be far off in saying iron could l)e melted. Some, to derive benefit from this escaping flame, make their charging- door as high as they practically can. Others try to construct lln' cupola so as not to generate this flame. To this cud, some cupolas are nuide with two rows of tuyeres. The princi[)le involved is simply the admitting of an upper volume of air or oxygen to unite with the carbon gas liberated from the fuel by the bottom tuyeres. The flame one sees at common cupo- las' charging-door or stack is gieatly caused by the escai)ing gas meeting with the oxygen of the air. If, instead of allow- ing this gas to reach the charging-door to receive oxygen, we admit oxygen about at the height al)Ove the first row of tuyeres, where the melting-point commences, we there generate the flame, or burn some of the gases that otherwise pass up the stack. This point is further treated upon p. .']()."). If we can confine the heat thus produced to the melting-point, instead of letting it i)ass u\) the stack, there should be some bent'fit derived. The amount of air admitted through the u[ipi'r tuyeres, to CONSTRUCTION OF CUPOLAS. 289 combine with the gas produced by the air passed througli the lower tiij'ercs, should only be sudicient to consume the gas generated. If more than this is admitted, the solid carbon, or, commonly speaking, the fuel, will be attached, and converted into gas, which will escape, thereby causing imperfect combus- tion. 7/*, by two rows of tuyeres, more gas is made than there is oxygen famished to consume, the fact can readily be knoion by the amount of Jiame seen at charging -door. The greater the distance in a cupola between the bottom and the charging- door, and the fuller it is charged, the less will be the flame seen. In order to conduct some experiments upon this subject of two rows of tuj^eres, I had in our cupola (shown in chapter upon " Melting with Coke and Coal," p. 273) four 2^" tuyeres placed about 1-i" above the top of the lower tuyere. In making these tuyeres, we simply cut four round holes in the top of the wind- box over the peep-holes ; then, after four round holes were cut in the cupola's shell, 2^" gas-pipe was used to make the con- nection ; and to make the turn, there was a T used having three openings. One opening was used as a peep-hole, which was closed by a plug screwed into it. In the centre of this screw- plug, there was a hole bored, ^|", through which was worked a ^" rod having a cast-iron conical round plug on its end suffi- ciently large to jiist admit of its sliding easily, and thus regu- lating the blast. My experience with these tuyeres was an observed imjyrovement in the speed of melting ; and by them at least as hot iron ivas obtaiyxed. Another advantage which might be well to notice is, that in running long heats the upper tuyeres are of much assistance in prolonging the life of a heat. Should the lower tu^-eres become to any serious degree bunged up, the top tu3-eres will admit much of their blast, thus letting air into the cupola which other- wise would be excluded. After running a month or so with the two rows of tuyeres, 2!tO rONSTIMTTION OF Cri'OLAS. I li:i(l two of the iijipi-r tiiycns raised up so us to I>o 20" altove the Ixjttoiii tuyeres. This, then, jjave us what iiii^ht lie teniieil three partial rows of tuyeres. Tlie hij^hest row I hacoKEcoV ,1 f Fig. 110. -Cheney Cupola. CONSTRUCTION OF CUPOLAS. 297 Bed, Cuiinellsville coke CIIAKCVS . . 1,200 lbs. iron, , 0,000 lbs, . . 572 " u 6,000 " . . 572 " u 6,000 « . . 704 " (; 6,000 " . . GGO " " 6,000 " . . 572 " " 6,000 " TOTALS. Iron melted 36,000 lbs. Fuel consumed 4,280 " Ratio of fuel to iron 1 to 8.41 Length of heat, 1 hour and 35 minutes. The iron is described as being very hot, and the cupola as giving entire satisfaction in both economy and speed. One- third scrap to two-thirds pig iron was melted. The blower is Root's No. G ; revolutions, one hundred and twenty-five per minute ; length of blast pipe, twenty-five feet. Fluor spar and limestone used for flux. The next cupola to be noticed is what is called " The Cheney Cupola." In its design, there are many admirable features ■which commend themselves to the practical man. The follow- ing is Mr. Cheney's description of his cupola, as published by the " Boston Journal of Commerce." " The cut illustrates the manner of constructing an economi- cal cupola of medium size, to melt four tons of iron per hour. It is thirty-four inches inside diameter, and will melt six or seven tons of iron without slagging. By opening the slag-hole after four tons of iron have been drawn, the melt may be con- tinued to twenty tons. To make the slag fluid, so as to run off freely, use thirty pounds of limestone to one ton of iron. "In charging this cupola with coke, put 600 pounds on the bed ; on that, 2,000 pounds of iron. In the subsequent charges, use 130 pounds of coke and 1,400 pounds of iron. In a melt of 208 roNs-j iMcrioN oi (tpolas. li'^lil tons, this cnpol:! will melt U'U jiounds of iron to ono poiiTnl of cokr. (II r\<^h\ poiuitls ol' iioii to (Hie jxdiiul <»f coal. If coal is iixmI |oi- fuel, till' sand-lu'd should l)t' iiKuk- alioiii tliicf or four iiiclns dci'iicr than when coke is used. '•'IMiis cupola is designed for ordinary foundry-work wluTO Kliar|) iron is wanted. For heavy foundry-work, such as east- ings, recpiiring several tons of iron in one piece, the l)ed may be made deeper l)y placing the tuyeres from four to six inehe-s higher in the large-size cui)oIas. '•Put on Mast as soon as the cupola is charged, and give this cu})ola al)out six ounces pressure of Idast for coke, and nine ounces pressure for coal. '• When the lining burns away, nnd the dimensions of the cupola are enlarged so that six hundred pounds coke fail to make the bed sixteen inches above the upper tuyeres, the coke in the bed must be increased ; also increase the iron on first charge in same proportion as the coke is increased. "Fig. 110 show^s a cupola shell 48" in diameter, continued same size to the full height; lined with 2" common brick flat- ways to top of charging-door, and inside these 4V' fire-ltrick. The common brick always remain, so that when the fire-brick gets thin the shell is protected. " A is the sand bottom. " li is the iron runner. " I is the slag runner. Outlet for slag is a 2" hole opposite the iron runner, ^" lower than the bottom of the lower tuyeres. " C, lower tuyeres. They decline inward ^" in G", and are 16" above bed-plate, 8^" wide at face of lining, and S^" vertical, made with flange on the upjier side to bolt to the shell. The opening through the shell to admit the blast into these tuyeres is 3" by 5". Each opening admits fifteen square inches. " D, upper tuyeres. They are 2" in diameter, decline inward 2.1" in C", and are 20" a])ove bed-plate at the inside of the lining. 'I'lie hole through the shell to admit Itlast is 1]'' in CONSTRUCTION OF CUPOLAS. 209 (linmotcr. Six of these tuyeres are made with a flange on the top side so as to bolt the tuyere to the inside of the shell. These tuyeres each receive 1^ square inch blasts through the shell. " F is an 8" blast pipe to connect the wind-chamber with the main pipe, which should not be less tiian 12" in diameter. If the blower is more then fifty feet from the cupola, the main pipe should be 14" diameter. The wind-chambers have open- ings opposite each tuyere, with peep-holes in the shutters. Fig. 111. " E, wind-chamber, is 10" deep and 22" vertical, made in two sections, each section to supply wind to three tuyeres. " H, charging-door, is twelve feet above bed-plaie. "Fig. Ill shows a sectional view through the tuyeres. They are the same width as the lining between them, and supply an equal force of blast to all the fuel. The tuyere must be twice as large as the opening which admits the blast, and must occupy one-half the space around the inner circumference of ,")0<) CONSTRUCTION Ol" CCl'OLAS. the ciiiKila. Ill laiircr oiijiola.^ incrcaso the nnnilior of tlir tnycivs. " Fij;. Ill' sliows a perspective vii'W of the wiiid-cliamlier, matle in stini-eirchs, so that when it is holted to the siu-ll it will exteud around it, forniinj^ one chamber to supply wind to all the Fig. 112. tuyeres, and dropping down m" in front of the lower tuyeres. If through carelessness the iron overflows the tuyeres into the wind-chamber, it is more easily removed tiian if the cham]»er is even on the bottom, aud the chamber is not iu the way of the slag-hole." COMMENTS ON CUPOLAS. 301 COMMENTS ON CUPOLAS. In closing the preceding chapter, it may be said there are tiiose who, no doubt, would prefer the author's giving a few comments upon the merits of the respective cupolas therein mentioned. Taking them in the order shown, the first we come to is that at the Callahan and Dearmon's works. The style of tuyere used is one that ought to work well in melting with coke alone in straight cupolas which range from 48" down. Below 40" the tuyere area should be decreased in proportion to the de- crease in diameter of the cupola. From 40" to 48" it would be best to use the tuyere area shown ; for, to enlarge them would deter the blast more or less from being forced through the fuel to the centre of the cupola, and thus fail to create that rapid combustion which should exist there as well as at the outer circle. In melting with coke, it is well that laige tuyere areas be used, as they serve to prevent tuyeres from bunging up in ruu- niug off heavy heats. With coal, it is not as essential to have a large tuyere area, for the reason its life is not so readily chilled and ''blown out" by blast pressure as in the case of coke ; and also it is often beneficial to have the tuyere area made smaller for coal, so as to induce the pressure more in among the centre body of fuel. We require more j)ressnre. or density of blast, in melting with coal, for the simple reason, from its compactness it forms a most dense fuel. It must be understood that this extra pressure is not to be created by a contraction of the tuyeres : whenever '.)()2 coMMKNTr; ON crroLA.'^. prorjsuro must lie increased, il must Ito done l)y incrrnainrj the jxnfvr upon the hlmcrr ; and tlie iiieiease llieie — is indeed employing a " necessar}' evil : "' for, since the blast is cold, it must be raised in temi)erature Itefore it ean enter into combustion ; and when more air than ean l>e utilized is forced into a cupola, extra he.at is absorbed, and therefore it must increase the retardins: of melting, for its influence is to reduce temperature. From this fact can be deduced, there is such a thing as delivering too much air into a cupola, a thing which many think eaiuiot l)e done. Combustion cannot proceed beyond a certain rate ; and an excessive sui)[)ly of air oidy causes a waste of hvixt. and an unealU'd-for destruction of the cujtola's lining. An insulllcient supply causes imperfect com- bustion. The first combination of car])on with oxygen produces carl>onic acid, and this, in passing up through the fuel, fre- quently takes up uiore carbon, and is converted into carbonic oxide ; which, if allowed to pass away in this state, causes con- sideralile loss of heat, as carbonic oxide is a combustible gas, and can be burnt by furnishing it with a supply of air; which, if it only gets as it reaches the charging-door, is then of course too late to be of any service. By having a requisite volume of air properly delivered, this carbonic oxide is much decreased, and therefore more carbonic acid, which is the prod- uct of perfect coml>ustion, created. To know when we have the requisite amount f)f air in our cupolas, can in every-day practice be l)ut approximately told. However, tiiere is much that might be done to help us in intelligently handling the supply of the recpiisite quantities of blast. When we think how few foundries Iheri' are that have anv idi-a of their blast BLAST AND COMBUSTION. 807 pressure, or the volume of air used, aud secondly how few arrange so as to assist the pressure in deliverino; its volume properly into the cupola, we are not surprised at their excessive cost iu runuiug. A great many cupolas are constructed so that the tuyeres cannot be examined to see if they are working freely. The first thing to know is the density of the blast which the blower is creating, and this is easily shown by attaching a blast- gauge to the pipe. The second thing is to know iT the right volume of blast is being delivered into the cupola. Tliis can be told to a certain extent by a practical man, by noting the cupola's action ; and he may judge very closely as to its work- ing. In fact, it is almost the only way of telling whether or not a cupola is receiving its proper amount of blast ; for the pressure of blast in a cupola cannot be known from that gen- erated in the blast-pipes. The pressure in a cupola is generally much less than that in the blast-pipes ; and the amount of difference will depend upon how close the iron is charged, and how high and full the cupola is, and also on tuyere area for the passage of air into the cupola. Towards the end of a heat, higher pressure will generally exist in the blast-pipes than in the beginning. This is caused by the tendency of tuyeres to become "bunged up," and the accumulation of slag and dirt in the bed. Chilled iron mixed with fuel and slag is not always the only cause of " choking-up " of tuj^eres ; often clean fuel will lodge so close in front of tuyeres as to choke them up considerably. This is one of the reasons why large tu3'eres are often recommended, for with them the fuel has less chance of preventing the free delivery of the blast ; for we must not lose sight of the fact that fuel packed up close to the mouth of a tuyere acts to a degree like a damper. A good thing in practice is, just before the blast is put on, to bar the tu\eres so that whatever pieces of fuel may be "choking" them up, will be pushed back, and thus give the blast a good chance to enter ; and then b}- watching the tuyeres, and keeping them 30« lil.ASr AM) COMmSTION. open (liirini; llic course of tlio heat, tlic rof(iiisi(f' voliunos of !iir cMii lie iiiKiT if:iililv :iiiii.i. Cubic fl. at til'"' F. 141 liillit. Coke, (IcsiccaletJ, . Coal, average . . . 2.r,i L'.4(i 1(1.7 14.l:;a That tlie pressure or density of the l)l:ist as mi'Msiired in tlie pipes is more or less rejiiihited \>\ the tuyere opeuiuifs. :iud closcneas and ivcifjht of liie charged iron, is heyond dispute. The conchision to bo drawn from the above would point to the advisability of charging iron closer for coal than for coke ; as, the closer tljp iron is charged, the longer should it take the air and gases to travel upwards, thus affording a better chance for the increasing of pressure or density of blast in among the coal, without being obliged to raise the temperature of the unused volume of eacapiiig air referred to in fore part of this chapter. The cupola may be said to be only the end of a blow- er's blast-pipe, and the charges of fuel and iron but a damper, which could be packed so close as to almost shut off the cscajw of the gases or blast. The more completely any blast-pipe's outlet is closed by means of a damper, the greater pressure there will be in the pipe, and the more power will be required to run the fan or blower. It is not intended by the foregoing t(» say iron and fuel should be charged so close as to form a damper, so that the gases generated liy the combustion of the fuel should not freely escape ; but simply to show how we can regulate pressure within limits. BLAST AND COMBUSTION. 309 The pressure of the l)l:ist used on cn[)()las ranges from three up to eighteen ounces. For coal, from one-fourth to one-third more pressui-e is required than for coke ; and for eltlier fuel, the larger the diameter of the cupola, the more pressure is re- quired. A good showing of the average pressure used upon different diameters of cupolas is seen in the following table, which is compiled from Sturtevant's experiments. As speed in melting is chiefly augmented by the blast, the cuix)la should be supplied with all the volume it is possible to use jyrojitahli/ ; for, the more rapid the melting, the better and hotter will l^e the metal produced. Diameter in Melting Capacity Cubic Feet rUESSUKE Inches inside PEPv IIOUU IN OF Air IN Ounces op OF Cupola. Pounds. PER minute. Blast. 22 1200 324 5 20 1000 507 30 2SS0 708 7 35 4130 1102 8 40 0178 1040 10 46 8900 2375 12 53 12500 o-j53 14 GO 10500 4410 14 72 23800 0304 10 &4 33300 8880 10 "The number of cubic feet of air per minute given against each size cupola is the result of numerous tests taken on cupolas. "The melting capacity per hour in pounds of iron is made up from an average of tests on a few of the best cupolas found, and is reliable in cases where the cupolas are well con- structed, and driven with the greatest force of blast given iu the table." — Stl'ktkvant. ;310 yLACJUiNc; cjlt cltulas. SLArxGING OUT CUPOLAS. As sl:\os- sible, to "keep a head" of iron in the cupola until a sullicient amount of slag has accumulated, and then to make a special tap to let it out. Having made a good-sized hole, then by means of the regular pressure of blast let the cupola blow out until all the accumulated slag is disposed of, and then stop up, repeating the operation as a suflicient body of slag accumulates. As the end of the heat approaches, the slag taps require to be made oftener. It may accumulate toward the end of the heat so that at every tap more or less slag must be let out. As a general thing, lK)wever, if onlinarily dean fuel and iron are used, "slagging out" is not conimeuced until from one-third to one-half of "the heat is down." This refers to what are termed " lu'avy heats;" for as a general thing, unless burnt iron or b:id fuel is used, " light heals " seldom n'(]iiiie any "•' slagging out." SLAGGING OUT CUrOLAS. 811 By the above expression "keeping a head," is meant to sinipl}' not permit all the iron to run out of the cupola before stopping up. Slag floats upon the top of iron : therefore, by keeping a head of liquid iron in the cupola, the slag cannot run out of the tapping-hole. In slagging out b}' means of a regular " slag-hole," tlie tap- ping-hole can be kept clean. Slag-holes are simpl}' a hole about 2" diameter made from 2" to 8" below the tuyere's bottom, as illustrated in many of the cupolas shown. You can allow the top of the slag-hole up within about one inch of the bottom of the tU3'eres ; if an}' nearer than this, the cold blast entering the cupola has a tendency to chill tlie slag, and, if your tuyere is a continuous one, blow it back. Should the tuyeres be such as have a space between them, then place the slag-hole about in the middle of the two tuyeres which are the farthest away from the tapping-hole. "When the slag commences to accu- mulate, the slag-hole, having been stopped up with clay, is " tapped," and, in some cases, is left open during the balance of the heat ; then the blast blowing out carries slag with it. In others, when slagging out through a slag-hole, opening and closing it is done at intervals, but before opening it the liquid iron is allowed to rise nearly to a level with the hole ; which brings the slag upon a level with the slag-hole, so that it can readily run out when the hole is opened. After the slag is nearly all out, if the metal has not risen so as to compel the cupola to be tapped out in order to keep the metal from running out of the slag-hole, the slag-hole is then stopped up, and the cupola tapped out. The height to place a slag-hole should be chiefly regulated l)y the class of work to be done. AVliere the metal must be car- ried away by small or hand ladles, the slag-hole should be lower than if the metnl is carried away V)y crane ladles. In using small ladles, it is not desiral)le to allow much of a head to accumulate ; whereas, with crane ladles, a body is 812 si,.\(;(;iN(; oir citolas. often :ill()\v('(l (i) nccnimilulc in oitlcr lluit tapK uv.xy yield ji larjjjf Mnionnt cadi tinir. In liie latter Ciise. this necessity may arise on accdinit of waiting for :i ladle to he retiirned ; then, ajfain, it may lie best to liayo larc excelled, especially in the lari^e cupolas. In fact, when properly fliixed, utiiggcil, and tuyere oiuiiiO , large cupolas could often l)e luadu to run lu* long aj< the lining would stand the cousUint heat AREAS OF TUYERES AND BLAST PIPES 315 AREAS OF TUYERES AND BLAST PIPES. It is evident from an examination of the table upon p. 321, giving the ratio of the area of the tn3'ere to that of the cupola, that there exists a great variation in the ratio of tuyere to cupola area allowed in the cupola-practice of America, ranging, as is seen, from 4.32 to as high as 30.75 per cent. By this some may be led to think that most any area will do for the admit- tance of air to a cupola. There is no question but considerable difference in the per cent of tuyere area can be used with but little or no ill results. While great variation in tuyere area is admissible, some ill effects will undoubtedly result from an indiscriminating adop- tion of tuyere area. The author is far from affirming tliat there would be no observable difference m the working of two cupo- las, both same diameter and run under like conditions, but one having a tuyere area of only four, and the other of thirty, per cent of that contained in tlie cupola. Upon general principles, small tuyere areas cause shorter- lived melting than large tuyere areas. This question will be found discussed upon p. 320, vol. i. and p. 302, vol. ii. What tuyere area is the best to adopt, the reader will be better able to understand, after reading the pages above referred to, and a study of the tables and formulas at the end of this chapter. In taking up the question of blast-pipes, areas, etc., attention is first called to the table of B. F. Sturtevant's for equalizing the diameter of pipes (p. 316); which is very valuable, as by it one can readily learn the number and diameter of branch pipes necessary in conveying of blast from the main pipes to a cupola. "i^»T°i*iTi¥i"iVi*iV at 5 1 1 CD oo>j|aft|Oi A u M M ':} ^ .y r V ^ ^, '." !^ -.'3,:i!$lc '¥i?r §s ts H -U 1 g|?i £885 1 iM "I - ■'M|-|* i|S - .,1 ! ^ -i| £ ^1 S! to {o & S{ « »• 1 GB l| I*" r ■ :i "Z- i|S]s e Si-.|g 1 1 ! S|8|8jS 10 00 V CO ^ e-to" b ^,-,s,i "1^ «| £ OD| ^^ s fe|fe|£3 .•g !3|8 SI Si to b -4 «• b ^ «- bo im o " g|li :5 ssjsjs !^ r* U"^ y !^ r' ^ b' b ^ CO io col b 1 ^,¥ '1 S|SJS5|g g|g « Si S 1 CO — « 00 a> bi CO -• A b 00' u cr i 1^ - E i;S g g g ff Sg|£ 4> u S « io 1 -« » «;> 4- b ^ ^ ^ u bo cslio|— '— -J Ui 4>' b '*- •2. !' ^ ? 5 1 = E- Si », S 4. 09 •i ^l*i"i" b 1 00 -4 p> bo !-i' b -l' 00' — »o| to bo' io -l' CO OD ? 2 E r £- 2 ^. .5 S 00 00 Oil *l. ►* 1 1 8 :s;s|k CO p> CI CI ^1 «. *.|co co]to|io — — © b' to' b b C' b -1 col X. Ot v\ »l li! — S o Si en, o> CO — Oo| -4 1 b' to f 1 CO CO CO lol to bo> io bol CO b b SO ^ SgjS to to o oo 1 °> b f- 4>. CO C3 to loj t b| ii. b bj t J-" bo «< ^\^ iiimniii o> * g 8 b 00 CO o>l o- bl -1 4^ CO b c.. to to to > bo| 4. Ll c O C). ii'tO ^ =■ ^ ^ ~" '"'' 2 '^.c. "■ f K 4. ' CO 1 IC c^f'=. " o!l-2;?^J S^l !g K S 1:1 i\t CO CO to to x* iol b b i-lt 1 1 oo b 5" -If- ocl o CO ►0 b to co| b *-| to'"- .J n „_B-£~£ O c r. S = p ■ "2. o£ rt=- ~ c-n-7 — •-: < " :: 's. 2 = o <-- § cr z £ X o> >- b> «. «- col to i-!b to to c :M = :: i f|:IJr w o> ^ CO CI' to b toj >-■ 4-1 b c; to Coj M lo| to — b! b "c-'l ^ ^ CO ds <* mi o MJ b[ b' bl io bo; «- r; CO c. to 00 Ol .fa «c' ts b ifa Ml »! M b bl cc to o S. 5 f £ J' o =• E: =• i =■ -; «i « U> O m! 1.5 col In io to CO " S 3- 3- " 3 n re S * e 2 5 " £. 3 i s3j? S ' ^ - :i HU « 5^ _ -^ c i K n 2 Z _ S H - ^< til-il f.|*-|Kil-|-;-:ta SSlssic Bczoit gresti la •:ibl^|giitli:!i:i« if ^5 5*1^1 liS^s ££:;ts* is?? sislslsisli: 00 t:l£|s|gis CO oj 8235feg gSaSS SSE^S rill 1?:? sii^!if|r:ig sifisgi "-.stft* s?£ssas cSstcS SioutoS i3SS3£ i^iblrfel ^n (X| MXmmm^. -^ 1 316 AREAS OF TUYERES ANT) IU,AST TIPES. 817 Sturtovmit's table is one which will not only save labor in cal- culating areas of blast-piiies, but is valuable in other respects; one of which is prominently showing the retarding elTect of friction in the deliver}- of volumes of air through long pipes, and the advisahiUty of placivg blowers near to a cupola in order to save cost iii motive 2)^'>oer, for the cost to suppl}' motive power to drive air through lo)ig pipes is something worthy of consideration. The nearer a blower can practically' be placed to a cupola, the better results in every way will be produced. Blast-pipes should be sudiciently large to convey the required volume of air without undue loss by friction. The longer the distance air is carried, the larger in diameter should the pipes be. Where small conducting-pipes are used, much more power is necessary, as a greater velocit}' is required to discharge a given amount of air ; the friction being increased in the ratio of the square of the velocit}' with which the air moves. The table of Baker's (p. 318), giving the diameter of maiu blast-pipes, will be found a valuable companion to the Sturte- vant's table, in determining the areas of maiu and branch blast-pipes. "Blast-pipe should, in all cases, be air-tight. A few small holes often cause trouble, the blower having to be run faster to make up for leakage, which is only waste of power, and, as the pressure in the blast-pipe Increases, the escape is also in pro- portion : therefore it will be impossible to force through the furnace the requisite amount of air. Diameter of blast-pipes should be in proportion to the size of cupola, so that the air delivered may not be forced to travel faster through the pipes than sixty feet per second. If the pipes exceed fifty feet in length, their diameter should be increased somewhat (on account of the friction of the air in the pipes). For every additional fifty feet it would be well to add one inch to the diameters siven al)ove." — Bakku. ;Us AKKAS OI- ITYKIM-.S AND P.I.AST I'lPES. IlAKKirs TAIILK. Civini; tlio Di;iii\ftir i<( .M:iiii lUast-PiiK's for all T'lipolas ransjinir from IS' to S4* iiisiilo (liaiiuttT. Lnii^tli of I'ipi's to Iw .'>(> feet. I>1A.MKTKII DiAMKTEU DiAMRTER Diameter I>IAMKTKI( I>IAllKTi:U OK OP OP OP OK OK Cupolas. riPE. Cupolas. Pipe. Cupolas. Pipe. Inches. Inches. Inches. Inches. Inches. Inches. IS r> 41 Hi (i;} 17J 1!) H 42 lU (A IS 20 5J 43 12 0."» IS} 21 G 44 m 0(5 isi 22 Oi 45 m 07 18} 23 6i 46 13 m 19 24 C| 47 13} GO lOi 2r> ^ 48 13i 70 10} 20 n ! 40 13| 71 20 2 1 -1 1 50 14 72 20} 28 8 : 51 1-H 7-! 20i 20 H i 52 14f 74 20} 30 Si 53 15 75 21 31 8J 54 15} 70 2H 32 9 55 15i 77 21} 33 Oi 50 15| 78 22 34 9i 57 16 70 22} 35 9| 58 16} SO 22i 36 10 50 lOi 81 22} 37 10^ 00 16| 82 2;? 38 lOf 01 17 SS 23i 39 1 11 02 Hi 84 24 40 Hi To foini)l('tc tliis clinpUM-, the author will o;ivo his original fornuihis for tiiidiug the area of the tuyeres for dilferent diame- ter cupolas, etc. The first is the maximum area advisable, and is simi)ly to construct tuyeres of such area that their sum shall be tweut}'- five per cent of the average area of the cupola, calculated on AREAS OF TUYERES AND BLAST PIPES, 310 its inside diameter. Tliis would give a 40" cupola six S^^" round, or a 2^" open flat continuous tuyere. To (ind tiie medium area of tuyere : Divide the area of the cupola by 9. This gives an area of tuyere of 11 ^ per cent of that contained in a cupola ; and gives a 40" cupola six of" round, or a 1^-" open flat continuous tuyere. To find the minimum area of tu3-ere : Divide the area of the cujiola by 20. This gives an area of tuyere of 5 per cent of that contained in a cupola ; and would give a 40" cupola six 3|" round, or a ^" open flat continuous tuyere. To find tuyere areas ranging from medium up to maximum, the ratio would of course increase in per cent by decreasing the divisor. The figure 8, used for a divisor, would give 12^ per cent; 7 would give 14f per cent; G would give 16f per cent; 5 would give 20 per cent. To find area of tuj'eres ranging from medium down to mini- mum, the divisoi's would of course increase from 10 up to 19. The 40" cupola is used ses an illustration of the different areas of the tu3'eres resulting from these formulas ; but if the first formula were emploA'ed, and the tuyeres were round, it would be better to increase the number of tuj^eres to seven or eight, as this will give a smaller diameter to each, and distribute the blast more evenly around the cupola, — a point worth con- sidering in designing a cupola. When, by any of the above formulas, the tuyere area is obtained, it will then be divided by whatever number of tuyeres are desired. Then, if the tuyeres are intended to be round, square, or flat, the dimensions of the tuyere can be readily found by referring to page 322, containing the areas and circum- fereuceS of circles and squares. Should the tuj'ere be of other shape, the subdivided areas would then require special figuring to obtain the dimensions of the form of tuyere desired. "With reference to which of the above formulas it is best to adopt, the reader is recommended to consider the conditions .S20 AUKAS ()!• TIYKUI-.S AND lil.ASI' I'lI'KS. rcforri'tl to in tlio fnrc-pHrl of this cliuitttr. and adopt tliat one Ik'sI snilin'4 tlic iri|iiiicnifnls »»r the intended 40.25 m 102.8S74 842..391 1072.502 40^ 128.0202 1304.20(5 1000.502 33 103.0728 855.301 1080. 41 128.8050 1320.257 1081. 33J 104.4582 808.309 110.5.502 41} 120.5010 1330.407 1701.502 m 105.2436 881.415 1122.25 4U 130.37(34 1352.055 1722.25 331 10(i.0290 894.02 1139.002 41| 131.1018 1300.001 1743.002 34 100.8144 907.922 1156. 42 131.0472 1385.45 1704. ;34i 107.5908 921.323 1173.002 42} 132.7320 1401.09 1785.062 34J ias.3852 934.822 1100.25 424 133.5180 1418.63 1806.25 34| 100.1700 948.42 1207.502 42^ 134.3034 14.35.37 • 1827.562 35 100.0500 002.115 1225. 43 135.0888 1452.2 1840. 3.H 110.7414 075.000 1242.562 43} 135.8742 1400.14 1870.562 3.-.^ 111.52(i8 080.8 1200.25 43^ 130.0500 1486.17 1802.25 3.-,| 112.3122 1003.70 1278.002 4-H 137.4450 1503.3 1014.002 3G 113.0070 1017.878 120(>. 44 138.2308 1520. .53 1030. 36i 113.8830 1032.005 1314.002 44} 130.0158 1537.86 1058.002 3tH 114.0084 1040.349 i:W2.25 44i 130.8012 1.5.55.29 1080.25 30^ 115.45:;8 1000.732 13.50. .502 44} 140.5800 1.572.81 2002.. 5(52 AREAS OF CIRCLES AND SQUARES. 325 CIRCUMFErtEXCE AND AREAS OF CIRCLES ; ALSO, THE AREAS OF SQ,\J ARES, — Continued. ? o Circum- Area of Area of S o Circum- Area of Area of E 2 .2 •- feionce. Circles. Squares. 5 ° ference. Circles. Squares. 45 14L3720 1590.43 2025. 53 166.5048 2206.19 2809. 45i 142.1574 1608.16 2047.562 53i 167.2902 2227.05 2835.562 4H 142.9428 1625.97 2070.25 53J; 168.0756 2248.01 2862.25 45| 143.7282 1643.89 2093.062 53| 168.8610 2269.07 2889.002 4G 144.513G 1661.91 2116. 54 109.6464 2290.2.3 2916. 40^ 145.2990 1680.02 2139.062 54i 170.4318 2311.48 2943.002 4G| 146.0844 1G98.23 2162.25 541 171.2172 2.332.8:3 2970.25 4Gf 14G.8698 1716.54 2185.562 54| 172.0026 2354.29 2997.562 47 147.6552 1734.95 2209. 55 172.7880 2375.83 3025. 47^ 148.440G 1753.45 2232.562 55i 173.5734 2.397.48 3052.562 47* 149.22G0 1772.06 2256.25 551 174.3588 2419.23 3080.25 47f 150.0114 1790.76 2280.062 55| 175.1442 2441.07 3108.062 48 150.7968 1809.56 2304. 56 175.9296 2463.01 3136. 48i 151.5822 1828.46 2328.062 56i 176.7150 2485.05 3164.062 4SJ 152.3076 1847.46 2352.25 56 1 177.5004 2507.19 3192.25 48| 153.1530 1866.55 2376.562 5G| 178.28.58 2529.43 3220.562 49 153.9384 1885.75 2401. 57 179.0712 2.551.76 3249. 49i 154.7238 1905.04 2425.562 57i 179.8566 2574.2 3277.562 41)i 155,5092 1924.43 2450.25 57j 180.6420 2596.73 3306.25 49| 156.2946 1943.91 2475.002 181.4274 2619.36 3335.062 50 157.0800 1963.5 2500. 58 182.2128 2642.09 3364. 50i 157.8654 1983.18 2525.062 58J 182.9982 2664.91 3393.062 50^ 158.6508 2002.07 2550.25 58J 183.7836 2687.84 3422.25 50| 159.4362 2022.85 2575.562 58| 184. .5690 2710.86 3451..502 51 160.2216 2042.83 2G01. 59 185.3.544 27.33.98 3481. 5U 161.0070 2002.9 2026.562 59J 186.1398 2757.2 3510..5G2 oH 161.7924 2083.08 2652.25 59| 186.92.52 2780.51 3.540.25 51| 162.5778 2103.35 2678.062 59f 187.7106 2803.93 3570.062 52 163.3632 2123.72 2704. 60 188.4960 2827.44' 3600. 52J 164.1486 2144.19 2730.062 GOJ 189.2814 2851.05 3630.062 52i 164.9340 2164.76 2756.25 G0| 189.0668 2874.76 3660.25 52f 165.7194 2185.42 2782.562 60| 190.8522 2898.57 3690.502 ;-jti AKKAS or ClltCl.IS AM» SQIAKKS. CIKCrMFKItKNCK AX!) Ai:KAS OF CIUfLES ; ALSO, TliK AicKAs OK s(^rAi;Ks, — r„;i/;/,,(c,/. 1 s ("Ircum- Area of y\r<'ii of o 5 ('irciim- Area of Area of fercncu. ClrclcH. Squuruii. ft o fcreiicv. Circles. BquarcH. Gl 101.6370 2922.47 3721. 09 216.7704 37:39.29 4761. cu 192.4230 294(5.48 3751.502 (59} 217.55.58 37(5(5.43 4795..562 Oli 193.2084 2970.58 3782.25 09J 218.3412 3793.08 4}S;30.25 61| 193.9938 2994.78 :J813.002 G9J 219.1206 3821.02 4S65.(J62 (52 194.7792 3019.08 3a44. 70 219.9120 3S48.46 4900. 02J 195.5646 3043.47 3875.002 70} 220.6974 3876 49:35.0(52 62i 196.3500 30(57.97 3900.25 70 i 221.4828 3[K)3.03 4970.25 62J 197.1354 30!)2.5G 3937.502 70J 222.2(582 3931.37 5005..562 63 197.9208 3117.25 3909. 71 223.0536 3959.2 5041. 6:H 198.7062 3142.04 4000.502 71} 223.8;J90 3987.13 507<5.502 caj 199.4910 316().93 4032.25 in 224.6244 4015.10 5112.25 6:]| 200.2770 3191.91 4004.002 71| 225.4098 404:5.29 5148.002 04 201.0024 3217 4096. 72 226.1952 4071.51 5184. Wl 201.8478 3242.18 412S.0(]2 72} 220.980(5 4099.84 5220.002 64^ 202.0332 32(57.40 4160.25 72i 227.7(500 4128.20 52.50.25 64J 203.4180 3202.84 4192.562 72| 228..5514 41.50.78 5292.562 65 204.2040 3:318.31 4225. 73 229.3.308 4185.4 5329. 65} 204.9894 3343.89 4257.562 73} 230.1222 4214.11 5:3(35.562 65i 205.7748 33(39. .50 4290.25 v>\ 2:30.9076 4242.93 .5402.25 65f 200.5002 3395.33 4323.062 m 231.69:30 4271.84 5439.062 ()(i 207.345(i 3421.2 4356. 74 232.4784 4300.85 5470. 60} 208.1310 3447. 17 4389.002 ■74} 233.2638 4329.90 5513.002 66^ 208.9104 3473.24 4422.25 74i 234.0492 4:!59.17 5550.25 6(iJ 209.7018 3499.4 44.55.502 74J 2:34. 8:U6 4:588.47 .5587.502 67 210.4872 3525.60 4489. 75 2:35.6200 4417.87 5(525. 67i 211.2726 3552.02 4522.502 75} 236.4054 4447.:38 5(502.5(52 67^ 212.0580 a578.48 4.5.50.25 75^ 2:37.1908 447(5.98 5700.25 G7f 212.84.34 3005.04 4590.062 75J 237.9762 450(5.07 57:38.062 68 213.0288 3031.(59 4(524. 70 238.7610 4530.47 5770. GS\ 214.4142 3058.44 4658.062 70} 2:39.5470 45(50. ;30 .5814.002 68i 215.1990 3085.29 4(592.25 70J 240.:3324 4590.;30 58r)2.25 68| 215.9850 3712.24 472(5.562 76| 241.1178 402(5.45 58iX).562 AREAS OF CIRCLES AND SQUARES. 327 CIRCUMFERENCE AND AREAS OF CIRCLES ; ALSO, THE AREAS OF aqu Aims, — Continued. 1 § Circum- Area of Area of S § Circum- Area of Area of E 2 5 ° ference. Circles. Squares. S u a J. C ° 85 ference. Circles. Squares. 11 241.9032 4656.64 5929. 267.0360 5674.51 7225. m 242.6886 4686.92 5967.562 85i 267.8214 5707.94 7267.562 rii 243.4740 4717.31 6006.25 85^ 268.0068 5741.47 7310.25 771 244.2594 4747.79 6045.062 S5f 269.3922 5775.1 7353.062 78 245.0448 4778.37 60S4. 86 270.1776 5808.82 7396. 78i 245.8302 4809.05 0123.062 86i 270.9630 5842.64 7439.062 78^ 246.6156 4839.83 6162.25 86 1 271.7484 5876.56 7482.25 78| 247.4010 4870.71 6201.562 86| 272.5338 5910.58 7525.562 79 248.1864 4901.68 6241. 87 273.3192 5944.09 7569. 79i 248.9718 4932.75 6280.562 87i 274.1046 5978.91 7612.562 71)^ 249.7572 4963.92 6320.25 Slh 274.8900 0013.22 7656.25 79f 250.5426 4995.19 6360.062 S7| 275.0754 6047.03 7700.062 80 251.3280 5026.56 6400. 88 276.4608 0082.14 7744. 80i 252.1134 5058.03 6440.062 88^ 277.2462 6116.74 7788.002 80| 252.8988 5089.59 6480.25 881 278.0316 6151.45 7832.25 80| 253.6842 5121.25 6520.562 8S| 278.8170 6186.25 7876.502 81 254.4696 5153.01 6561. 89 279.0024 6221.15 7921. 81i 255.2550 5184.87 6601.562 89\ 280.3878 6256.15 7965.562 8H 2.56.0404 5216.82 6642.25 89^ 281.1732 6291.25 8010.25 81| 256.82.58 5248.88 6683.002 89f 281.9586 6326.45 8055.002 82 257.6112 5281.03 6724. 90 282.7440 6361.74 8100. 82i 258.3966 5313.28 6765.062 90i 283.5294 6397.13 8145.002 S2h 259.1820 5345.63 6806.25 90i 284.3148 6432.62 8190.25 82} 259.9674 5378.08 6847.562 90f 285.1002 6468.21 8235.562 8a 260.7528 5410.62 6889. 91 285.8856 6503.9 8281. 8;H 201.5382 5443.26 6930.562 9U 286.6710 6539.68 8326.562 8;3i 262.-3236 5476.01 6972.25 91i 287.4564 6575.56 8372.25 831 263.1090 5508.84 7014.062 91| 288.2418 6611.55 8418.002 84 263.8944 5541.78 7056. 92 289.0272 6647.63 8464. 84i 264.6798 5574.82 7098.002 92} 289.8125 6083.8 8510.062 84^ 265.4652 5607.95 7140.25 92i 290.5980 6720.08 8556.25 84| 266.2506 5641.18 7182.562 92| 291.3834 0756.45 8002.562 ]-2H AKKAS OK CIItCl.KS AM) SQl'AUKS. Cinr'T-MFERKXCK AXD AREAS OF CIRCLES; ALSO, THE AKKAS (»K sgl'A RES, — r'.y,,. ■/(«/../. t § ' CIrciini- A roil of Arcii of « g circum- Area of Area of - O furcnce. CircloH. SiinuruH. S I. G ° ference. ClrclcH. BquurcM. 03 202.1688 6702.02 8640. 96J 303.9408 7351.79 9300.502 o:H 202.0542 0820.40 8005.502 07 304.7352 7.380.8:3 9409. mi 203.7300 6866.16 8742.25 07} 305.5206 7427.97 94.57.562 o;}j 204. .53.50 6002.03 8780.002 07J 306.3000 7400.21 9.506.25 04 205.3104 6030.70 88:36. 07f 307.0014 7504.55 9555.002 04} 200.0058 6076.76 8883.0<)2 98 307.8708 7542.08 9004. 04 i 206.8812 7013.82 80.30.25 08} 308.0(522 7581. .52 90.53.002 04J 207.6000 7050.08 8077.562 98^ .300.4476 7020.15 9702.25 05 208.4520 7088.23 0025. 98| 310.2330 76.58.88 0751.. 562 onj 200.2374 71 25.. 50 0072.562 99 311.0184 7007.71 9801. o-H 300.()22S 7163.04 0120.25 99} 311.8038 7730.0;3 98.50. .562 or)| 300.8082 7200.6 9168.002 99^ 312..5802 7775.06 9000.25 06 301. .5036 7238.25 0216. 99| 313..3740 7814.78 90.50.062 00} 302.3700 7275.00 9264.002 100 314.1600 7^54. 10000. OO.i '80.3.1044 7313.84 0312.25 Not only are the above taljles of areas for circle.s and squares useful for the purpose referred to on p. 310, Init also in fijiuring weights of castings ; for in the case of desired iveiijhts for sfjnare or round j)late)i not to he found in vol. i., referring to the aliove table will save the necessity of first figuring to obtain tluir areas before they can be multiplied by the weight of a cubic inch of iron as seen iu vol. i. pp. 37U, 376. AMERICAN CUPOLA TRACTICE. The following fortj'-six reports of cupola-workings have been carefully collected by the author from thirty States, reaching from Maine to Oregon. The reports will not only be found interesting, but very valuable to consult ; giving, as they do, so many different men's ideas and practice in mixing and melt- ing iron. In selecting the firms shown, those were chosen that the author thought used intelligence and system iu their prac- tice. These reports the author believes to be a practical account of the cupola- workings of the respective firms. Each firm's name, and the line of castings made, are given solely for the purpose of attaching authority' to the reports, and to enable foundryraen to classify the workings with their own or intended class of work or castings. In collecting the reports shown, the author would state that considerable stress was laid upon obtaining some knowledge of the fluidity of the iron melted. I believe the questions were conscientiously answered as far as such a thing could practi- cally be done. The XXX shown stands for what shops gen- erally term " good hot fluid iron ; " the XX stands for a medium fluid iron, such as is often suitable for pouring ordinary thick- nesses of machinery castings. When collecting the reports, the length, etc., of blast-pipes was also ol)tained. Only such portions are mentioned as were thought to be of service in giving ideas, etc. ; since, to publish all the bends and different crooks, etc., would only be adding confusion to the reports. The reports as shown argue well for the kind and lil)eral spirit of American foundrymen, iii letting their experience and practice be known ; and, no doul^t, many will feel that they shouhl be credited for their liberality shown. In this the author hearlilv coincides. sua :VM) AMIKKAN (TI'OI.A I'ltACTICK. PORTLAND, ME. COMMON 44" CUrOLA. Otitsiilo (lininctor .M" Tliickiipss of lining .%" Inside diiiinetfT at tuyeres ."'.T" Ijiir;^est iiisiile or lucdtiiij^-point diameter 4ii" Inside diameter at eiiar;;in;i-d()()r 44" Height from bottom |)lar V.M. Dottom dropped .... .".45 " Amount of iron melted, 14,000 lbs. Amount of fuel consumed, 2,250 " Katio of fuel to iron used, 1 to GyS'u- Fluidity of melted iron, XXX. Length of heat, 2 hours. Remarks. - described. "W as the above ; distance from three tons per pipe. Our. iron is jobbing castiu; 0( T. 2:5, 1SS3. -The above heat presents an average working of the cupola e have two other cupolas, one of which is of same diameter the other is M" inside diameter, having four tuyeres 8" X 3"; bottom plate to bottom of tuyere, 14". This cupola will melt hour. The three cupolas are all fed by the same IG" blast- mcltcd for making locomotives, marine, arcliitectural, an" diameter. Ileiflbt frouj bottom plate to boll(^im of lower tuyi^re, 'JO"; to uiiper tiiv(!re Ileij^ht of lower tuyere aViove sand bottom on back sid Ileijjlit from bottom plate to bottom of slag-hole . . Fuel used for bed: coke First charge of iron . . " " coke. . Second charge of iron . " " coke Third charge of iron " " coke Fourth charge of iron 1,400 lbs. 4,000 " 2(i0 " 2,500 " 200 " 2,500 " 200 " 2,500 " Fourth charge of coke P'ifth charge of iron . " " coke Sixth charge of iron . " " c(jke Seventh charge of iron " " coke Eighth charge of iron , 01" Hi" 44" 4t;" 44" IJ' :\'y' 14" IS" 2(X) lbs. 2,500 2(iO 2,.-)00 2(W 2,500 aw 2,500 Eight more charges, continued per order shown. No. 7 Sturtevant fan; diameter main blast-i)ii)e, 12 blower. Cupola 22' from Time of starting fire " charging first iron, 1.00 p.m Bla.st put on 2.40 " 12.00 A.M. First appearance of fluid iron 2. .55 v.m. Bottom dropped .... 7.05 " Revolutions of blower, 2,500. Pressure of blast, (\\ ounces. Kind of fuel used, Coniudlsville coke. Kind of llux used, limestone. Amount of iron melted . 41.500 lbs. Amfiunt of fu(d consumed, 5,o00 " Ratio of fuel to iron used, 1 to 7 nnf. Fluidity of nielle.l iron, XXX. Length of heat, 4li. 25m. Remarks. — Our iron is poured into architectural and light house-work moulds. The last of the iron was just as hot as the first of tho heat. We use limestone on every charge. After easting five t" Style of tuyer(!s: five tuyeres, iO" X 5", at inside; 7" x 5" where it joins the blast-i)ipes. Height from bottom plate to l)ottom of tuyere 15" Height of tuyere above sand bottom on back side 10" Fuel used for bed : coal . 1,800 lbs. Fourth charge of scrap First charge of pig . . 3,000 " " " coal " " scrap . 2,500 " Fifth charge of pig . " " coal . . 400 " " " scrap Second charge of pig . 2,500 " " " coal . " " scrap . 1,500 " Sixth charge of pig . " " coal . 400 " " " scrap Third cliarge of pig . . 4,500 " " " coal . " " coal. . 400 " Seventh charge of scraj Fourth charge of pig . 3,000 " 1,500 lbs. 400 " 2,200 " 1,800 " 400 " 500 " 4,000 " 400 " 4,000 " No. 5^ Baker blower; diameter main blast-pipe, IG". First appearance of fluid Time of starting fire . . 12.30 p.m " charging first iron, 2.00 " Blast put on 3.15 " iron 3.30 p.m. Bottom dropped .... (i.OO " Amount of iron melted, 31,000 lbs. Amount of fuel consumed, 4,200 " Ratio of fuel to irou used, 1 to 7 1'lfii- Fluidity of melted iron, XX. Length of heat, 2h. 45m. Remarks. — Our iron is used for turbine-wheels and mill-machinery castings. W. S. BEECHING, Foreman JJolyoke Machine Co.'s Works Foundry. Oct. 23, 18S0. .•304 ami:rican cri'oi.A iKAciifi;. WORCESTER, MASS. COMJAIJ 2t!" CUPOLA. Onfsido (li.imctor ... VI" Tliickncss of liiiiiis 8" Inside diamoter at tnynrcs 2(i" Larf^ost inside or ineltiiij^-jioint (liiiiinicr IMj" Inside diameter at charsin^i-door W lleiglit from bottom plate up to bottom of cliarfjing-t., J. A. Cvlviii Works Foundry. Kkh. 1, 1S84. AMERICAN CUPOLA PRACTICE. 335 SPRINGFIELD, MASS. COLLI AU 28" CUPOLA Outside diameter 42" Thickness of lining ... .- 7" Inside diameter at tuyeres 28" Largest inside or melting-point diameter 30" Inside diameter at cliarging-door 28" Height from bottom jilate up to bottom of charging-door . . , . 9' 6" Style of tuyeres : two rows of tuyeres, six above and six below. Lower row, oh" square; upper row, 1^" diameter. Height from bottom plate to bottom of lower tuyere 22" Height of tuyere above sand bottom on back side 14" Height from bottom plate to bottom of slag-hole „ . 15" Fuel used for bed: coke . 500 lbs. First charge of pig . . . 1,.500 " " " scrap . . 500 " " " coke . . IKJ " Second charge of pig . . 1,500 " Second charge of scrap . 800 lbs. coke . , IIG " Third charge of pig . . . 700 " " " scrap . . 1,881 " No. 5 Sturtevant fan; diameter main blast-pipe, 8". Time of starting fire . . 3.30 p.m. " charging first iron, 4.40 " Blast put on 5.00 " First appearance of fluid iron ........ 5.15 p.m. Bottom dropped .... 6,15 " Revolutions of blower, 3,150. Pressure of blast, 6 ounces. Kind of fuel used, Connellsville coke. Kind of flux used, oyster-shells. Amount of iron melted, 6,881 lbs. Amount of fuel consumed, 7.32 " Ratio of fuel to iron used, 1 to D^. Fluidity of melted iron, XXX Length of heat, Ih. 15m. Remarks. — Fifty-five pounds of coke was saved from dropped bottom; therefore the ratio of fuel to iron actually consumed would be 1 to 10|'ij"if' This heat was an exceptional one for its size. "With a heat of five tons we can melt 1 to 10 or 11 with ease. We use our iron for machinery and light castings. JAMES SIMPSON. Foi'cman iSiJviiujjidd Foundry Co. iUv 1, 1SS3. IVM', AMI'.UIfAN niol.V rKA( IICK. PROVIDENCE, R.I. MACKENZIE ;«" x n:!" fll'OLA. Outside (limonsions C2"xOf/' Tliickii.'ss of lining V,f," Insidu (liinciisions at tuyori'S ."JO" x 44" Largest inside or iuoltin;^-|i(iiiit W x r>:i" Inside diiiu-nsioiis at fharj^iiiK-door 40" XM" Iloiglit frniu bottom [ilate up to Ijottom of charginp-door . . . 11' G" Style of tuyeres: flat 1" opiMiiiig, continuous tuyere. Height from bottom plate to bottom of tuyere I'j" Height of tuyere above sand bottom on back side 7" FiU'l used for bed : coal . .1,100 11)3 First charge of pig . . . 1,100 " " " scrap . . (JOO " coal . . . 200 " Second charge of jiig . . 1,400 " " " scrap (iOO " " " coal . . 200 " Third charge of pig . . . 1,400 " " " scrap . . (JOO " No. 4J Bak or; diame Time of starting fire . . 1.20 r.M " cliarging first iron, 3.00 " Blast put ou 4.00 " llevohitions of blower, used, Lehigh coal. Kiud Third charge of conl . . 200 Ib.s. Fourth charge of pig . . 1,4(X) " " " s>" CLTOLA. Outside dimensions Insiile dimensions at tuyeres I^!ir;;est inside or ineltinf?-iH)int iliniensinns Inside dimensions at ,(i(Mi " " scrap . . 7,5()0 . . 500 . . 400 2,500 5,000 500 300 2,000 5,000 500 oOO fiOO lbs. :,(XI0 " " " coke " " coal . . Second charge of pig . " " scrap " " coke . " " coal . Third charge of pig . . " " scrap . " " coke . " " coal . Fourth charge of pig . " " scrajt coke. " " coal . Fifth cliargc of pig . . " " scrap . " " coke . " " coal. . Sixth charge of pig . . " " scrap . " " coke . " " coal . Seventh charge of pig . " " scrap No. G Mackenzie blower. First ajipcarancc of fluid Time of starting fire . . 12. .30 p.m. " charging first iron, 2.t. Our class of work is all kinds of engines, pumps, and machinery castings. FREDERICK SIBLEY, Furcmait, Ddamatcr's Iron }y\»ks Fhiindi-y. h'EU. 15, lSb4. AMERICAN CUrOLA PRACTICE. 339 YONKERS, N.Y. ODD STYLE OF CUPOLA. Largest outside diameter 70" Thickness of lining 5" Inside diameter at tuyeres 30" Largest inside or meltiug-point diameter 48" Inside diameter at charging-door 60" Heiglit from bottom plate up to bottom of charging-door 8' Style of tuyeres : two 4" X 12" oblong tuyeres. Height from bottom plate to bottom of tuyere , 16" Height of tuyere above sand bottom on back side 12" Two 10" diameter by 5' long branch pipes convey the blast from the main pipe to the tuyeres. Fuel used for bed: coal . 1,100 lbs. First charge of jiig . . . 4,000 " " " scrap . . 1,000 " coal. . . 400 " Second charge of pig . . 3,000 " " " scrap . 1,000 " No. 6 Sturtevant fan; diameter main blas1>pipe, 16"; length, 30'. Second charge of coal . Third charge of pig . . " " scrap . " " coal . Fourth charge of scrap 400 lbs. 3,000 " 1,000 " 300 " 2,800 " Time of starting fire . . 1.00 p.m. " charging first iron, 3.15 " Blast put on 3.45 " Revolutions of blower, 2,200. Kind of fuel used, Lehigh coal flux used, oyster-shell, at the rate of one peck to one ton of iron. First appearance of fluid iron 3.53 p.m. Bottom dropped .... 6.35 " Kind of Amount of iron melted, 15,800 lbs. Fluidity of melted iron, XX. Amount of fuel consumed, 2,200 " Length of heat, 2h. 50m. Ratio of fuel to iron used, 1 to 7yo%- Remarks. — The class of work made is elevators, gas-engines, and machinery castings. This cupola, in vertical appearance, is somewhat like that of a bulged barrel. 4" above the tuj-^eres it starts a taper that, in the height of 36", increases from 30" to GO" inside diameter. This 60" con- tinues in height for 36" more; at this point it then commences to decrease, and 36" higher up it is again the same diameter as at the tuyeres; this point being at stack, the 30" diameter is continued up to end of same. This style of cupola is not to be recommended as a success for long heats, and I would give a common straight cupula the preference. L. C. JEWETT, Foreman Otis Brothers & C'o.'s Works Foundry. Dec. 15, 1883. ;;4<> AMKUICAN (M'pol.A I'KACTKK. SYRACUSE, N.Y. COMMON 40" C'l'I'OLA. Ontsido (liamotor ^'■'•" Tliickiifss of liiiinjT s\" Iiisidi! (liain(!tor at tuyorcs •n" l^arj,'fst insido or lUfltiiif^-pniiit (liaiiictcr 4_'" Inside (liainotcr at cliarf?iii!^-iloor ."-'i" Hci;:lit from bottom plate lip t(j bottom of cliarging-door '.»' Style of tuyeres: four (I" x (i" triangular tuyeres. J leiglit from bottom plate to bottom of tuyere 11" lloii'lit of tuvere above saud bottom on back side b" Fuel used for bed: coal . 1,050 lbs. Second charge of i)ig . . 2,700 Ib.s First charge of pig . . 3,000 " " " .sera]) * IKJO " " " scrap . 1,000 " " " coal . •JOO " " " coal . . 400 " Third charge of scrap . . 2,o00 " No. 7 Sturtevant fan; diameter main blast-pipe, 12'' Time of starting fire . . 1..".0 p.m. " charging lirst iron, ;>.:i0 " Blast put ou 4.20 " First appearance of fluid iron 4.27 p.m. Bottom drojiped .... r>.j') " Revolutions of blower, 2,500. Kind of fuel used, Lebigb coal. Kind of flux used, tluor spar. TOTALS. Amount of iron melted, *),nOO lbs. Amount of fuel consumed, 1,650 " Katio of fuel to iron used, 1 to G. Fluidity of melted iron, XXX. Length of boat, lb. 35m. E.KMARKS. — The class of work made is for stationary engines. The heat is a small one for the cupola; therefore the percentage is not as high as it would be were the heat a larger one. PATRICK EG AN, Foreman The iilrai(jht Line Emjine Co. Foundry. Nov. 16, 1S83. AMERICAN CUrOLA PRACTICE. 341 ROCHESTER, N.Y. COLLIAU 48" CUrOLA. Outside diameter 02" Thickness of lining 7" Inside diameter at tuyeres 48" Largest inside or melting-point diameter 48" Inside diameter at cliarging-door 48" Height from bottom plate up to bottom of charging-door 12' Style of tuyeres: two rows of tuyeres; lower row, oblong; upper row, round; lower, 9"X 4"; upper, 2h" diameter. Height from bottom plate to bottom of lower tuyeres, 24"; to upper tuyeres 40" Height of tuyere above sand bottom on back side 21" Height from bottom i)late to bottom of slag-hole ITg" Fuel used for bed: coke . 1,400 lbs. First charge of pig . . . 1,515 " " " scrap . . 1,852 " " " scrap . . 1,852 " " coke 240 " " " coke . . 240 Second charge of pig . . 1,515 " Fourth charge of pig . . 1,515 " " scrap . 1,852 " " " scrap . 1,852 Second charge of coke Third charge of pig . 240 lbs. 1,515 " Seventeen more charges, continued per order shown. No. 9 Sturtevant fan; diameter main blast-pipe, 14" at blower, 12" at cupola. Time of starting fire . . 10.10 a.m. " charging first iron, 11.20 " Blast put on 12.30 p.m. First appearance of fluid iron 12.35 p.m. Bottom dropped .... 4.45 " Revolutions of blower, 1,800. Pressure of blast, 8^ ounces. Kind of flux used, limestone. TOTALS. Amount of iron melted, 70,707 lbs. I Ratio of fuel to iron used, 1 to lli%. Amount of fuel consumed, 6,200 " I Length of heat, 4h. 15m. Remarks. — In this heat the above amount was meltod, having a uniform tcmiierature from first to last. The metal was poured into car-wlieels. Oct. 23, 1883. EDWARD J. CAMPBELL, Superintendent Rochester Car-Wheel Works. ni2 AMl'.UICAX CT'l'OI.A l-KArTiri:. JERSEY CITY, N.J/ <"x42" L:ir^ist iiisido or incltiiiK-pxiiit irmiciisiniis lHi"x54" Tii.sidc tliiiicnsions at r ll(i"xr>4" Jlti^'lit fruiu bottom plate up to liottoin of ehar'^ing-door . . f>' "J" btyle of tuyeres; Hat 1" opeuiug, continuous tuyere. Height from bottom plate to bottom of tuyere, 12" front ami .S" back. Height of tuyere above sand bottom on back side 4" Fuel used for bed : coal . ".,000 lbs. First charge of iron . . . 14,000 " " coal . . . l,'-'00 " Second charge of iron . . 14,000 " coal . . l.oOO " Third charge of iron . . 14,000 " Third ehargo of coal Fourth charge of iron " " coal Fifth charge of iron . " " coal . Sixth charge of iron . . 1,200 lbs. .14,000 " . 1,:500 " .12,000 " . i,:joo " .12,000 " I. P. Morris Co.'s oO" x 24" blowing engine. Time of starting fire Blast put on . . . . 11.00 A.M. I First appearance of fluid . 1.00 P.M. I iron 1.20 p.m. Bottom dropped, 5.08 p.m. Stroke of blower, 70. Pressure of blast, 12 ounces. Kind of fuel used, Lehigh coal. Amount of iron melted, 80,000 lbs. Amount of fuel consumed, 9,.'KX) " llatio of fuel to iron used. 1 to S/o. Fluidity of melted iron, XXX. Length of heat, 4h. 8m. Remarks. — The iron is used for heavy engine and machinery castings. DAVID J. MATLACK, Foreman I. P. Morris & Co.'s Works Foundry. Oct. 25, issa AMERICAN CUrOLA PRACTICE. 345 ERIE, PENN. COMMON ?^'l" CUPOLA. Ontside diameter 40" TJiickuess of lining 5" Inside diameter at tuyeres ."51" Largest inside or melting-point diameter 35" Inside diameter at charging-door ;>0" Height from bottom plate up to bottom of charging-door 9' 8" Style of tuyeres : tlat, J" opening, continuous tuyere. Height from bottom plate to bottom of tuyere 15^" Height of tuyere above sand bottom on back side sy Fuel used for bed : coke . 120 lbs. Second charge of coke . 80 lbs coal 300 " " " coal . 70 " First charge of pig . . 900 '• Third charge of pig . . 900 " " " scrap . GOO " " " scrap . 600 " " " coke . . 80 " " " coke . 80 " coal . . 70 " " " coal 70 " Second charge of pig . 900 " Fourth charge of pig . . 900 " " " scrap. . GOO " " " scrap . . GOO " No. 4 Sturtevant fan; diameter main blast-pijje, 7". Time of starting fire . . 2.10 p.m. " charging first iron, 3.10 " Blast put on 4.00 " First appearance of fluid iron 4.0G p.m. Bottom dropped .... 5.05 " Revolutions of blower, 3,100. Kind of fuel used, Shamokin coal and Connellsville coke. Kind of flux used, Kirk's flux. Amount of iron melted, 6,000 lbs. Amount of fuel consiimed, 870 " Ratio of fuel to iron used, 1 to G^%. Fluidity of melted iron, XXX. Length of heat, Ih. 5m. Remarks. — The above workings show average results. Our work is chiefly engine-castings. DAVID SMITH, Foreman Skinner & Wood's Works Fonndry. Oct. is, 1SS3. 346 AMKiiir-AN rrrnT,A ruArTicE. PITTSBURGH, PENN. COMMON .M" cirni-A. Oiitsiilc tliiimiHcr T'J" Tliickiioss of lining U" Inside (liimifttT at tuj'crcs rii" Ijarp-st inside or mcltiiiti-pniiit (liaiiic'tiT r>>'," Inside diameter at cliar;;ing-door r.J" ITei^lit from bottom ]ilate up to bottom of eliart:iii;^-door 1-' Style of tuyeres : flat 1" opening, continuous tuyere. Heij^bt from bottom plate to bottom of tuyere 2<>" Hci^zbt of tuyere above sand bottom on back side 11" Ileiglit from bottom plate to bottom of slag-hole Hi" Fuel used for bed : coke . 1,400 lbs. Second charge of coke . . 200 Uis. First charge of iron . . . 4,.500 " " coke. . . 200 Second charge of iron . . 2,500 Third cliarge of iron . . 2,500 " " coke . . 200 Fourth charge of iroa . . 2,500 Six charges more, continued per order shown. No. 8 Sturtevant fan; diameter main blast-pipe, 12". Tiraeof starting fire . . 12.00 a.m. " charging first iron 1.30 p.m. Blast put on 2.50 " First appearance of fluid iron 3.00 p.m. Bottom dropped . . . . HA) " Revolutions of blower, 2,200. Pressure of blast, to 11 ounces. Kind of fuel used, Connellsville coke. Kind of tiux used, limestone or oyster-shells. Amount of iron melted, 27,000 lbs. Amount of fuel consumed, .3,200 " Ratio of fuel to iron used, 1 to S^^g. Fluidity of melted iron, XX. Length of heat, 2 hours. Re:mauks. — The class of work made is heavy steam and bla.st engines, and macdiinery. We have two 54" and one 'Mi" cupola ; also one air fur- nace. Our large cupolas can melt .'55 tons without trouble. "SYM. II. COXNER, Foreman Mackintosh & Hemphill {Fori Pill) Works Foiaidnj. Feb. 25, 1SS4. AMERICAN CUPOLA TRACTICE. 347 BALTIMORE, MD. COLLIAU 54" CUPOLA. Outside diameter ' 7-" Thickness of lining 9" Inside diameter at tuyeres 54" Largest inside or melting-point diameter 54" Inside diameter at cliarging-door 54" Heiglit from bottom-jdate up to bottom of charging-door 14' Style of tuyeres : two rows of tuj-eres, six above and six below; lower row oblong, upper row round, lower G" X 12", upper row 3" diam. Height from bottom plate to bottom of lower tuyeres 2(1" " " " " to upper tuyeres 45^ " " " " to bottom of slag-hole 20" Fuel used for bed : coke . 2,000 First charge of pig and scrap, 4,000 First charge of coke . . 240 Second charge of pig and scrap, 4,000 Second charge of coke . . 240 Third charge of pig and scrap, 4,000 Third charge of coke . . 240 Fourth charge of pig and scrap, 4,000 Five more charges, continued per Fourth charge of coke . . 240 lbs. Fifth charge of pig and scrap, 4,000 " Fifth charge of coke . . 240 " Sixth charge of pig and scrap, 4,000 " Sixth charge of coke . . 240 " Seventh charge of pig and scrap, 4,000 " Seventh charge of coke . 240 " Eighth charge of pig and scrap, 4,000 " order shown. No. 6 Baker blower; diameter main blast-pipe, 22" Time of starting fire . . 11.00 a.m. " charging first iron, 12.00 " Blast put on 1.30 p.m. First appearance of fluid iron 1.45 p.m. Bottom dropped .... 4.30 " Revolution of blower, 120. 'Pressure of blast, 9J ounces. Amount of iron melted, 52,000 lbs. I Ratio of fuel to iron used, 1 to lO^'g. Amount of fuel consumed, 4,880 " I Length of heat, 3 hours. Rkmauks. — We use 15 per cent wheel-scrap and 85 per cent charcoal jiig metal. Our heats range from 50,000 up to 150,000 pounds. Our iron is good and hot. WILLIAM HYSAN, Foreman Baltimore Car Wheel Co.'s Foundry. Feb. 14, lSS-1. MH A.Ml'.Kli AN Cfl'nl.A rKAC'lICK. WILMINGTON, DEL. Outsidn (lifimptor •'»0' TliickiH'ss of liiiiiiR (>' liisidc! diaiiiotiT at tuycrt-s Lar^ji'st iiisidi! or nicltinj^-point dianifter Iiisidt! diaiiU'ttT at (•liar'rin;'-door . ray ](>' 10" lleiKlit from Ijottom j)latn up to bottom of cliargiiifj-door . . Style of tuyeres: Hat '_'" ojteuiug, continuous tuj-cre. Height from bottom plate to bottom of tuyere l'.'" Iluiglit of tuyere above sand bottom on back side C" Fuel used for bed : coal . l,0.-,0 1bs. Third charge of iron .•!,n<»0 lbs First charge of iron . . . a.oix) " " " coal ].» " " " coal . . . 150 " " " coko . 75 " " " coke . . 75 " Fourth charge of iron . n/xx) " Second charge of iron . . .-..ooo " coal . l.V) " " " coal . . 150 " " *' coko . 75 " " " coke . 75 " Fifth charge of iron . . 3,000 " Sturtevant fan; diameter of main blast-pipo, 12' Time of starting fire . . 12.00 a.m. " charging first iron 2.00 v.M. Elast put on 3.00 " First appearance of fiuid iron .'!.07 p.m. Bottom droi>ped .... 5.15 " Revolutions of blower, 2,500. Pressure of blast, 8 to 12 ounces. Kind of flux used, oyster-shells. TOTALS. Amount of iron melted, 15,000 lbs. | Ratio of fuel to iron used, 1 to l^^g. Amount of fuel consumed, l,ii50 " | Length of heat, 21i. 15m. Rkmarks. — The above is the working of our smallest cupola. Our castings are for marine aud heavy machinery work. AVllJJA?*! STFAKT. Foreman Puscij tO Jv)us Co.'t! Ho/As Foiindnj. Nov. 25, 1SS3. AMERICAN CUPOLA PRACTICE. 349 CINCINNATI, O. COMMON 42" CUPOLA. Outside diameter 00" Thickness of lining \)" Inside diameter at tnyeres 34" Largest inside or melting-jioint diameter 42" Inside diameter at cliarging-door . . , 42" Height from bottom phite up to bottom of charging-door 8' Style of tuyeres : eight round tuj-eres, four 2" and lour SJ". Height from bottom plate to bottom of large tuyere IG" Fuel used for bed : coke 750 lbs. Fourth charge of coke . 100 lbs First charge of pig . . 1,100 " Fifth charge of pig . . 550 " " " scrap . 900 " " " scrap . 450 " " " coke 100 " " " coke. . 100 " Second charge of pig . 550 " Sixth charge of pig . . 550 " " " scrap 450 " " " scrap . 450 " " coke . 100 " " coke . 100 " Third charge of pig . . 550 " Seventh charge of pig . 550 " " " scrap . 450 " " " scraji 450 " " " coke . 100 " " " coke 100 " Fourth charge of pig . 550 " Eighth charge of pig . 550 " " " scrap 450 " " " scrap . 450 " Eleven more charges, continued per order shown. No. 5 Root's blower; diameter main blast-pipe, 15". Time of starting fire . , 1.00 p.m. *' charging first iron 2.00 " Blast put on 3.30 " First appearance of fluid iron Bottom dropped .... 3.35 P.M. 5.25 " Revolutions of blower, 150. Kind of fuel used, Connellsville coke. Amount of iron melted, 20,000 lbs. Amount of fuel consumed, 2,550 " Ratio of fuel to irou used, 1 to Ti^j^. Fluidity of melted iron, XX. Length of heat, Ih. 55m. Re.marks. — Our castings would be classed as light, the machine castings being prini'i[ially for wood-working machinery, and more than half of our total uut[iut being of lighter character. AVe frequently have irou hot euough for stove-plate. Our heats vary from 15,000 to 24,000. SAMUEL E. HILLES, Samuel C. Tatum & Co.'s Works. Nov. 23, 1883. a.')!) AMKUICAN CUrOI,A l'ltA( TICK. PORTSMOUTH, O. TAPER CUPOLA. Outside (liamotor 72" Inside diameter at tuyeres "-"J" Lar;;est iiisido or melting-point diameter 40" Inside diameter at charginj^-door ."Vj" Height from bottora-plate up to bottom of cliarging-door ](/ Style of tuyeres : six ;V' X 4" oblong tuyeres. Height from bottom-plate to bottom of tuyere 'J^»" Height of tuyere above sand bottom on back side LtK' Fuel used for bed: coke . . 500 First charge of pig and scrap . GOO First charg(! of coke . . . -M Second charge of pig and scrap . cm Second charge of coke . . 30 Third charge of pig and scrap . GOO Third charge of coke . . . ;» Fourth charge of pig and scrap . coo lbs. Four more charges, continued per Fourth charge of coke . , Fifth charge of pig and scrap , Fifth charge of coke . , Sixth charge of pig and scrap , Sixtli charge of coke . . Seventh charge of pig and scrap . Seventh charge of coke Eighth charge of pig and scrap . order shown. 30 lbs. GOO GOO 30 COO 30 GOO No. 4 Root's blower; diameter main blast-pipe, 12". First appearance of lluid Time of starting fire . . 2.00 p.m. " charging first iron, 3.."i0 " Blast put on 4.00 " iron 4.10 p.m. Bottom dropped .... 5.05 " Revolutions of blower, 120. Pressure of blast, 10 ounces. Kind of fuel used, Connellsville coke. Amount of iron melted, 7,200 lbs. Amount of fuel consumed, 830 " Ratio of fuel to iron used, 1 to i^xoo- Fluidity of melted iron, XX. Length of heat, Ih. 5m. Rem.arks. — This cupola is old style, drawn in at the bottom to save fuel. We use very little scrap, as it is siarce. We pour our iron iuto moulds for heavy maehinery and rolling-mill castings. THOMAS L. WniTE, Foreman Portsmouth Foiaulri/ and MaL-hinc-Works Foundry. Dec. 12, 18S3. AMERICAN CUrOLA PRACTICE. 351 AKRON, O. COMMON 38," CUrOLA. Outside diameter 50" Thickness of lining 7" Inside diameter at tuyeres 38" Largest inside or melting-point diameter 38" Inside diameter at cbarging-door 36" Height from bottom plate up to bottom of charging-door 9' Style of tuyeres : seven 5" round tuyeres. Height of tuyere above saud bottom t Fuel used for bed •. coke . 700 lbs. )n back side . . 9' Fifth charge of scrap . . 540 lbs First charge of pig . . l,f)'25 " " " coke . . . 150 " " " scrap . 1)00 " Sixth charge of pig . . . 750 " " " coke . . 150 " " " scrap . . 500 " Second charge of pig . 915 " " " coke . . 150 " " " scrap 500 " Seventh charge of pig . . GOO " " " coke . 150 " " " scrap . 500 " Third charge of pig . . 875 " " " coke . 150 " " " scrap . 500 " Eighth charge of pig . . 625 " " " coke . 150 " " " scrap . . 540 " Fourth charge of pig . 700 " " " coke . . 150 " " " scrap 540 " Ninth charge of pig . . . 650 " 180 " " " scrap . , 550 " Fifth charge of pig . . 800 " No. 5 Sturtevant fan ; diameter main blast-pipe, 12". Time of starting fire . . 3.00 p.m. " charging first iron, 3.45 " Bla.st put on 4.15 " First appearance of fluid iron 4.30 p.m. Bottom dropped .... 6.00 " Revolutions of blower, 3,000. Pressure of blast, 13 ounces. Kind of flux used, limestone. Amount of iron melted, 12,610 lbs. Amount of fuel consumed, IjOoO " Ratio of fuel to iron used, 1 to G/'u^u- Fluidity of melted iron, XXX. Length of heat, Ih. 45m. Remarks. — Our iron is used chielly for making engines and heavy machinery-castings. ADAM FRANCE, Foreman Webster, Camp, .>:i AMKincAN crroLA riiA( tick. YOUNGSTOWN, O. COMMON IS" crroLA. Oiitsido (lianiotrr W Tliiikiu'ss of linii)}; ('." Iiisiiln (liaiiictrr at tuyoros 44" l^arRf'st inside or mcliinp-point diaiurfcr 4s" Iiisidt; (liaiiH'tcr at cliar^iiii^j-door 4H" Height from Ixittoni platf! lip t(j bottom of charging-dtKir 11' Styl(^ of tuyert's : six 4" round tuyc^ros. Ileifjlit from bottom plato to bottom of tuyere 21" Height of tuyere above sand bottom on l)ack sidf Ki" Ileight from bottom plate to bottom of slag-hole IS" The blast-pipe is connected to a wind-belt 10"xl2"; the belt encircles tlie cupola, with the exception of about 24" in front at the spout. Fuel used for bed : coke . 1,500 lbs. First charge of pig . . . 4,.")00 " " " scrap . . 5<)0 " coke. . . ?m " Second charge of pig . . 2,000 " " " scrap . 2,200 " coke . . 300 " Third charge of pig . . . 2,000 " " " scrap . . 2,200 " " " coke . . oOO " Fourth charge of pig . . 2,000 " scrap . 2,2(K) " coke . . 'Am " Fiftli charge of pig . . . 2,000 " No. 7 Sturtcvaut fan; diameter of main blast-pipe, 12". Fifth charge of scrap . 2,200 lbs " " coke . . vm " Sixth charge of pig . . 2,000 " " " scrap . 2,000 " " " coke 200 " Seventh charge of pig . 2,000 " " " scrai ) . 2,000 " " " coke . 200 " Eighth charge of pig . 2,000 " " " scrap . 2,000 " " " coke . 200 " Ninth charge of pig . . 2,0(X) " " " scrap . 2,000 " Time of starting fire . . 12.00 A..M. " charging first iron, 2.00 p.m. Blast put on 3.30 " First appearance of Uuid iron 3.4.') p.m. Bottom dropped .... G.JO " Revolutions of blower, 3,000. Kind of flux used, limestone. Fluidity of melted iron, XX. Length of heat, oh. 20iu. Amount of iron melted, 37,800 lbs. Amount of fuel consumed, 3,(>00 " Ilatio of fuel to iron used, 1 to lOJ. Remarks. — Our work is heavy machinery-castings. When required, ■we have another cupola, .H" inside diameter, to helj) us out in very heavy heats. The small cui)ola is built upon about the same principle as the above, and both have always worked satisfactt)rily. WILLIAM NOLL, Foreman Uamilton's Works Foundry. Oct. 24, 1883. AMERICAN CUrOLA PRACTICE. 353 LANSING, MICH. COMMON 29" CUPOLA. Outside diameter 48" Thickness of lining . . 9^" Insitio iliainetcr at tuyeres 29" Largest inside or melting-point diameter 29" Inside diameter at eharging-door 29" Height from bottom plate up to bottom of eharging-door .... 8' 6" Style of tuyeres : three 4" X 9" oblong tuyeres. Height from bottom plate to bottom of tuyere 18" Height of tuyere above sand bottom on back side 11" Three 6" branch-pipes carry the blast from the main pipe to the cupola's tuyeres. Fuel used for bed: coke . IfiS lbs. coal . 200 " First charge of iron . . . 2,000 " " " coke . . 1G8 " Second charge of iron . . 1,000 lbs. " " coke . . (J4 " Third charge of iron . . 1,000 " No. 4 Sturtevant fan; diameter main blast-pipe, 8"; cupola 18' from blower. First appearance of fluid Time of starting fire . . 2.20 p.m. " charging first iron, 2.58 " Blast put on . ... . . 3.25 " iron 3.35 p.m. Bottom drojjped .... 4.41 " Revolutions of blower, 3,000. Amount of iron melted, 4,000 lbs. Amount of fuel consumed, 600 " Ratio of fuel to iron used, 1 to Gtoo- Fluidity of melted iron, XXX. Length of heat, Ih. ICm. Remarks. — Our iron is used for engine, saw-mill, and jobbing castings. Feb. 26, 1884. JAMES CROWNER, Foreman Jarvis, Barnes, & Co.'s Works Foundry. aoi A.Mi;i:i(AN cri'oi.A I'Iiaitick. INDIANAPOLIS, IND. COMMON :k;" ( ti'ola. Ontsifle (linmotor 5.;" Tliickiifss of liniiiR. . hj" iiiKKiifss oi lining. . I. Inside iliiiriK'tcr :it tuvorc I^iirp'st in.sitle or inrltiiifx-point diameter 3(1" Inside diameter at cliar;:iiif;-door X>" llei}:ht from bottom j)late up to bottom of tliar;;iiif^-d<)or 'y.i" Style of tuyeres : two .S" round tuyeres. IIeit;ht from bt)ttom ])late to botttun of tuyere Hi" Ileij^ht of tuyere above sand Ixtttoiu on back side 12" Two 8" branch i^ipes lead direct from the main pipe to the tuyores. Fuel used for bed : coke First charge of i)ig . . " " scrap " " coke Second charge of pig . " " scrap " " coke . Third charge of pig . . " " scrap . " " coke . Fourth charge of pig . " " scrap Thirteen more charges, continued per order shown. No. 8 Sturtevant fan; diameter main blast-pipe, li"; cupola 50' from blower. 050 lbs. Fourth charge of coke . . 12.'> lbs 700 " Fifth cliarge of pig . . . 700 " .S(K) " " " scrap . . 300 " 125 " " " coke . . . 12.J " 700 " Sixth charge of pig . . . 7tK) " .•MX) " " " scrap . . ;i00 " 125 " " " coke , . . 125 " 700 " Seventh charge of pig , . 700 " 300 " " " scrap . 300 " 125 " " " coke 125 " 700 " Eighth charge of pig . . 700 " 300 " " " scrap 300 " Time of starting fire . . 2.30 p.m. " charging first iron, 3.15 " Blast put on 4.00 " First appearance of duid iron 4.05 f.M. Bottom dropped .... 5.30 " Revolutions of blower, 2,000. Pressure of blast, strong. Amount of iron melted, 21.000 lbs. Amount of fuel consumed, 3,550 " llatio of fuel to iron used, 1 to Oi'Vu- Fluidity of melted iron, XXX. Length of heat, lb. ;K)m. Remarks. — The abo\e was an average heat during the busy season. Our castings arc for architectural work. CTIRIS. BAKER, Fuixinan Ilainjii, Ivi IcIkiiu, it (.'o.'s Wurks Fmindri/. Ai-KiL 1-J, 1884. AMERICAN CUrOLA PRACTICE. 355 CHICAGO, ILL. MACKENZIE «!" X 42" CUPOLA. Outside dimensions Thickness of lining Inside diameter at tuyeres Largest inside or melting-point dimensions Inside dimensions at cbarging-door . . , Height from bottom plate up to bottom of charging-door Style of tuyeres : fiat 1^" opening, continuous tuyere. Height from bottom plate to bottom of tuyere . . . , Height of tuyere above sand bottom on back side • . , Fuel used for bed : coke . GOO coal . 400 First charge of pig . . 2,500 " " scrap . . 2,500 " " coal . . . 200 " " coke . . . 200 Second charge of pig . 2,500 " " scrap . 2,500 " " coal . . 200 " " coke . 200 Third charge of pig . . . 2,500 " " scrap . . 2,500 " " coal . . 200 " " coke . 200 lbs. Fourth charge of pig " " scrap " " coal " " coke Fifth charge of pig . " " scrap " " coal . coke. Sixth charge of pig . " " ~ scrap " " coal. " " coke Seventh charge of scrap No. 6 Root's blower; diameter main blast-pipe, 14". 78" X 54" 6" CO" X ;5G" Gli" X 42" m" X 42" " Inside (liiUiichT iit tiiycn-.s iK)" Largest iiisiilc or lui-ltiiij^-poiiit diaiiK'tcr 'M" Inside diauieti-r at cliar;i;in}i-d(»ir ■'!'>" Height from bottom plate up to )i(itt(ini nf i-liar;,'in;^-doi)r '.»' Style of tuyeres : three (!" round tuyeres. Height from bottom plate to bottom of tuyere I'J" Height of tuyere above sand liottom on back side 9" Throe 0" branch pijies leail direct from main pipe to tlie cupola's tuyeres. Fuel used for bed : coal . COO lbs. First charge of iron . . . 1,500 " " " coke. . . 200 " Second charge of iron . . 1,000 " " " coke . . 100 " Third charge of iron . . 5(X) " " coke . . IfX) " Fourth charge of irou . . 500 " Fourth charge of eokcr . Fifth charge of iron . . " " coke Sixth charge of iron . . " " coke Seventh charge of iron " " coke Eightli charge of irou . KM) Dis. 500 " 100 " 5(W " 100 " 500 " 100 •' 2,000 " No. 5 Sturtevant fan; diameter of main blast-pipe, 8"; length, 25' Time of starting fire . . " charging first iron, Blast put ou 1.30 P.M. 2.45 " 3.25 " First appearance of fluid iron 3.."/) i-.m. Bottom driip]icd .... -l.'M " Revolutions of blower, flux used, limestone. 2,500. Pressure of blast, 10 ouuces. Kind of Amount of iron molted, 7,000 lbs. Amount of fuel consumed, 1,400 " Katio of fuel to irou used, 1 to 5. Fluidity of molted iron, XXX. Length of heat, Ih. 5m. TlKMAnKS. — Our work is very light, and lioure we require very hot irou. Our castings are for agricultural jiurposes. DAVID SPENCE, Forcmun G. W. Brown & Co.'s Works Foinidri/. Apnii. 3, 1884. AMERICAN CUrOLA PRACTICE. 357 BELOIT, WIS. COMMON 40" CUPOLA. Outside diameter 55" Tliickness of lining ; 7^' Inside diameter at tuyeres 40" Largest inside or melting-point diameter 40" Inside diameter at charging-door . o 40" Height from bottom plate up to bottom of charging-door 8' Style of tuyeres : four 7" round tuyeres. Height from bottom plate to bottom of tuyere 12" Height of tuvere above sand bottom on back side 8" Fuel used for bed : coke 300 lbs coal 300 " First charge of iron . . 2,400 " " " coke . . 120 " Second charge of iron . 1,200 " " " coke . 120 " Third charge of iron . 1,200 " " " coke . 120 " Fourth charge of iron . 1,200 " Fourth charge of coke . . 120 lbs. Fifth charge of iron . . . 1,200 " " " coke . . 120 " Sixth charge of iron . . . 1,200 " " " coke . . 120 " Seventh charge of iron . , 1,200 " coke . 120 " Eighth charge of iron . . 1,200 " No. 7 Sturtevant fan; diameter main blast-pipe, 12". Time of starting fire . . 1.30 p.m. " charging first iron, 3.00 " Blast put on 4.00 " First appearance of fluid iron 4.10 p.m. Bottom dropped .... 5.40 " Revolutions of blower, 2,400. Kind of flux used, fluor spar. TOTALS. Amount of iron melted, 10,800 lbs. I Fluidity of melted iron, XXX. Amount of fuel consumed, 1,440 " Length of heat, Ih. 40m. Ratio of fuel to iron used, 1 to 7-pu. \ Remarks. — The class of work made is paper machinery and jobbing ca.stings. The blast-pipe connects to a wind-belt 6" X 12", which encircles three-quarters of the cupola's circumferences. J. E. PARKER, Foreman Merrill & Houslin Works Foundry. Oct. 25, 1883. 358 AMi;in.')" Ijiir;,'c.st iiisiilc or iiicltinj^-iHiint diaiiiitcr •">■'" Inside diameter at fliarj,'in{;-duiir •"-'>" Heij^lit from liottoiu jilati; up to li'itiom of eliarjiiii^i-flDor 7' 4" Style of tiiyiTcs : four tuyeres, '<^" diameter at inside of liniii;;?, ami (>" diameter at shell. IIei;,'ht from hottoiii plate to bottom of tuyere 10" Heij,'lit of tuyere above sand bottom on baek side I'J" llei-^ht from bottom plate to bottom of slag-hole '.>" Fuel used for bed: coke . 450 lbs. First charge of pig . . . 1,200 " serap . . 1,200 " coke. . . m " Second charge of pig . . 300 " " " scrap. . 300 " coke . . 50 " Tliiril charge of pig . . . 300 " " " scrap . . 300 " " " coke . . 50 " Fourtli charge of pig . . 300 " " " scrap . . 300 " " " coke . . 50 " No. 5 Sturtevant fan; diameter main blast-pipe, 9'' Fifth charge of pig . . vm lbs " " scrap . . 300 " " " coke . .•50 " Sixth charge of jiig . . . .TOO •' " " scrap . , 3fl0 " " " coke . .50 " Seventh charge of pig . . 300 " " " scrap . 300 " " " coke 50 " Eighth charge of pig . :joo " " " scrap . 300 " " " cuke . 50 " Ninth charge of scrap . . 1,400 " Time of starting fire . . . 3.10 p.m. " charging first iron, 4.15 " Blast put on 4.40 " First appearance of liuid iron 4.47 p.m. Bottom dropped .... 5.40 " Revolution of blnwcr, 2,s00. Kind of Hux used, tiuor spar. After the first charge, then 7 pounds to every charge was used. TOTALS. Fluidity of melted iron, XX. Length of heat, 1 hour. Amount of iron melted, 8,000 lbs. Amount of fuel consumed, 850 " llatio of fuel to iron used, 1 to i),^,. Ri:m,vrks. — We have made quicker time than the above, but that shown is an average. What scrap we use, aside from our gates, etc., is of the best quality. The last charge, of 1,400 lbs. is mostly all scrap for sash- weights. Our general work is mill-machinery and steam-engines. P. L. SIMI'.SON, Foreman Norlh .S7ar Iron Works Foundry. Oct. 31, 18S3. AMERICAN CUrOLA PRACTICE. 359 BURLINGTON, IOWA. COMMON '25" CUPOLA. Outside diameter 40" Tliickiicss of lining S" Inside diameter at tuyeres 2.")" Largest inside or meltinj^-point diameter 2(i" Inside diann^ter at eharging-tloor 24" Height from bottom plate up to bottom of charglng-door 10' Style of tuj'eres : two 5" round tuyeres. Height from bottom i>late to bottom of tuyere 12" Height of tuyere above sand bottom on back side (J" Fuel used for bed : coke . 400 lbs. First charge of pig . . . 300 " " " scrap . . 300 " " " coke. . . 100 " Second charge of pig . . 300 lbs. scrap . GOO " " " coke . . 100 " Third charge of scrap . . 800 " No. 5 Sturtevant fan; diameter of main blast-pipe, 10"; length, 31', Time of starting fire . . 2.30 p.m. " charging first iron, 4.00 " Blast put on 4.:j0 " First appearance of fluid iron 4.45 p.m. Bottom dropped .... 5.45 " Revolutions of blower, 1,400. Kind of fuel used, Connellsville coke. Amount of iron melted, 2,300 lbs. Amount of fuel consumed, 000 " Ratio of fuel to iron used, 1 to 3nfV- Fluidity of melted iron, XXX. Length of heat, Ih. 15m. Remarks. — The class of work made is small castings and general machinery. Most of our work requires metal at white heat. The blast was not put on as strong as it could have beciu had we been able to take care of the iron. The heat bciing small does not, of course, show the economy it would were the heat a larger one. W. L. SCHTJCK, Foreman Heimlcn & ISchuck Works Foundry. Mat 19, 1884. 360 AMKUICAN (Tl'fjLA I'UACTICK. GRINNELL, IOWA. CnMMdN :;i" ci |-(»|, \. Outside (liamofor 4.s" Tlii<-kn<'ss of lining; 7" Insidt- (liaiintcr at tny,200. Kind of fuel used, Lehigh and Counells- ville coke. TOTALS. Amount of iron melted, Cifir>0 lbs. I Ratio of fuel to iron used, 1 to 7^^. Amount of fuel consumed, 8;?5 " I Length of heat, 4(3 m. Remarks. — I am running a cupola that we put up sis months a^o. The most fuel I ever used Wiis 1 to 7, and I am now meltint; with 1 to 8. I believe in using all the fuel required to melt good iron, but I do not believe in waxting it. The tirst fifteen hundred pounds of iron i.s run into mower wheels. These wheels have wrought iron spokes in them, and the rims have to be poured first, to give lliem a chance to shiink before the hul) is ixinred. We have a light seat, and also a light gc-ar cover, and several other pieces that liike hot iron, and we have no trouble in running them. The tuyeres in my cupola are 14" from biise. I have used them ae low uo 12", but for coke 1 prefer to have them higher. In some foundries they use tire-clay, weakened with sand. I use common clay mixed with the burned sand that comes from the castings. It is hard to mix, but makes a good lining. The sand preveuLs the clay from cracking, and it stands fire equal to lire-clay. J.\COB OTT, Foreman Cravcr, Steele, & Auslln's Agricultural Works Foundry. Fuel used for l)ed: coke . mi lbs. coal . - 200 " First charge of pig . . . 7(X) " " " scrap . . 300 " " " coke. . . (iO " Second charge of pig . . (;.-.o " " " scrap . . ;;oo " " " coke . . ."m " Third charge of pig . . . or.o " " " scrap . . ;kx) " " " coke . . .w " No. 5 Sturtevant fan ; diameter Time of .starting fire . . .S.25 P.M. " charging first iron, 4.00 " Blast put on 4..!0 " June 3, 1884 AMERICAN CUrOTA TRACTICE. 361 OMAHA, NEB. CAR-WHEEL DEPARTMENT : COMMON 50" CUPOLA. Outside diameter 024" Thickness of lining (>" Inside diameter at tuyeres 50" Largest inside or melting-point diameter 52" Inside diameter at charging-door 50" Height from bottom iilate up to l><)ttom of chargiug-door .... 10' 2" This cupola has six 3.^" x 8^" oblong tuyeres. Height from bottom plate to bottom of tuyeres, ISg". The hottest melting-point is aljout 18" above the top of the tuyeres. "Main blast-pipe to branches is 200' long. There are two branch-pipes, 16" diameter, IS' long. MACHINERY DEPARTMENT: COMMON 50" CUPOLA. The cupola in this department is the same size as the cupola in the ■wheel department, with the exception of the tuyeres. This cupola has three rows of tuyeres, all of which are 4" in diameter; there are six in each row. The respective; distance of each row from the bottom plate is 1.5", 24", and 34". The two upjier rows of tuyeres are at an angle of 40°, so as to throw the blast downwards. The hottest melting-point is about 14" above the top row. The lining between the two lower rows is also burned out a little from the effects of the blast. From the fan to the branch-]iipes, the main pipe is 68'. The length of the two 18" brant-h-pipes leading to the cupola is 22'. The diameter of maia pipe is 24". It has a No. 8 double Sturtevant blower, making 1,700 revolu- tions. The wind-belt surrounding the cupola is 30" deep by Og" wide. Cak-Wheel Cupola. lbs. Fuel used for bed : coke 1,200 First charge of pig 2,000 " " wheel-scrap .... 2,480 " " coke 215 Second charge of pig 1,000 " " wheel-scrap . . . 1,240 " " coke 21a Third charge of pig 1,000 " " wheel-scrap . . . 1,240 Fourteen more charges, coutiuued in the order shown. TIME. Time of starting fire .... 11.00 a.m. " charging first iron . . 11.5.5 " Blast put on 12.15 p.m. First appearance of fluid Iron . 12.22 " Bottom dropped 2.50 " TOTALS. Amount of iron melted . . . 40, .".20 lbs. Amount of fuel consumed . . 4,040 " Italio of fuel to iroji used, 1 to 8.G9. Length of heat, 2h. 35m. Machinery Cupola. lbs. Fuel used for bed : coke 1,200 First charge of scrap 5,000 " " coke 210 Second charge of scrap 2,000 coke 210 Third charge of scrap 2,000 " "^ coke 210 Fourth charge of scrap 2,000 Fifteen more charges, continued in the order shown. TIME. Time of starting fire .... 2.00 p.m. " charging first iron . . 3.00 " Blast put on 3.20 " First api)earance of fluid iron . 3.27 " Bottom dropped 5.40 '♦ TOTALS. Amount of iron melted . . . 41,000 lbs. .Vmount of fuel consumed . . 4,980 " Uatio of fuel to iron used, 1 to 8.23. Length of heal, 2h. 20m. Remarks. — We h.ive melted as high as 1 to OJ (with Connellsvillo coke for fuel) in the wheel furnace. Our iron is when melted very hot and fluid. 1 find very little differ- ence in the two cupolas, with the heats we are running, e.xcept the machinery cupola melts the fastest at the end of heat. Were the two cupolas run above the general capa- city of such sized furnaces, then the cupola with the three rows of tuyeres would pro- duce the hottest iron, and perform the fastest melting, if they were both charged exactly alike. EDWARD RICHELIEU, Foreman Union Pacific Itailwaij Foundry. April 4, 1884. 802 AMKKK AN rtl'fil.A I'KACTK'K. DENVER, COL. COMMON :•.-•" ni'oF.A. Outside (liaiiu'tcr Tliitkiiitss of liiiiii;; Inside fliaini'lrr at tiiyrrcs J^arjicsl iiisidi! or iiicltiii;4-ii<)iiit^ diaiiH't Height from bottom jdatc to bottom t r slag-hole . . IS' Fuel used for bed : coke 1,500 11)3 Fourth charge fif coke . . 20») lbs First charge of Fifth charge of pig and scrap 7,000 " pig and scrap . 3,000 " " " coke . . . 200 " " " coke . . . . 200 " Second charge of Sixth charge of pig and scrap 3,000 " pig and scrap . 3,0f»0 " " " coke . . . liOO " " " coke . . . • '_'(;<; AMi;i;iC.\N Ctl'Ol.A l'HA< I ICK RICHMOND, VA. COLLIAU 40" CUrOLA. Ontsido (lifiinetor TliirkiH-ss uf lining Iiisidf di:iiiu!tcr Jit tuynros Larf^cst inside or meltinjJi-poiiit diameter Insidt! (liaiiieter at cliarjiiiiji-door lleij,dit from bottom iilate up to bottom of rharpnnR-doMr .... 11 Style of tuyeres : two rows of tuyeres, six al)ov(; aiioo " 40" 40" 40" ' (;" 1,000 lbs. 170 " 500 " 2,000 " 170 " 500 " 3,000 " No 5\ Baker blower; diameter main bla.st-pipe, 12". Time of starting fire . . 12.00 a.m. " charging first iron, 2.00 r.M. Blast i)ut on ..... 3.15 " First appearance of fluid iron 3..'V) r.M. Bottom dropped • . . . 4.45 " Revolution of blower, IHj to 100. Pressure of blast, 7 ounces. Kind of fuel used, West Virginia coke. Kind of llux used, scraps of marble, 40 pounds to each charge. Amount of iron melted, 15,000 lbs- Ajuount of fuel consumed, 1,480 " Batio of fuel to iron used, 1 to lOi'J'y. Fluidity of iron melted, XX. Length of heat, Ih. 'Mm. Rkm.msks. — Our last iron is hotter than the fir.st. The coke used was rather soft and mashy. The castings made are for engines and saw-mills. L. FOX, Foreman Tuiincr and Dduncif Ewjinc Co.'s Works Foiimlnj. Xov. 3, 1883. AMERICAN CUPOLA TKACTICE. 367 SALEM, N.C. COMMON 26" CUPOLA. Outside diameter 42" Thickness of lining 8" Inside diameter at tuyeres 2(3" Lariiest inside or melting-point diameter 2G" Inside diameter at charging-door 22" Height from bottom plate up to bottom of charging-door 7' 6" Style of tuyeres : flat, 2" opening, continuous tuyere. Height from bottom plate to bottom of tuyere 12J" Height of tuyere above saud bottom on back side 4" Fuel used for bed : coal . . 400 lbs. First charge of iron .... 500 " " " coal .... 50 " Second charge of iron . . . 500 " Second charge of coal Third charge of iron " " coal Fourth charge of iron . 50 lbs, .500 " . 50 " .500 " Ten more charges, continued per order shown. No. 4 Sturtevant fan; diameter main blast-pipe, 8". Fau within 8' of cupola. Time of starting fire . , 1.00 p.m. " charging first iron, 2.00 " Blast put on 2.15 " First appearance of fluid iron 2.20 p.m. Bottom dropped .... 3.50 " Revolutions of blower, 3,000. Kind of fuel used, Lehigh anthracite (egg). Amount of iron melted, 7,000 lbs. Amount of fuel consumed, 1,050 " Ratio of fuel to iron used, 1 to 6/'u'o. Fluidity of melted iron, XXX. Length of heat, Ih. 35ui. Remarks. — We use Low Moor and Longdale, Va., iron. The Low Moor is very refractory to melt. Our work is saw-mill and general machinery castings. E. BABINGTON, Furanau Halan Jrou Works Foundry. Oct. 1-2, 1883. 308 AMI.IilCAN CII'OI.A rUAClKJE. NASHVILLE, TENN. COMMON r,C," CUPOLA. Ontsido (lisimetor Tliiikticss of lining: Inside iliaiiu'tcr at tnycrt's DiainiMor lU" above tlie centre of tlic tuyeres Largest inside or melt inp;-i>oint diameter Inside diameter at cliarj;infi;-door Height from bottom plate up to bottom of eliargmt; i\" w Fuel used for l)cd: coke . l,.">00 1bs. Fourth cliargc of coke . 2."2 First charge of pig . . . 2,0()0 " Fifth charge of pig . . . 1,4(W " " scrap . l,Of)0 " " " scraji . , 800 " " coke . . . 2.-.2 " " " coke . 252 Second charge of pig . . 1,300 " Sixth charge of pig . . . 1,400 " " scrap . . 700 " " " scrap . . 800 " " coke . . 252 " " " coke . 252 Third charge of pig . . . 1,400 " Seventh charge of pig . . 1,400 " " scrap . . 800 " " " scrap , 800 " " coke . . 252 " " " coke . 252 Fourth (iiarge of pig . 1.400 " Eighth charge of pig . . 1,400 " " scrap. . 800 " " " scrap. . 800 12" S" two lbs. Eight more charges, continued per order shown. No. 5 Root's blower; diameter main blast-pipe, IS". Time of starting fire . . 12.00 a.m. " charging first iron, 1..30 p.m. Blast put on .3.00 " First appearance of fluid iron .3.00 Bottom dropped .... 5.;50 Revolutions of blower, 150. Kind of fuel used, Alabama coke. TOT.4LS. Fluidity of melted iron, XXX. I-ength of heat, 2h. 30ni. Amount of iron melted, .35..'W 11 is. Amount of fuel consumed, 5,080 " llatio of fuel to iron used, 1 to Tjxas- Bem.akks. — The l)lower does not run as fast as it should to do its best work. We melt fnmi 7.j to S tons jx-r hour. There is abnut two thousand ptuinds <(f metal left in the cupula when the blast is stopped, which stands iifteen In twenty minutes until it can be poured. It has to be poun'tl into iiiiiuMs that recpiire dull iron, and is bandleil by few men; hence the delay. 'J'lie eastiiii^s we uiakt! are stoves, manttds, and hollow-ware, therefore our iruu mu.-5t be very hot. CHAKI.ES PRESTON, Forc7nan l'hiUij>s it Baltorff :>(ove Works Foundry. Feb. 27, 1884. AMERICAN CUrOLA TRACTICE. 369 CHATTANOOGA, TENN. (^OMMON 28" CUPOLA. Outside diameter 40" Tliickness of lining 6" Inside diameter at tuyeres 28" Largest inside or nieltinpi-poiiit. diameter 28" Inside diameter at charging-door 20" Height from bottom plate up to bottom of charging-door V &' Style of tuyeres : two 3^" X 7" oval tuyeres. Height from bottom plate to bottom of tuyere 15" Height of tuyere above sand bottom on back side 12" Two 6" branch pipes carry the blast frona main pipe to the cupola tuyeres. Fuel used for bed : coke . 400 lbs. Fourth charge of coke . 50 lbs First charge of pig . . . GOO " Fifth charge of pig . . 600 " " " scrap . . 200 " " " scrap . 200 " " " coke . . . 50 " " " coke 50 " Second charge of pig . . 600 " Sixth charge of pig . . 600 " " '* scrap 200 " " " scrap . 200 " " " coke . . 50 " " " coke 150 " Third charge of pig . . . 600 " Seventh charge of pig . 600 " " " scrap . . 200 " " " scrap 200 " " " coke . . 50 " " " coke 150 " Fourth charge of pig . . 600 " Eighth charge of pig . . tiOO " " " scrap . . 200 " " " scrap. 200 " No. 1 Root blower; diameter main blast-pipe, 8". Time of starting fire . . 2.00 p.m. " charging first iron, 2.45 " Blast put on 4.00 " Revolutions of blower, 600. First aiipearance of fluid iron 4.05 p.m. Bottom dropped .... 5.40 " Kind of flux used, limestone. Fluidity of melted iron, XXX. Length of heat, Ih. 40m. Amount of iron melted, 6,400 lbs. Amount of fuel consumed, 950 " Ratio of fuel to iron used, 1 to Q^io. Rkm.akks. — The blower is too small for our work, and has to run too fast. W'ln'ii (jur heats are heavier than 4,800 jiounds, we make the sixth charge of fuol 150 pounds instead of 50. We find that we cannot make good fluid iron with certainty ev(!ry heat with much less coke than 1 to Ty-. We have melted as high as 1 to !), and quite freiiuently melt 6,000 pounds by ha\ing4()0 pounds coke on bed and .50 pounds for all the charges; but we prefer to use a little more coke, as it makes more certainty of ol)taining economy in the end. Our iron is used for making engines, turbine-wheels, and mill-castings. G. W. WHEELAND, Proprietor, W. S. BURGER, Foreman, j^tna Foundry and Machine, Works. Maucu 18, 1884. :u A.MKKKAN ( I r" Inside diaiiicttr at tuyeres 2.S" Lart^cst insido or iii(ltiii;;-I><>int diameter .■<()" Insider diaiiiftcr at eliarj;in;i-door L'H" ll(ij;lit from bottom plate up to bottom of charging- " *' " scrap 300 " " " scrap . 500 " " " coke . . 75 " " " coke 7.5 " Third charge of pig . . . 200 " Sixth charge of pig . . 200 " " " scrap . . 500 " " " scrap . 500 " No. 3 Root blower; diameter main blast-pipe, \1 Time of starting fire . 2..".0 r.M. I First appearance of Ihiid " charging first iron, 4.00 " Blast put on 4.iX) " 4.45 r.M, Bottom dropped .... 5.30 " Amount of iron melted, 4,200 lbs. Amount of fuel consumed, 725 " Ratio of fuel to iron used, 1 to S/Jj. Fluidity of melted iron, XXX. Length of heat, Ih. Remarks. — The above is not as good a sliowing as wo can make. I take it as an average of heats run last s]>ring, of wliidi we kept a record of liltcuu beats. Our iron is used for geneial jobbing castings. Nov. c, 18S3. R. I. jMEALOR, Foreman Montyomcnj Iron Vo.'s Works Foundry. AMERICAN CUrOLA rRACTICE. 371 COLUMBUS, GA. COMMON 30" CUPOLA. Ontsido (lianiotor 42" Thickness of lining (>" Inside (liiinieter at tuycros 30" Larjiest inside or melting-point diameter 30" Inside diameter at charging-door 30" Height from bottom plate up to bottom of charging-door 8' Style of tuyeres : flat ^" opening, continuous tuyere. Height from bottom plate to bottom of tuyere 11" Height of tuyere above sand bottom on back side 7" Connected with the ^" opening tuyere is an air-chamber, 8" X 2^", inside the cupola shell. The blast is carried to this l)y means of one branch pipe, 4" X 8" where it connects with the chamber, and 8" X 10" where it joins the main blast-pipe. Fuel used for bed : coke 175 lbs. Fifth charge: scrap . . 300 coal 400 " " " coke . . 75 First charge: pig . . . 1,800 " Sixth charge: pig . . 300 coke . . 75 " " " scrap . . 300 Second charge : pig . • 300 " " " coke . . 70 " " scrap . 300 " Seventb charge : jiig 300 " " coke 75 " " " scrap. 300 Third charge: jiig . . 300 " " " coke . GO " " scrap. . 300 " Eighth charge : pig . . 300 " " coke . . 75 " " " scrap . 300 Fourth charge : pig . . 300 " " " coke (iO " " scrap . 300 " Ninth charge: pig . . 300 " " coke . . 100 " " " scrap. . 300 Fifth charge: pig . . . 300 " Six more cliarges, continued per order shown in last two charges, 48" shell, four-blade, home-made blower, main blast-pipe 12" x 12", Time of starting fire . . 1.30 r.M. " charging first iron, 3.45 " Blast put on 3..50 " First appearance of fluid iron 3..59 p.m. Bottom dropped .... 5.51 " Revolutions of blower, 1,600. Fluidity of melted iron, XXX. Length of heat, 21i. Im. Amount of iron melted, 10,200 lbs. Amount of fuel consumed, 1,525 " Ratio of fuel to iron ii.sed, 1 to (t-fjyij. Remarks. — One piece for an .amraoni.i ice machine in this he.it weighed .'i.SOO pounds, and had to be poured with clean, hot iron, in order to stand a test of 273 pounds press- ure. We cast every day, but never use our large cupola, 00" x 36", unless we have some one jtiece that takes over 7,000 pounds of metal to pour it. Our chief work is engines, saw-mill and cottou-machiuery castings. ROBT. E. MASTERS, Foreman Columbus Iron Works Foundry. Ai'KiL 10, 18S4. .)(li AMKlilCAN cri'Ol.A ritAClKK. PALATKA, FLA. COMMON -l-l" CII'OF.A. Outside diainoter .'V." Tliickiicss of lining 7" In.si(k' (liiinictiT at tuycn-s '.'J" Liir^icst in.si(l(! or intiltint^-poiiit dianit'tcr '_"_'" liisiili' (lianiottT at c'liarKin;i-" Height of tuyere above sand bottom on back side 15" Three 4" branch-pipes carry tlie blast from the main pipe to the cupola's tuyeres. Two of the branch-pipes arc 8' long and one 2' long. Fuel used for be (liniiiptcr 4.'." TliickiH'Ss of liniiirj ,1'." Iiisidi- (liaini'tt-r ill tiiycri's :n" Larp'St iiisidt- (ir iiu'ltiii;;-|i()iiif (liaiiiftcr "M" Iiisiilf (liamottT iit cliarjiiii^i-ilddr '.W Ilcit^hf from liottoin plati- ii|) to hottoiii of (•liarj;iii;;Hloor K/ Style of tuyeres : seven 4" round tuyeres. Height from liottoni plate to l)ottoiii of tny" Inside diameter at tuj'eres 22" Largest inside or melting-point diameter 24" Inside diameter at cliarging-door , . . . 23" Height from bottom plate up to bottom of charging-door „ , o , . 9' Stylo of tuyeres : four i" round tuj'eres. Height from bottom plate to bottom of tuyere o » . 12" Hoight of tuyere above saud bottom ou back side , „ 6" Fuel used for bed : coke 2r>0 lbs. First charge of pig . . 1,000 " " " coke. . 150 " Second charge: pig . . 500 " " " scrap . 500 " Time of starting fire . 2.00 P.M. " charging first iron 2.30 " Blast i>ut ou .... 3.30 " Second charge of coke . , 75 lbs. Third charge: scrap. • . 1,000" " " coke ... 50 " Fourth charge: scrap . . 500 " First ajipearauce of fluid iron 3.45 p.m. Bottom drojiped .... 4.30 " Revolutions of blower, 1,500. Kiud of fuel used, English coke. TOTALS. Amount of iron melted, 3,500 lbs Amount of fuel consumed, 525 " Ratio of fuel to iron used, 1 to Gyo%- Fluidity cf melted iron, XXX. Length of heat, Ih. Remarks. — The class of work made is stoves, hollow ware, and jobbing. The blower used is an old-fashioned wooden one, made by hand. The iron came down very hot. The pig used is Glengarnock Scotch. Feb. is, 1S34. JOHN MONTAG, Foreman Novelty Iron Works Foundnj. aid MMI.IINf; STKKI,. MKI/riNC STKKL IN AN (»l:l>IN\KV :;o" cC I'ol, A n>H. FutI for licil : rokc . roikl . KlrHt chnrKc : |>iK • " " t<(k(! . Bccuiid charge : \nu " " Hcni] " " coke Third charge : piir . " " fcrikp " " coke II.H. •J.HI •."(HI I/.'IHI fiO :iiH) iVI 50\k • " " Kcrap " " coke . Sixth ciiargc: pig . " " Hcrap " " coke . Seventh charge : pig 3cnip " " coke " " coal Ninth charge: Hle«-I . " '• coke . Tenth charge : cleel . lb*. , :iu) .'lO , :iuu . atw . luu , IOO . 1 ,a»o . 1(10 . 1,(J(W Time of xtartiiig lire . . . Time of cliargiiig tirct iron BluDt put un ■J.l'> I'.M. I Kirnl appeanmce of Ihiiil iron ;!.!:■> " HotUjiu ilropjted 4. -JO " •4 .•27 &.4d RevolutionH of blower wlien on nteel, l.iKXi. Kind of fuel u«cd, liirmlnghain coke and I.ehigh coal. The cupola in which we niell*'d thin heat in the one given on p. 'Ml; an the diiueuHionH of cupola can there be seen, it is not Hhowu with tbia re{Kirt. Amount of Mco\ molted " iron melted . . " fuel conijunied TOTALS. 2,2nn Ills. I Hatio of fuel to iron unod :'.,4U0 " Length of heal. . . . 1,100 " I . . 0.66 Ih. 2.0in. Remarks. — The " American Sfacbitiist" of .\i)g. 2;?, 1SS4, contained an account of my " Meltiiii; Steel in an Ordinary Cuitola." Since then, by experinientini:, we have learned sometbiiig of the nature of it, ancJ not only found the class of work we can use it in to best advantage, but have also made a decided improvement in manner of meltinu' and fluidity of metal. The metliod of charging is different from the account I gave in the aiticle referred to. We have not melted a heat of steel alone, not having occiision to melt more than 1,000 to •2,U0ii pounds at a time. \Vc continue to melt it right iK'hiiid the cast-iron portion of the heat, as above shown. As soon as tlie last charge of cjist-iron begins to settle away from the charging-door, we keej) the cupola fiiH of sti-el up to the charging-doors until the last has been put on : this gives it the benetit of a long heat, and when it reaches the melting point it comes down (to use the expression of a moulder here) "hot enough to run a needle with the point up." It is very fluid when it first comes from the cuitola. While it does not remain fiuid as long a.« cast-iron, I am satisfied a very large piece could be poured with it. I notice, by agitating it iu the ladle, it " gums up" around the ladle quicker than cast-iron. Charging in above manner, the cast-iron all melt.s down ahe.id of the steel. Then there is a cessation in the melting for a few minutes before the steel starts : once started, it melts very fast. The appearance of the metal is so different from cast-iron in the lluid suite that we can tell it as soon as it starts from the cu]>ola. The steel scrap used is of a class known as " slab, or agricultural steel ; " and we have melted (50,(1(10 out of the 7.5,000 pounds we had on hand, besides using un all the scrap tliat has been made since then. I>y itself, tlie steel runs ])orou8. I?y adding one-sixtu cast-iron to the charges, we find it runs the castings very close and solid, and hardi-'r than the steel alone. For furnace-liners, back-plates, grate-bars, brake-shoes, etc., it is superior and more serviceable than cast-iron. In light castings annealed, 1 feel sure it would make stronger castings than malleable irtni. Last f;ill, J. C. Albrecht, master marliinist of the railroad shops here, complained alxnit the chilled truck-wheels, shii)i)ed him here for section masters' handcars, cutting out and u'ettini; Hat places in them in a sIku^I time. We asked him to let us make a set of steel-rim wheels for a trial. lie i)laced an order with us for two sets of wheels, steel rim, 2(i" diameter, .'!.\" face, 3" thick, 1" llaniie, ten J" round wroutrht-iron spokes, set zigzag in hub: hub of cist-iron; weiijht of wheel, V20 iwunds. They have had to stand the test through the most severe winter we have had in the South for years, lluviug filled orders for loO since then, is evidence of the satisfaction they are gi\lug. The above heat was melted April '24, 18S5. UdRT. E. MASTERS, Foreman Columbus Iron Wvrks Founilry, Coluiuhus, Ga. MELTING AND MIXING STEEL WITH CAST-IRON. 377 MELTING AND MIXING STEEL WITH CAST- IRON TO OBTAIN STRONG OR CHILLED CASTINGS. As a supplement to the })ie\ioiis page, the author offers the following few notes, which the readers wall no doubt find inter- esting and of value. The union of steel with east-iron has of late j'ears been much practised for the purpose of either adding strength to or increasing the depth of chill to cast-iron, ideas and notes upon which will also be found in vol. i., pp. 272, 297, 298, and 299. It might be well to here state that wrought-iron has also been used in mixture with cast-iron, sometimes being melted in the cupola and again mixed in with the cast-iron after it was melted. I have heard of its being used as high as 33 per cent in mixture with cast-iron melted in a cupola. The greatest per cent of either steel or wrought-iron which can be mixed in with liquid cast-iron after it is melted, will, of course, depend on how " hot" the fluid cast-iron is, and what it is intended to be poured into. I would not have the reader understand b^' the above term, "greatest per cent," that the more steel or wrought-iron there is mixed in a ladle or cupola with cast-iron, the stronger should the product be. As far as strength is concerned, I would be led to say there is a limit, and that it greatly depends upon what grades of steel and cast- iron are mixed together. The cast-iron, in order to obtain the greatest strengtli in product of mixtui-e, will be greatly affected by the amount of carbon the steel and cast-iron contain. A soft or low carbon steel should produce a much stronger product than a hard or high carbon steel ; and I have no doulit but that. o7S MF.I.TINf; AND MIXINH STKKI, WII H fAST MION. from :i cMn'riil iiiixtiut' of loiv-Cdrhon stpcl v:Hh Idv-mrho'n rnat- irnii, iiiiiporlioiialcly .slioii;^; c';i.stiiiu;s c-oiild l»i; pioduci'd. Tlic ri'sult ohtiiiiic'd from :i iiiixtiirc of liigli-carlxjii sU'el with cast- iron can l>o .such :ls to iin[iair the original strength of the ( :l^t-il■oIl. Stcd, lus is ^Yc■ll known, conhdns less carbon tluin cast-iron, and mny-c than tcrom/Iit-iron, the latter sometimes containing )iut a trace, ('arl)on is lu-ld in cast-iron in a combined and in an iMicombined stale. "When coml)ined, it is chemically united with iron, as seen in hard or white cast-iron ; and when uncoiTl- liined. tin- carbon ap^icars in the foiin of graphite, a,s seen in No. 1 grailes of foundry soft gray iron. Ca-st-irou containing carbon in the graphite or uncombined state requires a higher temperature to melt it. than when it is chemically combined with the iron ; and the larger 2)er cent of chemicallj' combined carbon iron contains, the less heat is required to melt it. The more carbon there is in wrought-irou, steel, and hard cast-iron not only causes it to be melted easier, but also makes it retain its life or fluidity longer. Carbon can be given and taken awa}" from iron or steel. Fiicl will su2^phj it, and air eliminate it. AVhcu wrought-iron or steel is melted in a cupola, both of the above agencies are at work upon it ; and while we can in one sense say they are being weakened through oxidation, we can in another sense say they are also weakened through carbonization; for when steel, etc., is mixed in among fuel, and there melted, it cannot Init be affected by it, as the oxygen of the atmosphere comluning with fuel in a cupola creates carbonic acid and carbon oxide, which, when liberated in concert with other gases, — such as sulphur, etc. — which fuel contains, all go towards destroying the original strength of scrap-steel or wrought-iron scrap. "When we see, in the manufacture of steel, that the slightest jxr cent of a component can so materially change its nature, wiiat can we expect in the way of certainty in producing grades MELTINC. AND MIXING STEEL WITH CAST-IRON. 379 out of a, cupola, wIkm'c stocl is tuml)lcd in with n, conglomera- tion of cast-iron and fuel, of wiiosi; clieniical analysis we know nothing or have no control ? To procure a homogeneous product from the mixture of steel with cast-h'on, as a rule, seems to have been poorlj- accom- plished. The steel mixes with the cast-iron in such a manner, that, when castmgs arc turned or bored, hard spots or mottled surface often appear. In melting steel with cast-iron there are, however, a few things which can often be done in assisting to ol)tain a uniformity m i)erceutage of the material charged : as, for instance, did one desire castings made of one- fourth steel and three-fourths cast- iron, the material should be carefully weighed and charged ; and in charging the cupola, adopt the method set forth in vol. i. p. 304. The method there descril)ed will at least insure the production of the mixtuie as charged. Of course, if there were enough weight of the steel mixture to make a heat by itself, then the mode above referred to would not be necessary. Another point which might be well to mention in regard to obtixining a uniform mixture is, that the more metal there can be collected in a large ladle, and agitated by stirring with a "mixer" or wrought-iron rod, the better homogeneous castings will be produced. No one should expect, that, by catching and pouring the metal into small work as fast as it melts, the cast- ings produced can contain the uniformity m mixture they would where largo bodies of the metal are first collected before the pouring commences. Of course, in the case of large castings the metal would, through necessity, require to be collected in large bodies. For small castings the metal would, in being collected, require covering with dust, etc., in order to " hold its life ; " or, where it was to be made a stead }• business, a closed reservoir could be used ; the iron as it melted, running into it, could, after a body was collected, be taken out in " small taps " as required. There are of course many castings which will not .'1^0 Mil TINC AM) MIXINC SIKKI, WllU CAST IKON. 1)0 imicli iiijiind tlironixli irrc^xiilaiity in uiiiformity of mixtiiro. TIu' ahovi' jxiiiits ail! himply lo yica idms to iis.si.st tlioroiijili and C'cuKil mixing in cases wlioic fiiio work is required. In clinrginir sled mixc*! with ejust-iron, or nlonr. in .1 cniirtla. the .stool caniioL Iml lu' injund tin'ougii cnrhonization aiul oxida- tion. Were air-fnrnaees or cruciMes used (which I l»clieve could be made practical for the purpose) for melting steel, the above injuries steel receives would be greatly overcome. I sim- \>\y liore suggest ''air-furnace" and "crucible" for the piu-pose of presenting something that may be of value to those inclined to experiment with scrap-steel to the end of o])taining strong castings. Sanuicl i\I. Carpenter of Cleveland, O.. who holds K-lters- patent No. 173,1;J9, awarded him Feb. 8, ISTft. upon a prociss for the iuimersiou of steel into liquid cast-iron, claims that cast- iron, in order to be strengthened by a mixture with steel, can only be done by melting the steel immersed in liquid cast-iron, thereby preventing it from contact with the blast of air which oxidizes the steel and impairs its strength when melted in cu- polas. Upon this point I greatly agree with Mr. Carpenter ; for in my experience with steel melted with ca.st-iron in a euix)la, I cannot say I thought it as a general thing to add strength to cast-iron. Whenever I have used or seen steel melted with cast- iron in a cupola, it was generally for the purpose of hardening or giving a deep chill to castings. Fur this purpose^ steel mixed vjitli cast-iron is at least effective. To melt scrap-steel without mixture with cast-iron in an ordinary cupola, as creditably pcrfonncil liy l\()1)ert E. Mastei'S, seen on \). ."70, and described in "American Machinist," April 2"), 1885, has attracted great attention throughout the United States, and will no (loul)t be the cause of starting many others to utilize scr((p-steel for making eastings. Most all kinds of scrap-steel can be melted (l)orings, etc., are I)e,st melted by being packed iu cast-iron pots, etc.), and classes of castings MELTING AND MIXING STEEL WITH CAST-IRON. 381 found in which it may often be well utilized. The melting of cast-steel in cupolas, as far as manipulation is concerned, is in principle the same as melting cast-iron. For steel, more fuel and blast i)ressnre may often be required than for iron. Scrap-steel when melted in a cupola produces a product somewhat similar in nature to "white iron;" and as Mr. Masters writes under the head of " Remarks," p. 376, if small castings were annealed, I should say they would no doubt be similar to malleable iron, thereby making them suitable for hardware purposes. There still remains one thing to be done, and that is to have scra[)-steel produce, direct from the melted state, castings some- where near as strong as was the scrap-steel before it was melted. Who can best accomplish this (whether mixed with cast-iron or not) could, I assure them, "reap a harvest." There are thousands of tons of scrap-steel lying idle in the country. The industry of utilizing it into castings once started, there is no telling in what success it will end. 382 sTKAM-rowHit cu am:s. FOUNDRY CllAxXES. STEA:\I-rOWER TRAXES. As an introduction to tlic followiiij^ chapters upon cranos, the nutlior wisiu's it uiuliTstood that n(j patents c-ovlt any of llic devices siiown. and that any one is at lilx-rty to use and prolit l)y any of tlie ideas set forth. Tlie author's mode of dealiiii^ with the construction of cranes is one which is not only oriirinal, but also one which he thinks all will agree is practical, anr:ilf. I'.iil for shops tliat iiialvf a line of m.itliiiui V cjislings, the ahility l<» opfiatc liy hand as well as power will often he found valuahlc ; for tiieii the crane can he operated, when. throiiLrli acci(h'nt to tiie hoiler or pipes, or otherwise, steam could not he ohlaincil or used. n FRICTION rOWER CRANE. 387 FRICTION POWER CRANE. The Griffith & Wedge (Zanesville, O.) power crane shown opposite, is used in the foundry of tlie Niles Tool Works, Ham- ilton, 0. Several cranes stand in a row, and are all worked by one line of overhead shafting, to which power is transmitted b}' belt. The top gudgeon A, being hollow, admits of the shaft B passing through it ; and being engaged b}' the mitre wheels at S, the shaft R revolves the driving friction pulley Y. To throw the crane into power-hoisting gear, the lever D is pidled, which presses the friction drum against the friction pulle3's F and Y. To throw the crane into hand-hoisting gear, the shaft // and gear V slide out, thereby engaging the clutches at X. The pinion Z, also gear 3f, is keyed to the sleeve : this sleeve, of course, revolves upon the shaft H. When driving the crane by power, the gear F, which is keyed to shaft H, being, as shown, engaged with gear G, drives the pinion L, which then transmits power to gear 3/, thereby revolving the sleeve and pinion Z. The gears L and G, being keyed to the brake-shaft, make the brake operative, whether the crane is worked by hand or by power. These cranes have a platform at the rear, so the operator revolves with the crane. This also places him high enough to handily reach all the levers. The crane's frame is made of pine. One special feature is that of the carriage. It is not only a handy carriage, but a short one. Many cranes lose nearly half of their workinoj lloor area through having a long carriage. «Sl. uiHjn om- axle. With Fig. 1 1.') style of j> FRICTTOX rOWKR CRANE. 389 cnn'iiigc, one can often see the hoisting-clmiii hanging ont of par- allel, as shown. Bringing the chains close together, as at E, is often done in this style of carriage, for the purpose of making a short carriage. When the hoisting-chain in two parts, as here seen, is contracted out of parallel, as at E, there is more or less trouble caused when turning the hook R. I have often beea obliged to lower down and take part of the weight off a crane before I could turn the hook without twisting the chains. Such bother as this is very annoying, besides causing loss of time. I think tluxt it is evident that a shorter carriage can be i)racti- cally worked, made after the style of the Griffith & Wedge carriage, than tlie one shown in Fig. 115. Another point which would be well to notice is that of the moving or racking of the carriages. There are many devices for this purpose. With chains there is often much annoyance, caused through their stretching; and, again, the chain will be so situated as not to move the carriage steadilj-. I see by the Gritlith ct Wedge design, the carriage is made, as far as prac- ticable, to overcome these evils. It is hardl}' to be believed that a chain will stretch as much as it does. I have often been obliged to cut out from one to two feet in rack-chains during the first week or so they were used. Many carriages are made with no provisions for taking up an}' slack. As will be seen at TF, a simple arrangement for this i)urpose is provided. Having a slack rack-chain often causes much bother, and, where there is no provision for taking it up, it has to be often taken down and cut otT, involving much labor. As will be seen in the plan of carriage in Fig. 114, the two sheave wheels are carried to one side of the carriage, in order t«) allow the hooks, IT and C, to which the rack-chain is hitched, to have a pull as near to the centre of the carriage as is prac- tical. jMau}' cairiages are moved b}' a rack-chain upon each of their sides ; again, othei's will have onl}' one at the extreme out- side or in the centre. The thing to be sought for, in moving a carriage, is that it shall move along steadily, and have no more :v.iO ri;i( ri'»N ro\vi:i: ciiANi;. frii'tion tipon one side of jih than upon th<' other. A jjood wav to accuiDplisli tliis is to pull with one cliaiii as lu-ar the ci'iitrt! of (Miiiage as possildc. To [iiiU with two chains wonld l»e lictter, were it possiMe to have thcni always pnll even and alike. This, J think it is safe to say, is seldom clone, even with the Hat link ehain which is the l>cst to adopt for that pnipose. "Where there aic two comnioii link chains pnllinir a carriage, one will often see first one and then the other ijnllinir, every change causing .1 jerk. Were the links of chains all of an excvt length, and if llicy wonld not stretch, then with a true pitch-chain sheave they could he depended upon to pull alike. The blocks of cranes ofti-n cause us moulders trouble. They are fi'eiiuenlly made so light that it requiies the hanging-oii of weight to pull thcni down. Again, they will be made without any guard, as shown at Y.*, Fig. 115, p. 388. AVitli such blocks trouble is often caused l)y their getting out of the sheave grooves. As seen in the blocks of the Grillith & Wedge crane, there is not onl}' a guard, but the blocks are heavy enough to pidl the chain down. It is not necessary that a large sheave be used in order to make weight. Should a small sheave be used, the cheeks of the blocks could readily be made heavy enough to aid tile weight of sheave in pulling down the chain. There is not quite the objection to the chains Iianging out of parallel, caused through small lower blocks, that is stated with reference to the chains narrowtng up at the upper or carriage blocks shown on p. 388. However, when practicable, it looks and works better to have the lower sheaves large enough to cause the chains or ropes to hang parallel. Driving-power for cranes is not limited to the two modes here shown : some use hydiaulic power. The latest means is the enii)loynn'nt of electricity. How successful or practical its a|)i)lication for foundry cranes will prove, is yet to be seen. Tile i)rinciple involved in regard to power, as shown in the two cranes previously described, is no doubt at present the most practical ones for foundry use. i-ii of J J, HAND-rOW2R IRON CRANE. 391 HAND-rOWER IRON CRANE. Although power cranes have mauy advantages over hand cranes, the simple mechanism of the latter is alone a factor which will always command attention. The simplicity of hand cranes is snch as to allow their being made by almost an}' Arm, whereas the power crane will often require to be " built out- side." A few years since, the frames of cranes were almost entirely made of wood ; but at the present time many are made entirely of iron, the low price of iron making this construction nearly as cheap as when made of wood. Iron is really the proper mate- rial. Iron cranes not only look neat and light, but they are durable, and will keep their original shape. Wooden cranes, tln'ough unequal shrinkage, get more or less out of shape, thereby often causing trouble with carriages, gears, and chains. The iron hand crane (Fig. 116) of Messrs. "Webster, Camp, & Lane, Akron, O., which I am enabled to here show, is very simple in construction and readily worked. The end elevation shows the crane as one would see it if viewed from the front. The gears are shown engaged for " fast motion." To engage for " slow motion," the pinion A is pushed into contact with the gear B. The cranks, or handles, are removable, so that for either speed two handles may be used. Some cranes are so arranged that the handles always remain upon the one shaft. In such cases they are generally secured by means of a nut or pin upon end of, or through, the shaft. "NVliere handles are not thus secured, they should, as sliown at F, have plenty of shaft length. In this, as well as other fea- E^dtfJA Fig. 116. '?f>2 llAND-roWllH IKdN (IIANK. tiircs of the crane, the cxin'riciiic of inactical iiicii is seen. Sonic may think this sliaft (picstion one of minor importanfo. I know it's a simple thin;:;, and one to wliicli, liy many (U'si<;n- ei-s, no allcntion is jjaid. A liandlc that rctjuircs to be ehanj^cd from one siiaft to anothei' necessarily re(|niros a very easy fit. Where tiie sipiare part of sliaft is so short as with many cran<'s, the handles can readily work off without its l)ein<; iioticc4 llAND-l'OW i;il IKON CllANi;. ol'ti'ii irivc waiiiiiig licfoit' tlicy lucik'. Alioiit tlic only oltjce- tioii Id llitir use is tlicii" n'(|iiiiiii;^ siicli hirirf j 5.^ H 1^ n 54 78 11 17 13 155 6i 7i 4 .'J 5 n n 44 G4 <» 13 12 14i 5 6 5 45 4'1 n n 39 55 8 11 Hi 13i 45 5i 5i ^ - n - 83 - Gi - lOi - 4i - 6 4 4 u n 27 39 5i 8 9^ IIA 4 5 7 3i 3^ li n 20 30 4 8 10 3i 4i 8 U Sk 1 1 16 24 3 5 7 9i 3 4 9 2J 2? I I m 20 2i 4 6 8 25 35 K) 2i 2i 5 i 8. 04 13 1? 3 5 6i 2i 3i lOi 2 2 » £ 5.13 9 li 2 4i 5i 2 3 io.i IS 1^ rs rl 4.27 0.^ i li 4 45 15 25 JO^l H n 1 J. 3.48 5), 1 1 o 1 4i li 2 i;OF.F.LIX(; S XOTKS 0\ TilK TSKS OF WT1!E IJOPE. Two kinds of wire rope art- iiiaiiut'aclurcd. Tlu' most pliable variety contains nineteen wires in the strand, and is generally nsed for hoisting ROEBLINGS NOTES. 395 and running ropo. The ropes witli twelve Mires, and seveti wires in the strand, are stiffer, and are better adapted for standing-rope, guys, and rigging. Kopes are made up to 3" in diameter, both of iron and steel, upon special application. For safe working-load allow one-fifth to one-seventh of the ultimate strength, according to speed, so as to get good wear from the rope. AVhen substituting wire rope for hemp rope, it is good economy to allow for the former the same weight per foot which experience has api)roved for the latter. Wire rope is as pliable as new hemp rope of the same strength: the former will therefore rmi over the same sized sheaves and pulleys as the latter. But the greater the diameter of the sheaves, iHilleys, or drams, the longer wire rope will last. In the construction of machinery for wire rope, it will be found good economy to make the drums and sheaves as large as possible. The minimum size of drum is given in a column in the table. Experience has demonstrated that the wear increases with the speed. It is therefore better to increase the load than the speed. Wire rope is manufactured either Avith a wire or a hemp centre. The latter is more pliable than the former, and will wear better where there is short bending. Steel ropes are, to a certain extent, taking the place of iron ropes, Avhere it is a special object to combine lightness with strength. But in substituting a steel rope for an iron running rope, the object in view should be to gain an increased wear from the rope rather than to reduce the size. Wire rope must not be coiled or uncoiled like hemp rope. All untwist- ing or kinking must be avoided. To preserve wire rope, apply raw linseed oil witli a piece of sheepskin, wool inside, or mix the oil Avith equal parts of Spanish-brown or lamp- black. H'X) HAM) r(J\Vi;ii WUUJJKN C'ltANhS. IIAND-rOWER WOODEN CRANES.^ Altiioucii iron is the inodcni nintciial for crMiic fiaines, wood will, lio doultt, contiime lo be imicli iist'd. 'llic fact that tiiiibor i.s ol)tuinabk' in almost any section, that it is chc:ii) in first cost, and that local skill is easily available to design and frame it, are points which will command attention, and keep W(jod from falling into disuse. The timber chiefly used for cranes is pine, maple, and oak. There are probably more piiic cranes than all the others combined. The species of i)ine gen- erally used is the yellow or red i)iiie. The red Canadian i)ine, found from the Pacific to Canada, is the j-ellow pine of Nova Scotia and Canada. The limber is much esteemed for its streugtli and durability, and is used greatly for ship-masts, etc. The i)itcli pine of Carolina and Georgia is noted for its strength and dura))ilitv, in which (pialities it surjiasses others of its class. Maple is chiefly found in North America. For strength, it is superior to pine, and by some authors is placed ahead of oak. jNIaple being a sweet "wood is apt to "doze ; " l)nt if in good shape when framed, and given a coat of paint, it will i-cmain sound much longer than were it not thus treated. Oak, like pine and maple, has several species, and for its strength and durability is greatl}' prized. It is especially adapted for exposure to the weather in a damp climate. Its species ai'c found in almost all parts of llie countiy. Live oak is genei'ally considered the lu-st. It grows on the coasts of the (iulf of INIi'xico, and as far north as Virginia,. ' 'riiis and (lie followiiii; iIiiim- cIiMptcrs, with i-xci'ption of some addilious, the aiilliur liiij (ii'tel ajipi-ar in " Iron Trade I{e\ii.:w " of C'levt-land, O. HAND rOWER WOODEN CRANES. 397 The timber used for cranes is generaUy reonlnled more by uliat can be readily procured, ajid in the best shape, than from choice or preference of kinds. The following table, show- ing the transverse strength of woods, is deduced from United States Ordnance Department experiments, conducted by Hodg- kinsons, Fairbairn, Kirkaldy, and Haswell ; power reduced to uniform measure of one inch square, and one foot in length ; weight suspended from one end as illustrated by Fig. 117. Breaking weight. Brealiiiig weight Ash . . 108 lbs. Oak, white . . . . 150 lbs. " English . . . 160 a live . . . . . 160 *' Canada . . . 120 a red, black . . 135 Beech .... . 130 (« African . . . 207 *' white . . Birch .... . 112 (160 English . Canada (105 ■ i 157 . . 146 Cedar, white . 160 Pine , white . . . . 125 Elm .... . 125 " pitch . . . . 137 '• Canada, red . 170 (. yellow . . . . 130 Maple .... . 202 " Georgia . . . 200 In the construction of foundry cranes, the strains timber is Fig. 117. Fig. 118. subjected to arc chiefly transverse strains. The transverse strength of a timber is that which it would stand were it laid horizontally, being supported at one or both ends, and loadcid until it broke, as illustrated by Figs. 117 and 118. It is often remarkable how strength and liahtncss can be :)!is iiAND-i'owi.u \vooni:N CIIANI'.S. coiiiliiiitMl liv tlif jiii- ;iiiy oilier kiiwl dI' m.-nliiiin v. 1 lie form u;ivrii to tiinlK-i-, uikI the way it is fraiiu-tl, liave niiicli to do with its relative strength, as ^^ will 1)0 si-en by the followiiif; example for Figs. 11 !» and 120. To ascertain the relative sectional strength of tinil»er. mnlUjiJy the square of the dcplli hi) the ..„ ^ ^„ thiikuess. Fig. 120. Fig, 121. EXAMPLE. Square of depth 16 Thickness 4 Kelative strength 04 Fig. 122. Square of depth 04 Thickness 2 Kclalive strength 128 In the sections, Figs. 119 and 120, we have the same area, or niunl)er of square inches ; but by having the area in the oblong or rectangular shape, as per Fig. 120, we have a timber that will stand double the load that such a one as Fig. 119 would, were both to have the load applied on their respec- tive surfaces, ^and K; the timbers to be either supported at one end, as per Fig. 117, or supported at both ends, as per Fig. 118, and same length between, or from their support. There are many cranes whose frames would have been much stronger had the above principles been more strictly adhered to in construction. The following gives the fundamental princi- ples for finding the transverse strength of beams : — " Trcoisverse streu(/th of a beam is inversely as its length, ajid directh/ as its breadth and square of its depth, and, if cylin- drical, as the cvbe of its diameter. That is, if a beam C long, 2" broad, and 4" deep can carry 2,000 lbs., another beam of HAND-POWER WOODEN CRANES. 399 the same inatcrial, 12' long, "2" broad, ami 1" deep, will only carry 1,000 Ihs., luMnu; iuvi'rsoly as its lenutli. Again, if a beam G' long, 2" broad, and 4" deep can snpport a weight of 2,000 lbs., another beam of the same material G' long, 4" broad, and 4" deep will support doulile that weight, being directly as its breadth ; but a beam of that material G' long, 2" broad, and 8" deep will sustain a weight of 8,000 lbs., being as the square of its depth." — Templeton. " Ti'lien one end is fixed and the other projecting, strength is inversely as the distance of the weight from the section acted upon ; and stress upon any section is directly as the distance of weight from that section. " IVJien both ends are sv]^ported only^ the strength is four times greater for an equal length, when the weight is applied in middle between supports, then if one end only is fixed. " When both ends are fixed^ the strength is six times greater for an equal length, when the weight is applied in the middle, than if only one end is fixed. " Beams of wood, when laid with their annular layers vertical, are stronger than when they are laid horizontally, in the pro- portion of eight to seven. " The lower end of a tree will furnish the best timber." — II AS WELL. Accompanying this chapter, two wooden framed cranes are shown, which will not only give ideas in framing, but present valuable points in constructing jib-cranes for foundry use. The twent3'-five-ton crane (Fig. 121), shown on p. 400, is "triple-geared," /i" being the "first motion," B the second, and P the third. The shaft of pinion, A", is such as will slide out, thereby disengaging the first motion wlienever it is desir- able to operate the crane by its second or fastest motion. The " third motion " is not operated by crank. For some it might 400 ir AM) i'()\vi;u \vf>oi)i',N cuani'.s. ]}Q well to s:iy thai the tliii(l iiiotiun is ikMimI for tlic purpose of iiK r(:i>iii;j, the pmsir. Sumr immv tliiiik it <>(1<1 that tlio l)itclics wi-n' not more pioportioiu-d, the lirsl :in (stay in ahunt the Hanie position. The manner in which top j^ndccoon.s are gonerally inca.sed in cranes canse.s tiieni to Iteconic more or less honnd when cranes get ont of pluml). To overcome the i-vils arising from such efftH'ts. we use, as seen at 7', a round cap which tin; gudgeon can ri'adily accommodate to any incline tt) which the out-of' pluml) crane may oscillate it. The cap T, as si-en. covers tlic gudgeon so as to kci'[) it free of the dust which collects upou beam, etc. In this cap are tw(j small oil-holes for the purpose of keeping the gudgeon well lubricated, a thing which nuist 1)e attended to before one need expect to have a crane revolve easily. On p. 393 the question of cranes, when loaded, getting the control of the operators, was touched ui)on. That such things have often luipi)cntMl. most users of cranes can testify. In some cranes a ratchet (shown, old style, p. 402) is used. This is only of service while the hoisting is being done. ShouUl the crane through any cause "get away," it cannot be stopped until all the " mischief is done." Shown b}' plan and side views is the sketch of a brake and ratchet wheel wdiich are attached to the crane as shown. This brake is formed of two parts, best seen in plan view. The outer part V. which contains the internal ratchet, is loose upon the shaft. The inner part, which contains the springs and tia- ra tchet pawls seen at YY, is fastened by set screws or keyeil to the shaft. Sui)posing the crane to lie hoisting, the direction W would take would be that shown by the arrow. The pawls 1"!' turning the leverse way of the ratchet notches are sprung out by the sjjrings in IT. as they pass them liy. Now, should tlu' crane through any cause attempt to " get away," 11' would HAND-rOWER WOODEN CRANES. 405 then, of course, turn the opposite way to the iirrow directions, and in doing so the pawls would catch the notches ; and as the ratehet part V is held by the brake straps G and L^ the crane cannot, of course, run down. Should it be desirous to lower by means of the brake, all that is required is to operate the brake wheel A, seen in side elevation of the crane. The ihechanisni of this little machine rightly entitles it to the des- ignation of a safety ratchet-brake. The racking device of this crane is one woi'thy of notice, as it is no doubt the best that could be adopted for carriages that are pulled l)y two chains. The trouble that pulling car- riages with two chains generall}' causes having on p. 390 been commented upon, the subject will not be here discussed. This carriage is pulled by having a chain composed of malleable- iron links (one of which is seen at Fig. 123) passed over the sheaves 3fM, and bolted cto the wrought-iron bar K, seen in the plan of carriage. These links being all of the same pitch, and the sheaves 7^/3f, over which the chain works, being very accurate in pitch also, the carriage must necessarily pull very s juare, and without causing much friction upon the sides of the track. The track, as will be seen, is formed by railroad- rails. The way tracks are generally made is by simply using flat bars. The using of the rails shown not only makes a rigid track, but also helps to strengthen the jib, and presents very little friction surface for the carriage-wheel rims to work upon. Altogether this original idea is one that works well and is worth noticing. Shown by the ten-ton crane, the distance of the shaft upon which the crane handles are seen is 3' above the floor-level ; this is about the right height to place shafts for convenient working, or operating of the crane's handles. While the above is the most convenient height for shafts, the^' can be worked higher or lower ; the limit to their convergenc}' from the above heiiiht should not exceed 8" below or above the 3'. [OC) HAND I'OWI'.U \V0()1)I;N (•I(\NKS. Tilt' •icnrs ill all of these oniiics sliow the arms jind rims stioiiLjiY eoiistnicttil. This is soinethiiig too often neglected. 1 liave yet to see any goar'.s teeth fraetiireil from '•'■ jiure. strains," hut many arms aii|i JMw J>I> iiKidr from (')" to s" longer, so ns to fjivc moro SUpliiill to the jilt. It will lie noticed lliiit tlio top jaw (Fig. \2'>) is in:i(lo so that it can he placed after a post has heen set up. After the jaw is set ui)on the collar ^1, the piece // is placed in, ami a l)olt put through the two. The jaw then answers the same purix>se as if it were one solid casting, as per plan seen in Fig. Tilt. The plan Fig. 12!> is, of course the; strongest, and often the West to adopt where circumstances will i)ermit. The constructing of a post-crane does not always necessitate the erecting of a post csi)ecially for that purpose. It may be that one would like to use sf>ine post that is standing. With a jaw, as per Fig. 12.">, it can be utilized without the post being taken down. Should the post reipiire to have a collar to sup- port the top jaw, ideas aie illustrated in Fig. 130, showing how a casting made in halves could be bolted on to a square or round wooden or iron column. The diameter or square of a post or column may not be entirely regulated by the capacity of crane desired. The weight a post will have to support may often call furpose of making a crane : it is only where a post is required to support a build- ing, and in the same locality a crane is desired, are they advo- cated. The building of post-c-raucs is not advised except under circumstances which w ill not permit the use of a clcar- swiugiug pivot-crane. rOST-CRANES. 411 A peculiar fcnturo of this crane, which will no doubt attract the eye of many, is that of the racking arrangement. The movement of the carriage is done by means of an endless "racking-chain " passing over two 6" loose sheaves at K; from thence over the sheave E. This sheave, as shown in Fig. 131, is chucked into the pinion H, and both the pinion and sheave are loose upon the shaft D. As this sheave and pinion is made to revolve by means of the racking-chain, the spur-wheel S which is keyed on to the shaft iV revolves the truck-wheels W TF, thereby' moving the carriage. Cast on to the wheels WW are pinions which mash into the rack seen upon the side elevation of the crane. This rack is used for the purpose of insuring the carriage travelling square when heavil}' loaded. While this form of a racking device is quite a novelty and a success, as far as working is concerned, in point of cheapness it cannot be said to have much advantage over the style used in the twenty- five-ton jib or travelling crane shown (pp. 400, 414) , From this it must not be inferred that the style shown in post-crane would work well upon the ten or twenty-five ton cranes shown. For loads over three tons, such a style of carriage should give place to those shown with the heavier cranes. For holding up the lower jaw i^, bolts TT, as shown, are used. The construction of this lower jaw is simplified by making the cheek-pieces M so as to be secured to F by means of set screws. A plan view of the lower jaw F is seen in Fig. 12G. The width between the cheek-pieces M in con- structing a crane will be regulated by the length of drum required. For the same number of feet in height of hoist, the length of a drum can be much loss where wire ropes are used lustead of a chain for the sustaining cord. This, of course, means tliat in both cases tlie same diameter of barrel is used. For wire rope it is best to use the barrels as large as practica- ble, and they should be larger in diameter for wire than chain sustaining-cords. As this crane is only intended for loads up ^112 I'OST-rUANKS. to one Mini a half tons, it is liiit sin^lo-f^carcd. To fonstnict one for loads iaiigiiM4 from two to five tons or upw.iids, it would rt'fjiiiiv that the crane ho doiilik'-gcarcd, a thin;^ which can he applied to post-cranes as well as pivot swinginif-cranes. The construction of the lower jaw in the crane shown is such as to l)rin!j; the lower end of the hraces up fully five feet clear of the lloor. This will allow one's moukling up within ahout two feet of the ci'ane's post ; and also give good height in hoist when working undei' the hraces. The fraine-W(»rk of the crane shown is sti'oiig enough to carry a hjad of three tons, and is, as all frames of cranes should be, stronger than the sustaining- cord. TUAVELLING-CIIANES. 413 TRAVELLING-CRANES. Tnr. hnnfl tvaA'olling-crane sliown on p. 414 is one which tlie author has dcsigiiod to ilhistrate principles and ideas Avliich he thinks would work well in hand-travellers for foundry use. The capacity of the crane is intended to be ten tons. The arrangement of the hoisting and racking of the crane is in principle similar to those used in jib-cranes. For moving the traveller upon its longitudmal track, a shaft connected to the two wheels /SaS is operated by the bevel-wheel X and the pinion F, as shown. Moving "hand-travellers" lengthwise in a shop, is usually a troublesome performance to arrange ; so much so that I doubt if in this particular point a ten-ton "hand-traveller" can be made a success. I do not call a traveller a success that requires an army of men to move it when heavily loaded, nor are they a success when they cannot be made to travel much faster than a snail. A^'llile this crane is presented for ten tons eapacit}', it should be understood that such loads should be handled only occa- sionally. If it is desired to handle daily from six to ten tons, I would advise the traveller be operated by other than hand- power. In reality I do not believe " hand-travellers" can be made to move properly for foundry use with much more than five-ton loads. In designing the gearing for this crane, I thought it best to make it " triple-motioned," in order to save the necessity of employing six or seven men to clin)b up into the pendant to do the hoisting for heavy loads, which would be the case if the 11 1 I UAVI'I.I.INC (U ANKS. ^I'Miin^ well' oiilv '' c'ed .sliuil H. For •iiiiSi Fig. 132. linliter loads the hnndli-ts can lie used upon citlioi' of the olhcr two shafts shown ; and tiiiib tluy can hoist with incivased I'RAVF.LLING-CRAXE.S. 415 speed. In lowering; loads the crane can be manipiilalcd by 11i(> brake s^)\vii, if desired. The bra]fvel-" radius is nearly equal to 2(>A- revolutions jx-r minute. In dt'signing tlic gearing foi- a ci:uu>, it nuist lie remembered lh:(t to rjain poirer vithont a sacrifice m spe^d can only lie done by iiiereasiiig the iiioti\ e power liy wliieli the eranc is operated. GEARINO UP CRANES. 421 The " power of a crane" is but the product oi force ^ lever^ age, and time. The heavier the weight to be hoisted, the hunger time will be necessary in proportion when the same motive force is used. A crane which would require twenty revolutions of its crank to hoist the block one foot high has but half the power of a crane where forty revolutions of a crank are necessary to hoist the block the same height ; this of course means where both cranes have the same amount of friction. The loss of power in cranes through friction ranges from twenty to fift}' per cent. A crane may be so badly constructed that where a hundred pounds of force are exerted upon its cranks, onl}' fifty pounds are effective in hoisting the load, the balance being used in overcoming friction. To construct a good working crane, much judgment and care should be exercised in the construction of its gear- ing and shaft-bearings, and when used they should be kept well lubricated. To increase the power or pull of a crane without increasing its motive force, can be accomplished by any means which will decrease speed in hoisting. Plans which are generally adopted to accomplish this end are, first, by means affecting the " gear- ing-up " of a crane ; second, by means of multiplying parts in the sustaining cord, as set forth in chapter on page 42G. Ol3taiuing power or leverage in the crane by gearing is not, as some suppose, confined to the multiplication of "motions." The different number of motions given to cranes are simply for the purpose of increasing or diminishing its speed, and for convenience in procuring power by the use of the limited space allowable in the construction of cranes. A crane, if it were piactical to use enough space, could be made as powerful with one motion as if it had two or three motions. To illus- trate this idea, we will suppose the ten-ton crane seen upon p. 402 constructed so as to have the same power or leverage with '' one motion" as it now has with its " two motions." As the 422 fil-.AKIN'i II' (UANI'.S. crano i.s now goaiod, lln' crank wl en njon its first motion travels about is/i" for every 1" it raises the l)locks. To liave this same leverage oi power ol' 1 to l^i.'i in the al)ove crane \s\\\\ a sinjiU' motion or speed, llie wheel upon tiie drum's shaft would re(iuire to l>e made with the 1/" i)ilch, having '»28 teeth; and the iiinioii. li;i\iiiLC eleven teeth, as thei'e shown, would then re(iuiie tiie crank to turn the same numlter of revolutions it now does in raising the l)locks one foot high. Now, to show the impracticahility of using a wheel having ;328 teeth (leaving out the question of utility in ha\ uig differi-nt s|)eeds), it is only necessary to state liiat a wlu'cl 1^" pitch, having 528 U-eth, would be al)out 24' G" diameter. In gearing a crane, the jtitch generally used ranges from 1" to 1|". The pitch of u gear is the distance from centre to centre of two adjacent teeth measured u[)on their pitch-line. The j)itch-line of a wheel is the line tangent to the circum- ference of a circle i)assiiig througli the point of contiict of the teeth of two whe-jls when engaged, and is about midway ])etween the extremity and root of a tooth. The extremity of a loolli is the outmost face, and the root that which joins or forms the face of the rim of the wheel. The class of wheel-gearing most used for cranes is that termed ^ si)ur-w heels." There are two other kinds of gear- ing, — bevel and mitre wheels, which are also sometimes used. A spur-icheel is a wheel having its teeth per[)endicular to its axis. A mitre-wheel is a wheel having its teeth at an angle of 45'^ witli its axis. A bevel-zcheel is a wheel having its teeth at an angle with its axis. " To compute the pitch of a xoheel. — Dividi' the circumference at the pitch-line by the number of teeth. " ExAMiM.K. — A wheel 4U" in diameter reqniies 7.') teeth: Wiial is its pitch? ;5.1I IC x ID ^ 7;") = l.(;7.'):)". GEARING UP CRANES. 423 " To cnmpvto the cluimcter of a ivhed. — Multiply tho numbei' of toc'tli by the piU-li, and divide the product by 3.1416. "• KxAMi'LK. — Niiinl)er of teeth iu a wheel is 75, and pitch 1.G705'''. What is the diameter of it? 75 X 1.G755 ^ 3.141G = 40"." Haswell. "Where two gear-wheels engage each other and one is smaller than the other, the smaller is called the " pinion," and the larger the " wheel." When in contact, the ratio of their revo- lutions IS regulated l)y the number of teeth each contains. To Jind the number of revolutions m a j)inion to one of a tcheel. — Divide the number of teeth in the wheel b}' those in the pinion. With a wheel having 96 teeth, and a pinion with 16 teeth (96 -^ 16 = 6), we see the pinion makes six revolutions to every one of the wheel. In cranes the smallest pitch is used for the " first motion," those used upon the last motion being larger. This is done because the nearer to the pull of a drum a gear is, the greater strain there is upon the teeth of the wheel. The strength of teeth, and relative proportion in depth of face to pitch of teeth, are well illustrated b}' the following formulas, given b}' the Walker Manufacturing Company, Cleveland, O. "The durabilit}- of the teeth of gears, under the same cir- cumstances, is nearly in a direct proportion to their breadth, and inversely as the pressure. The strength of the teeth of gears is directly in proportion to their breadth, as the square of their thickness, and inversely as their length. For example, if we double the bi'eadth we only double the strength ; but if we double the thickness, or in other words double the pitch, keep- ing the original length and breadth, we increase the strength four times : but as the length of teeth commonly increases with the i)itch, this circumstance must be taken into view ; for if we 42^ (iKAKINf; ir CUANKS. (loiililc till' tliickiirss :iii(l 1i'IiS, Fig. 13.S, there is the greatest strain upon the hook: the b(jttoin, as at /*, Fig. 138, when the hook is loaded with two chains, is also greatly strained, and such strains have been known to break hooks at P. To construct a well-proportioned hook, tlie sections 2^ and *S' should be larger in area than that of any other portion, from the fact that there is the point which has to stand the greatest strain. Theoretically, a really well-proportioned hook would be one so constructed that an expert would be luizzlcd to rightly guess the part Jirst to break. While the above is true proportion, I do not think it advis- able to have hooks so finely constructed. It is well to have the section at N or S a little the weakest ; for then there will be a chance to watch and note any overloading of the hook, which can be told by any opening out of the jaw. It is advisable, in an}' tool that can endanger life, to have it, if possible, so con- structed that its user can be forewarned of any tendency to break. From the above tests, two things arc to be deduced. One is, that the flattened hook is the stiffest ; while through this very element it may be said to be the most treacherous, from the fact that they are often apt not to open sulliciently l)efore breaking to attract attention, while the round hook generally alTords ample warning of an overloading. The strength of the hook de[)ends greatly upon the mechanic who forges it. There is such a thing as abusinif and tlistortinir the librcs of iron so as HOOKS. 431 to leave the hook strahied within itself when finished, and no doubt many hooks have been broken that would have stood a much greater load if there had been more skill used in their construction. One may have hooks made from the same bar that, when tested, would give such different results as to cause doubts of the same bar having been used. Hooks should never be loaded to any thing like what may be thought their ultimate strength, and in designing them a large factor of safety should be allowed. Heretofore there has been, as a general thing, but little thouglit given to the question of proportioning hooks, as can be readily seen by considering the varieties in use. To Henry R. Towne of Stamford, Conn, (manufacturer of hoisting-machinery) belongs the praise of presenting, in his work upon cranes, a " standard hook ; " and through the courtesy of Mr. Towne the hook, accompanied by his formula for its construction, is here shown. It is no doubt a hook which will by practical men be received as one worthy of imitation. ..." Fig. 139 represents, to a scale of one-sixth natural size, a 5-ton hook of the dimensions and shape determined by the following fornnilte, which give the dimensions of the several parts of hooks of capacities from 250 pounds (or one-eighth of a ton) up to 20,000 pounds (or 10 tons). For hooks of larger sizes the formulae become slightly different, the general propor- tions, however, remaining the same. "For economy of manufacture, each size of hook is made from some regular commercial size of round iron. The basis, or initial point, in each case, is therefore the size of iron of which the hook is to be made, which is indicated by the dimen- sion A in the diagram. The dimension D is arl)itrarily assumed. The other dimensions, as given by the formula?, are those which, while preserving a proper bearing-face on the interior of the hook for the ropes or chains which may be 432 HOOKS. passed thr tiltiiiiatc rupture which tlie amount of material in the original l)ar achnits of. The symliol A is used in the formuhu to indicate the )iominal cupacity of the liook in tons of 2,0UU Fig. 139. pounds. The formulae which determine the lines of the other parts of the hooks of the several sizes arc as follows, the measurements being all expressed in inches : — D = OJi^ +1.25 £' = 0.64A + 1.60 F = 0.3:3A + 0.85 i7=1.08^ I = l.ZZA J - 1.20.1 K= \.V?,A G = 0.752) O = 0.363 A + 0.66 Q = 0.G4A + 1.60 L = 1.05.1 3/= 0.50.1 IT = 0.866.4 " Example. — To find the dinicnsiun IJ for a 2-toii hook. The formula is : — Jj = U.5A + 1.25, HOOKS. 433 and as A = 2 the dimension D by the formula is found to be 2^ inches. "The dimensions A are necessaril}" based upon the ordinary merchant sizes of round iron. The sizes which it has been found best to select are the following : — Capacity of hook . \, \, \, 1, \\, 2, 3, 4, 5, 6, 8, 10 tons. Dinicnsioii ^ . . f , \\, f, l^V, U, If, 1|, 2, 2^, 2^, 2|, 3^ inches. " The formulae which give the sections of the hook at the several points are all expressed in terms of A^ and can there- fore be readily ascertained b}' reference to the foregoing scale. "■ P^XAMPLE. — To find the dimension / in a 2-ton hook. The formula ifi 7=1.33^4, and for a 2-ton hook ^1=1| inch. Therefore /, in a 2-ton hook, is found to be l^f inch. " Experiment has shown that hooks made according to the above formulae will give way first by opening of the jaw, which, •hoAvevcr, will not occur except with a load much in. excess of the nominal capacity of the hook. This yielding of the hook when overloaded becomes a source of safety, as it constitutes a signal of danger which cannot easily be overlooked, and which must proceed to a considerable length before rupture will occur and the load be dropped." . . . Figs. 140 to 145 are cuts of hooks very useful for foundries. The hooks. Figs. 140, 141, maybe propely termed crane-hooks, as thej- are chiefly used with cranes. The cuts of Figs. 140, 141 show both ends of their hooks as being parallel to each other : in practice they are generally made so that the lower hooks L will stand at right angles to the upper hooks X. Hook Fig. 140 is one which is handy to hitch to crane-hooks in order to save labor and trouble in handling lighter loads than the capacity of crane-hooks is intended for. In heavy cranes the benefit of such a hook is much felt, as the bending and turning of heavy 434 l!()f)KS. hooks and Mocks in liilcliiii;^ onto lii^Iit IoikIs is more or less ii nuisance. Jn sonic cases it is well {<> have two of these hooks, one to lie lighter than tiie otlii-r : the larjier of the hooks can often be nsed to good advantage if iikhIc marly the capacity of the crane's hook. Fig. 141. Fig. 142. Figs. 141 and 142 arc what arc commonly known as " changing hooks," on account of their being used in changing loads from one crane to another. Fig. 142 may be termed the safest hook from the fact that it is welded to the shank as shown. Fig. 141 is the most popular hook, no douljt because' its double hook-end presents the least interference when hitching Fig. 143. Fig. 144. Fig. 145. on. Fig. 14;3 is well known as the S-hook, and is one found to be very handy in many ways, and can be made from Hat iron as well as round. Figs. 1 14 and 11") arc a style of link and hook M'Idoiii to be found. Tlicy are simiily niadc from llat iron, ranging from h" up to 1" in thickness, and in width from 1" up to 3". They make the stilTest kind of a hook, and would, no doubt^ be uuich used were their strength mure fully known. i BALANCING AND HOISTING MOULDS. 435 BALANCING AND HOISTING MOULDS. TiiK l):ilancing and hoisting of moukls is an operation tliat often involves experimenting, and sometimes resnlts in loss of life or limbs. Of course there are a large number of moulds that one can readily hitch to, but again there are a large num- ber that require good mechanical judgment and knowledge in hoisting ; for such, the following notes and ideas set forth will be of value. In hitching to moulds, there is one thing that is very apt to be overlooked. The general impression is, that, if the crane- blockg hang directly over the centre of a mould's weight, it will hang level when hoisted up. This idea is not correct, as will be seen by the simple example illustrated in cut marked "Test," Fig. 14G. This block, instead of being suspended by an over- head fulcrum, is let rest upon an underneath fulcrum. The block is divided by a dotted line. Each of the parts B and A weighs exactly alike. Still you have to deduct 6.76 pounds, or nearly 7 pounds, from JB, and add it to A^ in order to make the block balance, as shown. This will be readily understood b}' those who have studied the princi[)le of the lever, and illustrates that a mould's centre of weight is not always its balancing-point, and that, instead of guessing for the centre of weight, we should guess for its centre of gravit}'. Some may ask, Is there not a more intelligent way to hitch to a mould than by mere guess- work? There is no practical way. Of course the weight might be figured, and its balancing-point be determined; but the time involved makes such a course generally impracticable. As shown by the plumb-bob line, tiie fulcrum or lifting-chain 43(3 I'.ALANCINfi AND llOlSTINf; MfXI.DS is (liivctly over the centre of t^iavify of tlu,' wcinlit. This is oI>t:iiiU'(l tlmuiuh tlu' ic^fiilntion of the slings sliown liitclictl to tlif lifUng-lH-aiii. Tiic icguliitioii of slings to nialvO :i mould lialnncc, allliongh ni>|iai(iitly so simple, is an operation that sometimes puzzles a Tnoiilder. It often tronlilcs him to tell which way the slings should he moved upon the lifting-beam, when they find a nmuld hanging similar to the weight that is shown at M. in dottccl lines l)elow /?, A. The cause of such unlevel balancing woulil be, that the fulcrum ov lifting-block was hung over the point l\ seen in B A, the riuhl-liund sling being set in the beam's notch No. 1, and the left-hand sling set in No. 1. To make the ■weight hang level, they must be placed as shown ; remembering that moving a sling towards the centre of a beam lifts up ilte moiikVs side or end, and that moving a sling towards the end of the beam lowers it. I have often seen lirst-dass moulders obliged to study for quite a while before they could tell which way the slings should be moved. About the most dangerous class of moulds with which we have to deal are those similar to the one marked Cylindi-r. In lifting such moulds, extra care must be taken, or the mould will turn over on account of the weight being all above that portion by which the mould is lifted. In hoisting any mould, as long as we can have tlie largest portion of its weight below the point by which it is lifted, there is generally little danger of its cap- sizing. Some, in hoisting such a mould, will drive wedges beneath the cross or beam, as seen at X. This is, no doubt, a good i)lan to adopt in hoisting top-heavy moulds. The farther from the lieam the point from which the crane-hook is hitched to it, the more weight will it recpiire to pull tlu' lifting-beam out of balance ; that is. if the puiut by whicli the lie.-im is suspended is rigid, so tliat it will alwnys i-emain in its own relatic^n or angle to the beam. In the biam shown lifting 7i and ^1, the chain-hook is hitched in an upright rigid beam at riizht angles BALANCING AND IIOISTINO MOULDS. 437 to the main beam. In this upright beam are four holes. The fourth or upper one is the fulcrum [)()iiit now used. To illus- trnle how we can regulate this point, we will suppose that this beam has no weight upon it, thereby allowing us to roek it back and forward. After noticing how much weight it A\ill take to make one end come down to a given point, we will then cut off the top down to hole No. 3. The hook being hitched in this hole, we again try it, and so on down to No. 1. Now, I think it is very evident that with the top three holes cut off, and No. 1 used for the fulcrum, it will not require much force to turn the beam entirely over, did the chain seen not prevent it. This explanation will, I think, prepare for an understanding of the principle and advantages of the cross shown. The ideas embodied in this cross, and its lifting slings and hooks, are such as can be applied to all classes of beams. The rigging, as shown, was devised by R. B. Swift. It is the first cross of the kind I ever saw ; and, as I am seeing it used almost every day, I know it to be a valuable appliance. The ordinary plan of lioisting with crosses is to hitch to an eye S. By this plan the fulcrum is but little above the centre JVof the beam. Now, as we have seen, that, the higher we raise the fulcrum, the harder it is to tip up a beam, we must acknowledge that by hitching at I", and having the hook slings spread apart as shown, it would be a hard matter to tip over a mould, even in hoisting top-heavy moulds similar to the cylinder shown. In using this lifting-cross, we rarel}' use any wedges between it and the mould X. So, if the latter is not exactl}' balanced at the point where it is hitched on, there will be little danger of its tipping over if the mould does not lift in a level position. Another feature of this beam is that its straight face V is underneath. This construction is good, as it gives a more reliable surface to wedge against when using the cross for bind- ing a mould together to be cast. Still another good feature is 438 lUI.ANCINfi AM) 1I()ISTIN(; MOULDS tlio " l1. ciilliii^ ciiiMdas' liiiiiii;H, 277. "()C,, ni2. as r(M|uin'(l for coal ami coke, 277, .'Jol, '.]{)'>, :iW. for 12" to I.S" cupolas, 2(J(). ilHTcrciicc of, in cupolas and blast-piiK's, 307, 308. f,Mui,'in<,' of, .'{07, .'](KS. mild for cupola.s, 2(18. objections to usinu. ;502, 300, ISturtovant's tabic for, .JOtl. Blast Tipes, detachable leather or rubber, 2G8. diameter r.s. length for, oKi. friction of air in, ;{l(j. reference-points upon, 320. table for equalizing the diameter of, 317. table for the diameter of main, 318. value of air-tight, 310. Bloweks, ' location for, 310. driving-power for, 302, 308, 310. r.i.ow-noLES, caused from pouring; dull iron, 0. produced by chaplets, 52. generated through mould-blowing, 41. BOLTIXO, down binders, — plans for making, 204. down floors for green-sand work, 221). down loam moulds, Oo, 88, 438. half cores together, 02. up a difficult loam core, 259. r>ui:NiN(i OF Castings, 217. amount of iron to use in, 223. grade of iron to use for, 22;3. Bkusiies, camers-hair, 171, 210. Candees, use of, in closing moulds, 57. ('AliiiON, in blacking, 211. in fuel, 280, 305. in steel, 377. INDEX. 441 Caiikiages, anti-friction boann<2;s for axles in, 233, 416. devices for imllins; crane, '.iSi), oUO, 40."), 411, 415. for delivery of large castings and ladles, 2ol. ill-constructed crane, oS8. short, advantages of, for cranes, 387. tracks for cranes, 4U5. Castings, cheaii bought, 15. cold-shut, 11), 109, IGl, 213. designing, points of value in, 2, 20, 21, 54. "dirt in gated end of, 114, 127. dirt, injury it can cause to, 10, 19. dirt, provisions for collecting and confining it in, 16, 42, 50. dirt, rising to upper surfaces of, 41, 44, 239. dirt, where generated from in, 15, 122, 127. filleting for strength in, 3. finishing up, allowing stock for, 114, 118, 132. good, uncertainties in producing, 24, 31. large, specimens of, 72, 76. over-shot, 100, 138, 159, 259. poured with hot and dull metal, 38, 41. round edges on, 161. smooth, points in procuring, 13, 38, 40, 45, 102, 210, 214, 226. sound finished, science of making, 39. sound, difficulties in producing, 3, 13, 46. strains on, 55, 147, 163, 230, 256, 263. strength of, 1, 8, 14, 19. strengthening, 3, 54, 377. strong, heavy scrap for making, 280. weights of, errors in figuring, 247. well proportioned, 4. Wrinkles in Mouldinrj Small: — core arbors for small, 141. cores, making of, for small, 102. flask hinges for small, 139. mould-boards for small, 134. making joints on moulds for small, 1.59. making patterns for small, 165, 167. printing blacked moulds for small, 209. procuring "good lifts" on moulds for small, 159. skimming-gates for, 123. Ill: INDKX. Chains, (lilt link rarkinq. 4(11, Inn. iiiiilli|ilir:ili(in of jmrt.s in cniiic-lioisliii;,', 4l'(>. slifiiKlh of, .isj, 42C.. stii'tcliiiig of oarriaj;*' pulliit;,', .".sO. trcacluTOUsnoss of, ."J'.t.!. nii-lianillcl liaii,i,'iiig of (•raiir-lioisting, .TSS, 300. ('ii.vi'i,i:rs, (lislanoe to allow for wodfjinp;, IS;]. iron stands for support in j,', (>4, 1S;5. improper settin^j and wod^ini; of, 17'^. loose and tight heads on, 184. sharp pointed, 183. stem for, 184. vooden blocks for supporting bottom, IS;"}. CHAPLETIXG, green-sand pipe-cores, 141. slanting core surfaces, 184. wrinkles of value in, 52, 57, Gl, 93, 2G0. Chilled Axle BEAUiNdS, 231. Chilled Rolls, blacking for surface of chills for, 237. handy flask for small, 238. novel flask for long-necked, 234. ride and table for thickness to make chills for, 235, 230. utility of whirl-gates in procuring clean, 23l>. ClXDEUS, beds under moulds, 132, 103. fine, power of, to resist pressure, 103, in cores, 58. in loam-work, 50, 02, 07, 83, 258. use of, in venting deep-sided moulds, 101. ClUCLES, rule for division of, 32, 34, 263. table for areas and circumference of, 322. CoMiirsTioN, 305. chemical action of, in cupolas, 300. creation of, in centre of large cupolas, 301, 303. economy of, in core ovens, 227. forced, in deep drying-pits, 01. increased in cupolas by use of "upper tuyeres," 288. pound of air required per pound of carbon for, 305, 308. ■^ INDEX. 443 CONTKACTIOX, (lofiiiilion of, for foundry praclico, 400. of long' riinnor gatos, !K). strains caused to castings tluough, 220, 400. Cores, bank sand in, 102. beer on, 103. blacking small, saving labor in, 102. centring of vertical set, 58. cinders in, 58. cylinder, port and exhaust, making of, 52, 104. difficult loam, 02, 67, 250. dry sand, expense of making, 140. filing a taper on roimd, 170. fine sand for, 102. flour in, 102. flour and rosin in, 103. gas in, cause of, 101. green sand pipe, 140. green sand, for arms in wheels, 203. green sand, advantage of, for pipe castings, 145. making and venting of, 101. pasting of, to form air-tight joints, 92. rosin in, 102. sagging of, 103. segments of, 05, 250, 254. setting and centring of ordinary, 173. setting of cylinder, 51, 57, 108. sleeking green-sand pipe, objections to, 142. splicing and securing vents in " butted," 118. suspending a heavy dry-sand, 91. thin, making of, 101. weighting down, rules for, 198, 202. Coke Arboks, for green-sand pipe cores, 141, 144. long skeleton, 91. self-forming print and supporting, 144. thickness to allow for green sand on pipe, 142. tnmnions on, 143. vent-holes in, 142. 444 INDKX. ('. ability of, to run loiii: licats, 2S8. biin,^'in,!,'-ui) of, 2(17, 'JtM^, 277, 2S(), ^^0'2, :]\2. capacity of a 12", 15", and 1»", 270. capacity of 20" to 80", ;'.14. cliari^ing-doors, advantag*; f)f liij^li, 2ss, ;;n|. comments on, •]0L constructed for coal or coko, 271, 27<">, ;;o;!, :;20. llame at cliarging-doors, diminisiiing of, 2'.Mt. fluxes for, see Fluxing, p. 44!). hanging-up of, 2()7. illustrated, 2C(), 274, 27S, 292, 294, 290, 29S. illustrated wlnd-chandjers for, 274, 292, :j00. large, points for consideration in making, o03, liquid iron accumulating in, 277, ;J12. oblong, construction of, 303. oddity in designs of, 287. original plan for small, 200, 271. peep-holes in, 208, 29;1, 304. picking-out and dauhing-up of small. 207. "scaffolding" of, prevention for, 277,313. shells for, construction of, 208. small, advantage of, 205. small, preparing of, for loiig heats, 207. small, successful melting in, 207. stacks for small, 209. styles used in small, 200. taper in small, advantage of, 208. tuyeres for, and meUiny in, etc., will all be found under their respective heads. ClIAKGIXG UP CUPOLAS, closeness of, when using coal or coke, .308. difference in weights to use with coal and coke, 270. 270. descri)ili\e modes of, 270, 27">, 279, 293, 295, 297, 330-370. effects of random, 285. INDEX. 447 Charging up Cupolas, — Continued. Avciglits for 12", 15", and IS", 270. with heavy scrap-iron, 278, 280. CUPOT.A-MXIXGS, blast cutting out, 277, 290, 291, 306. daubing for, 267, 860. diminishing the diameter of large cupolas by false, 272. fluxes, benefit of, in preserving, 312. improper daubing of, 312. thickness of, for small, 267. Cylixdeks, blow-holes in, 41, 52. cast slanting, 52. cast with one head in, 51. gating of, 42, 44, 53, 59, 63. grades of iron used for, 52. horizontal and vertical casting of, 39, 48. jacket, 60. locomotive, 43, 49. marine, 54. obtaining of a clean bore in, .38, 51. obtaining of a clean valve face on, 48, 55. scabbing of, 38, 45, 50. unequal wear and cutting of, 52. lui-parallel port and exhaust openings, 55. unsound riser-heads on, 46. Dull Liquid Iijox, as used in pouring heavy work, 122. causing cold-shut wavy castings, 161, 213. cause of holes in castings, 9. caused through delays in handling, 283. liable to be caused through melting heavy scrap, 280. lifting pressure of, 190, 193. reason for pouring castings with, 38. Dryixg, a cylinder in a pit, 60. loam mould on the floor, 80. kettles for, 61, 172. temporary enclosers for, 86. P'ACING-SAND, causing veined castings, 213. 448 INDKX. FAci.\(i-s.\M>, — ('otitinnril. for skiiiHlricd grocn-sand moulds, 170. iiiiiiiipuliiliun in .llsill^, l:!:>, l.VJ. iiuxiny of, for green-saiul work, 214. Fkkdino, hy '' (lowiiii^ olT," 47, 5;]. iiiaiiiitulatioiis in, 7. poron.sncss caused througli ill, 41. solid, 2, 47, o6. unpractical, 3. Fi".r.i)iNi;-iii;Ai>.s, below j, 35S. marble-yard chips for, 334, 366. oyster-shell for, 339. utility of, in cupolas, 312, 314. Foundries, facilities for handling metal in, 283, 284. good control of, 30. labor-saving rigging in, 140. machine labor in, 240. railway-tracks in, 230. Foundry Facings, compositions of cheap, 211. machinery used in manufacture of, 212. the use of, 211. Foundry Practice, hydrostatics applied to, 195. literature upon, 25, 30, 283. novelties in, 22. patents for, 22. progress in, 29, 241. specialties in, 29. Fuel, best for melting hot iron, 273. for skin-drying green-sand moulds, 172. kindling of, in cupolas, 271, 276. natural gas for heating ovens, 227. per cent economically used in melting iron, 284, 297. per cent of carbon in, .305. slack or soft coal for heating ovens, 226. Gaggebs, castings lost through ill setting of, 1-58. manipulations in using and setting, 1.57. preference for cast or wrought iron, 157. setting of, in skin-dried copes, 170. Gas, cushions formed in moulds, 213. in rosin and flour, 102. natural, as used in a core oven, 227. 450 INDKX. (iATKS, rontrartion of long, 00. nii^liinii of, its. cutliiig inoiiMs, iircvciitioii for, ITd. kiiiil easiest upon moulds, 117. for fylinders, 4;}-4(}, 5J>. for chilled rolls, 2;]9. " How off," 53, 194, 218. horn, no, 123. table of, equivalent areas in round and square, etc., 244, 240, top pouring, 129. skimming, see Skimming-Gates, p. 457. styles conmionly Tised, 129, ISO. which distribute and confine dirt, 116. whirl, 18, 90, 237, 239. underneath pouring, 117. Gear-wueels, contraction allowed for large, 203. construction of arms and rim for, 400. dcGnition of phrases used for, 422. device for moulding, 242, 201. form of tooth recommended for large, 204. pitch used for cranes, 422. objections to core cast, 201 r strains In cast, 400. strength of teeth in, 423, 424. tables for computing pitch, etc., of, 42.3. table of standard faces for spur, 424. utility of worm, 425. HEMP Rope, circumference of, to equal strength of wire rope, 394. objections to, 304. substituting wire rope for, 395. Hinges, for small work flasks, 139. Hoisting Moulds, 435. a difficult loam core, 07. determining centre of gravity in, 435. propeller-wheel copes, SO. Hooks, 428. crane "changing," 434. INDEX. 461 Hooks, — Continued. designing of, 431. experiments on strength of round and flat, 428. forging of, 480. formulas for constructing crane, 4ol-43o. proportioned construction of round, 421). sizes of iron for crane, 433. S, O, and C, 434. true proportioned, 430. weakest portion of, 429, 430. Iron, Cast, benefit of agitating fluid, 10. formulas upon strength of, 14. specific gravity of, 196. strength obtained from hot-poured, 8, 9. three essential factors to determine in, 11. welding of steel and wrought iron to cast, 217. Joints, ability required to make irregular-shaped, 155. bead for hiding overshotness at, 100. blacking of dry sand and loam, 90. charcoal blacking for parting, 256. difference in finning dry-sand and loam, 90. for small castings, 134, 158. objection to patched, 156. paper for forming, 117. points in forming loam, 100, 259. proper ways to form deej) pocket, 157. raised, 112. rule for slope in slanting, 157. Ladles, cause of sulliage gathering upon skimmed, 127. melting iron in a, 249. screw crane, 232, 238. Level, how to test and use xmtrue, 154. Level Beds, how to make a true, 153. made Avith pulley rims, 251. Loam-cakes, for forming grooves, 80. for absorbing moisture, 83. 452 jNi)j;x. LoAM-WOKK, liuililin;! onpos, R5. cindtrs in, 8^5, 2."»S. fiilse hul» iiiiuli" of, S2. puitlfs for dosing, •.»;!, 200. iiiuking i)lat<\san(l rings for, ^fy-^"!, 200. means for obtaining rcquinMl lliicivncss in, 00. odd ways of i)uilding, 5il. pits used for moulding in, 00, TO, skeleton copes for, 85. springing of moulds, 5.">. stiffening plate for, 00, 07, 200. ^fAcniXK-MorLDixu, advantages claimed for, 14S, 140, 240. Melting, advantages of coal for, 270, 273, 278, 280. advantages of coke for, 273. benefits derived from coal and coke mixed, 274, 27S. capacity of cupolas from 20" to 80" diaiueter, 314. economy in, 2&j, 287, 205. escape of heat in, 288. fluxes, aiding, 312, 314. Lea\y block or scrap in cupolas, 278, 280. iron hot, 274, 284. long heats, 274, 270, 277, 288, 280, 313, 314. small quantities of iron, 248, 205. speed in, 273, 280, 300. scrap-steel in cupolas, 370-381. ■\vn)ught-iron scrap in cupolas, .377. ■wrought or steel borings in cupolas, .381. vith all coal,* 270, 330, 331, 333, 330, 330, .340, .344, 307, 372. ■with all coke,* 270, 275, 203, 207-, 332, 334, :i;J5, 341, 342, 340, 347, 349, 350-352, 354, 358, 359, 301-360, 308-370, 375. with coal and coke,* 270, 275-, 293, 337, 338, 343, 345, 348, 353, 355-357, 300, 371-374. Mklters, superstitious and intelligent, 282. Molasses, blacking for chill rolls, 237. water on cores, 103. water on skin-dried moulds, 171. * Total meltiuga with all coal, tun ; with all cuku, Uccnti/-nine ; villi coal aud coke, eighteen. INDEX. 453 MOUI-BERS, bench, 158. good reliable export, 2G. ignorance of many, 32. making cores, 101. mental and iihysical development of, 20. progressive, 23, 283. Moulding, a curved pipe from a straight pattern, 250. a jacketed cylinder, GO. a large piston, 179. device for sweeping gear-wheels, 2G1. difficult loam cores, G2, 67. elbow and branch pipes, 143, finished castings horizontally, 114. hydraulic hoists, 89, 114. large air-vessels, 256. l^ipes on end in green sand, 252. propeller-wheels in loam, 81. true gear-wheels, 149, 261. MOULD-BOAKDS, composition for making, 13G. making match plate, 137. making plaster-of-Paris, 134. making sand, 136. mended with beesw^ax, 13G. patent elastic, 137. styles commonly used, 134. wooden, 136, 149. Moulding-machines, patent gear, 242. utility of, 240. Moulding-sand, elements in, 211. in cores, 102. oil and litharge in, 136. sharj) sand mixed with, 170. strengthening of, 169. wet with beer for mending loam-moulds, etc., Ill, 118. Nailing, around core-prints. 111, 175. corners of loam-moulds, 59. l.'ii INDEX. N.VII-IXf!. — f'outinvrd. ctli^cs of saiiil iiioulil-lioanls, 1.17. joints of f^rcfii-siiiKl moulds, I.")."), skin-tlricil gi-ffii-sainl iiiouMs, 170. OVKNS, construction of a nioilcrn, li'25. heated witli natural gas, '2'21. OXYOKN, causing "sulliago" upon liquid motal, 127. union with carbon in melting, 2tys, 302, oO-j, .300, .'J78. Pasti:, discretion in use of, 108. mixed with clay-wash and blacking, 100. mixed with oil, 109. to properly mix, 110. Patterns, abuse of, 1.50, 104, 1G6. brass and iron, 107. constructed for bedding-in, 154. draw irons for, 100, 108. draw screws for, 104. facilities for drawing of, 107. formed of sand, 117. liollow elbow and branch pipe, 140, 143. "loosening-bar" for rapping, lO-'j. lack of taper to, 104, 108. *' pounding-block " for preserving, 104. pulley rim used for moulding-pipes, 251, 254. rapping of, 159, 10.5. rapping plates for, 1.59, 1015. segments of, 251, 2-54. skeleton frame for, 117, 132. PATTKKN-MAKEI5S, attainments of, 104. doing moulder's work, .32. making i)atterns for linished castings, 41, 42. remarks for, 181. thought and skill required of, 108. unskilled, 104. INDEX. 455 TirES, elbow and branch, 140, 143. points of value in horizontal moulding of, 20. pulley-rim used for moulding, 251, 254. Pits, casting deep work in shallow, 93. desirable location for, 229. fitted up for drying loam-work in, 61. formed with cast-iron rings, 228. for moulding loam-work in, CO, 70. v vent channel, 228. rLASTER-OF-PAEIS, composition of, 134. making mould-boards of, 134. POUKING, air vessels, 260. chilled rolls, 238. condensers, 67. creation of "sulliage" when, 127. cylinders, 44, 53, 59, 65. green-sand pipes on end, 255. grooved drums, 76. heavy castings, — temperature of metal used for, 122. large volumes of metal, 120. moulds having extremes in space for metal, 44. momentum effect in, 187. propeller wheel, 88. slow filling-up by vertical bottom, 192. thin pipe vertically, 45. top and bottom, advantage of, 44, two "open sand" plates in one mould, 260. POUUING-BASINS, construction of, for skimming, 18, 117, 130. cutting of, 130. error in making long, 131. for chilled rolls, 237, 239. ^ height above "flow-off risers," 194. made in loam, 65. patterns for forming, 255. 4')«) INDKX. rKKssritr, OK I.ii^i ii> Ii;(>\, inoinoiitiini, dofinition of, 104. updii rliapli'tcd <'on's, risks from, 1S2, upon l)ott(nn and side of moulds, rule for liudint;, llifj. upon sides of flasks, 140. statical head, dclinilion of, 20-L wlu'n pouretl dull, I'M, I'M. Prtxts, chanifcrinc: roro. 177. cores fonninj,' tlu-ir own, W'l. discussions upon, 173. for pipe or column patterns, 14.'?. gaggering and securing around core, 111, 175. making cylinder core, 110. setting cores without, 133. taper, 174. vertical loam core, 57, G7. Printing of blacked green-sand moulds, 200. Ka.m.mixcj, hard, 150, 103. to obtain "good lifts," 1.58. up loam moulds, 05, 88, 228. PlSERS, "blind," 126. current of air through, 126. "flow-off," 49, 218. influence of in lessening pressure, 190. PiODDING, green-sand cores, 251, 254. loam mould, 59. KOLLIXG OVKK, advantage of, 140. bad work caused by, 140. Avrenching flasks by, 148. Rosin, in cores, 102. Scabs, friction at gates causing, 117. loam moulds, part most liable to, 50. range for thickness of, 19. sticky blacking causing, 210. top pouring causing, 45. INDEX. 457 Screws, for adjusting and centring loam-cores, 64. pitch of, definition for, 7-4. swivel, 438. Slag, accumulation of, 310. creation of, causes for, 2G7, 291, 312. cold blast eifecting, 312. tapping out, 311. Slagging out, table showing benefit of, 314. Slag-iioles, position and height for, 311. Shrinkage, definition of, for foundry practice, 406. holes caused through, 2, 7, 41, 46. per cent in, experiments to determine, 4. percentage, rule for figuring, 7. round balls, 5. Skimming-gates, bad elements in ordinary, 125. "blind risers" attached to, 126. castings gated to each other acting as, 127. cores for forming, IS, 116, 117, 125, 130. heavy and light work, 120. illustrated forms for, 17, 121, 123, 125. iong channel, advantage of, 127. patterns forming, 124, 126. patent, 125. positive acting, 132. relative proportions for, 17, 122. utility of, 19, 122, value of, for heavy work, 122. whirl, 120, 123. Skin-drying, green-sand moulds, 169-172. loam moulds, SO. Spindles, arms, novel plan for, 68. anti-friction arm for, 82. for horizontal sweeping, 89. for revolving loam cores, 66. l.")S INDEX. t>ri;. Stkel Sckat, annealing of eastings made from, ^>1C>, oSl. carbonization and oxidation of, 378, 3SU. carbon, high and low in, 377. castings made of, 370. • heat required for melting, 378. melting of, in cupolas, 370-381. melting of, in crucibles and air-furnaces, 380. mixed with cast iron for chilling-purposes, 377-380. principles in melting, 381. procuring homogeneous castings, 379. soft, best for making strong castings, 377. strengthening cast iron with, 377. Steel, welding of, to cast-iron, 217-220. Sti;aiuut-ki)ges, how to make level beds with, 153. parallel, 153. squaring beds with, 30. Sweeps, air-vessel, 275. balance weight for raising and lowering, 87. cylinder, 58, 02. dry-sand taper-core, 92. groove-drum, 73, 77, 79. gear-wheel, 202. ill gauging of, 09. ironing wooden, 78. lathe face-plate, 132. revolving loam-core, 07. SWEEI'ING, adjustable guide for, 82. dinii'ult loam-cores, 02, 07, 257. device for geai--wheels, 242, 201. INDEX. 459 Sweeping, — Continued, grecn-saud pipe-cores, 140, 142. grooves iu drums, 72, 70, 78, 80. large lathe face-plate, 132. long irregular dry-sand cores, 92. manipulations in green sand, 117, 132. revolving loam-cores, 70. "thickness" on loam-moulds, 84, 256, 258. under sm-face of loam-moulds, 60. Testing, bars, moulding of, 13. bars, size for, 8. burnt or mended castings, 218. machine, 10. pig-iron, 2G5. pipes, 20. spring of "bolting-down binders," 204. table giving strength of hot and dull i)oured bars, 9. value of cupolas, 287. Tubes, connecting vents of butted column-cores with, 118. securing core vents with, 64. TUYEKES, areas of for small cupolas, 268, 271, 320. areas adaptable for coke and coal, 301, 320. area and construction of, rules for finding, 319, 320. choking-up of, 307, 313. dimensions for a 12", 15", and 18" cupola, 271. equal division of, in cupolas, 319. height to adopt for coal and coke, 276, 303. large, advantage of, 307, 313, 315. kept open for long heats, how to, 307, 313. ratio of areas to that of cupolas, table on, 321. two rows of, advantage of, for long heats, 289. two rows of, experiments with, 289. two rows of, speed gained in melting by using, 289, 361. top I'ow, rule for height and area of, 291. top row, objections to, 291. valves for closing top rows, 289. -1»;0 INDKX. Vknts, caiisps of iron pottinp info core, OS, f»1, T).', ns, 107, US, isl, 2(M5. (•arryiii.;,'-off of vcrticiil set core, 177. coiw^triiclioii of i)ort ami exhaust eorc, ."iS, KXJ. fornietl by roils and strings in cores, KMS. metal burstin:^ through core, 107. risk of metal getting into under core, isi, 2fK5. securing core, 5S, (14, !t2, lOS, liy, ISl, 200. splicing or connecting, lis. Vkntincj, cores, 101, lOn. joints of mouhls reliably, 101. moulds that require hard ramming, 103. sides of deep moiUds, 101. Vent-Wirks, size of, 102. using rods for, 2.32. Wedges, breaking of iron, 182, dimensions for iron, 184. Weights of Castings, error in figuring, 247. table for saving labor in figuring, 328. Weighting Doavn, binders for copes, 204. copes, rules for, 196, 108. horizontal set-cores, rules for, I'JS, 201, 202. vertical set-cores, 202. Wire, size for twisting, 51. used in tying brickwork, 86. Wire Kope, advantage of, for cranes, 393. for sustaining cords, 302. objections to, 202, 304. I.lial)ility of, 305. preserving of, 303, 305. Itoebling's table, etc., on strength uf, 304. sheaves and drums, size for, 302. Wheels, causes for crooked holes in, 173. WuEEL Gearing. See Gear wheels, p. 450. INDEX. 4G1 Wood, kinds used for crane framos, 30(3. relative sectional strength of, 308. table on strength of, 397. transverse strength of, 398. Wrought-ikon, melting of, in cupolas, 377. welding of, to cast-iron, 217-220. THE LARGEST FACING MILLS IN THE WORLD; CAPACITY. 650 BARRELS PER DAY. Eagle Facing Mills, MANUFACTURERS OF AND DEALERS IN ALL KINDS OF FOUNDRY FACINGS, Blackings and Foundry Supplies, PLUMBAGO OR BLACK LEAD FOR ALL PURPOSES, HEAVY MACHINERY FINE STOVE PLATE FACINGS A SPECIALTV. ALSO SHIPPERS OF THE CELEBRATED CINCINNATI MOLDING SANDS Fop Stove Plate, Heavy and Light Machinery, Agricultural and Brass "Work. AGENTS FOR MONK'S CELEBRATED MOLDERS' TOOLS. Send for Illustrated Catalogue and Price List. No charge for Samples. S. OBERMAYER FOUNDRY SUPPLY MAN'F'G CO, CINCINNATI, OHIO, U.S.A. THE MACKENZIE PATENT CUPOLA AN^^^^ SMITH & SAYRE MFg! CO., 'PropHetors,' 245 Broadway, N. Y. TliU Cupola liu.t iii.ulu a (iru.'it riv.jlutlon In miltlij^ Irun. It Ji(Ti-ra from all utlurn 111 liaviu({ u < ontinuouh ivmeuil, or id utln-r KonU, tUe blaal eiitcrM tlic fuel at all uolutn. Above oue ton capacity pur Lour, tlii-jr are made oval In form. ThH brlnKS the blast to the center of the (uriiace with the l.a-t re.iUlaiice anil -luallesl po.-,»ible amount of power, and in cuniblna- tl'iii with tl.ecoritlnuousTuyereeau>e»coniplelctrength and durability, and sought to make them as convtn- ieut for working and repairs as our '■\vn and the experience of our cii'^tomers could suggest. Steam-Heating for Bnildings ; or A Treatise on Tootlied Gearing. Hints to .'^team-Fitters. By Wm. .J. Baldwin. Price, i-2.M. Ueing a Description of Steam-Heating Appara- tus and for Warming and Ventilating Private Ilouses and Larsie Buildings, with Remarks on Steam, Water, and Air in their Relatiims to Healing ; to which are added useful miscella- neous tables. Fourth edition. With many illus- trative plates. 1-Jmo, cloth. Extracts from Chordal's Letters. limo, Cloth (nearly 4iX) p.ifresj, g-J.tiO. Dis- count of i.') per cent, on orders for Ave or more copie-i. New and enlarged edition with addi- tional llluslr.itions The name "Chordal" has become a familiar word among mechanics everywhere, for the graphic and humorous sketches of shop life and character, and the fund of information which it represents. The author's writinjis should be prized and enjoyed by everyone interested in mechanical pursuits. Steam Engine Catechism. utriT^.?' practical questions and :inswers arr.inged so .as to give a yoiini; EnL'ineer.iusi the information required to lit him tor properly riinniiiir :in en- gine. By RohertGrimshaw. Ismo, cloth, $1. Cf.ntaining complete Instructions for Design- ing, Drawing and Constructing Spur Wheels, Bevel Wheels, Lantern Gear, Screw Gear, Worms, etc.,and the proper formation of Tooth Profiles. For the use of Machinists, Pattern Makers, Draughtsmen, Designers, Scientific Schools, etc. With many plates. By J. How- ABD Cboitwell. 12mo, cloth, $3.00. A Treatise on Belts and PuUeys. Embracing full explanat ions of Fundamental Principles; proper Disposition of Pulleys; Rules for determining widths of leather and vulcanized rubber belts, and bells running over covered pulleys; Strength and Proportions of Pulleys, Drums, etc. Together with the prin- ciples and necessary rules for Rope Gearing and t ransmission of power by means of Metallic Cables. By J. Howabd Cbohwtll, Ph.B., Author of A Treatise on Toothed Gearing. (Shortly.) Pl-TRIJSnED BY JOHN WILEY & SONS, 15 Astor Place, N. Y. ,l/.s(), the ycio and Cnmi'ht'- Cntdhiijiir "f thrir Publications will he mailed to any a