^MfirL-'^H ^ EBNo, LfXJOS NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS WARTIME REPORT ORIGINALLY ISSUED August 1945 as EeBtricted Bulletin L5G03 EFFECT OF VAEXATION HT DIAMETER MD PITCH OF RIVETS ON COMEEESSIVE STREHGTH OF PANELS WITH Z-SECTION STIFFEHERS I - PANELS WITH CLCSE STIFFEHER SPACING THAT FAIL BY LOCAL BDCKLING By Norris F. Dow and WiUlam. A. Hickman Langley Memorial Aeronautical Laboratory Langley Field, Ta. WASHINGTON NACA WARTIME REPORTS are reprints of papers originally issued to provide rapid distribution of advance research results to an authorized group requiring them for the war effort. They were pre- viously held under a security status but are now unclassified. Some of these reports were not tech- nically edited. All have been reproduced without change in order to expedite general distribution. DOCUMENTS DEPARTMEWJ Digitized by the Internet Arcliive in 2011 with funding from University of Florida, George A. Smathers Libraries with support from LYRASIS and the Sloan Foundation http://www.archive.org/details/effectofvariatioOOIang NAG A RB No. L5G05 "^ NATIONAL ADVISORY COMIilTTES FOR AERONAUTICS RESTRICTED BULLETIN EPRUCT 0? VARIATION IN DIAiMETER AND PITCH OF RIVETS ON COMPRESSI'/E STRENGTH OF PAI\"ELS WITH Z-SECTION STIFPENERS I - PANELS WITH CLOSE STIFFSNER SPACING TPLAT PAIL 3Y LOCAL BUCKLING By Norrls P. Dov/ and Vifllllam A. Hickman SmiMARY An experimental Investigation is being conducted to determine the effect of varying the rivet diameter and pitch on the compressive strength of 2J4.S-T aluminum-alloy panels with longitudinal Z-section stiffeners. The panels were selected on the basis of available design charts, and the panel proportions were limited to the region of these charts in which the panels have the closest stiffener spacings and the smallest values of width-to-thickness ratio for the webs of the stiffeners and have such length that. failure is by local buckling. The results showed that for these panels the compressive strengths increased appreciably with either an increase in the diameter of the rivets or a decrease in the pitch of the rivets. Data are also presented from which the rivet diameter and pitch required to develop a given stress in the panels may be determined. INTRODl^TION The design and analysis of sheet-stiffener panels for aircraft structures have been the subject of extensive experimental and theoretical investigations, but the deter- mination of the size and pitch of rivets for attaching sheet to stiffeners is a problem that has not been ade- quately solved. In reference 1 charts and procedures are presented for the design of Z-stiffened panels to carry a given intensity of loading at a given panel length. The test data on which these design charts were based, however, were obtained for an arbitrary dLsoneter and pitch of the MCA R3 Ko. L5G05 rivets, Becaure reference 2 indicates that the local- buckling str.angth of panels having triangular stiffeners at close spacings could, be varied over a Vv'ide range by var3^ing the rivet diameter, an investigation is being con- ducted in the Langley structures research laborato\-^y of the FA-C'I to determine the effect of var7;'-ing the rivet diameter and pitch on the strength of 2l;_3~T aluminujn- alloy panels with longitudinal Z-section stiffeners of the type for which the design charts of reference 1 vvere orepared. Results are presented of the first series of tests for the investigation. The tests of reference 1 were made on approximately 7OO panels and, since any number of com- binations of rivet diameter and pitch is possible for each panel, the results of the tests on the 1I4.O panels reported hr'irein can cover only a very small region on the design char'bs of reference 1. The region so far investigated is the local-buckling region of panels having the closest stiff ener spacings and the sm-allest values of width-to- thickness ratio for the webs of the stiffeners of the panels given in reference 1. Panels of these proportions were chosen to be the first investigated because the con- nection of, stiffeners to sheet Is believed to be most critical for such panels. Further testing will be required to determine the effect of rivet size and pitch on longer panels, particularly on panels of maxim.um structural effi- ciency in which failure occurs as a combination of local buckling and bending as a column. SYMBOLS L length of specimen p radius of gyration l/p slenderness ratio W width of specimien bg spacing of stiffeners on sheet b_j^ width of attachment flange of stiffeners b;Y width of web of stiffeners bp width of outstanding flange of stiffeners NAG A RB No. L5G05 5 tg thickness of sheet t.j,j thickness of v;eb of stiffenei^s d diameter of rivets p pitch of rivets h depth of countersink for rivets €S Qj compressive yield stress for the material cff average compressive stress at failing load for any speclm.en P3_ coFxpresslve load per inch of panel width c coefficient of end-fixity in the Euler column formula TEST SPEGIRENS AND T^STKOD OF TESTING The specimens consisted of 2ij.3-T aluminum- alloy panels having Z-type stiffeners, as shown in figure 1. The stlff- eners on all panels v;ere identical. The sheet thickness v/as varied to give five selected ratios of stiff ener thickness to sheet thickness: -^ = 1.25, 1.00, 0.79, O.63, bg ^S b,., b i^ and 0.51. The proportions -r- = 25, -~- = 2C, — = 9.5j and bp ' ^S t^V % - — = O.li were chosen to give the oanels of reference 1 b-v that have the closest stiffener spaclngs and the smallest values of vifldth -to -thickness ratio for the webs of the stiffeners. The lengths of the panels were so chosen {■^= 20j that no column failures occurred. Accordingly/, the term " com.pressive strength" as aonlled herein to the panels of the present investigation is used to Indicate "the strength of the panels for local buckling failures of the general types shown in figure 2. Tests vjevii made to determine the miaterial properties of the sheet used for the construction of the specimens. The mazlmum with-grain compressive yield stress c^y obtained for the material used for the construction of the 1^ _ NAG A RB ¥,o, L5305 spsclmens was 1_;.6.0 ksi. The minirn'om value of c^j obtained 'iz/ss i;.1.5 ksi. The average value of o^^r obtained froi.i the results of tests of stress-strain specimens cut fro:ri all the sheets of material used in the construction of the panels v;aa l|-3.7 ksi. The foregoing values of a^^^ represent the with-grain properties of the flat- sheet material before forming. The rivets used thx^ougho^t the investigation were A17S-T flat-head rivets (A1t14;2aD) . Both the size and ■Ditch of the rivets wer^e- varied for each Fstio of sheet thickness to stiffener thickness- as is shovvTi in table 1. The minimum, rivet oitch used in all cases was equal to three tines the rivet diameter. On all panels the rivets were driven bj the FACA flush-riveting process in which the rivet is inser-bed with the head opposite the countj^r- sionk end of the hole, the shank of the rivet is driven into the- -cavity formed -hy the c oun-ter sink, ■ and the exces'S- material is removed with a milling tool. A c-ounterslnk angle or"60° v;as u£ed throughout. The depths of counter- sink used are given in table 1. Ultim;ate co-iiipress-ive. loads for the ll|_0 specim.ens were determined in a hydrauli-c testing machine ha-ving ajn.-accu-- rac^'- of one-liBlf of 1 .percent of -the load. The ends of the specimens were- ground flat and paraJ.lel- -before testing to ins-are an even distrihution of load over the oanel. ■c-:r'crT-rr'TC! Because the. preseni:-' investigation covers .only the local-buckling ragion of a small part of the design charts of reference 1, no attempt- is made herein to present the data by plotting the average stres-s at failing load Of against the parameter - ———, as in reference 1. Instead., _ 1/vc Of is plotted against, tha ratio of rivet -diameter to the sum of thicknesses o£ sheet and. .stiff ener -r — in t Q + ^^, figure 3 i-n order to present the results in a manner simi- lar to that used. in reference 2, numerical values of Of Pi ^ and — T— r, nowever, are listed in table 1. Figure 3 shows that, LA/c for all values of t-y/ts investigated, the ccm.pressive strengths increase-d' apr)reciably with either an increase in the diameter of the" rivets or a decrease.in the- Ditch of the rivets. HACA RB No. L5G03 Figure 3 Indicates that no substantial increase occurs in p^-nel strength for values of — - greater to+tv' than approximately 1.5' Although the vilue of 1.5 is somewhat higher than the value of 1.25 obtained in reference 2 for triangular stiffeners w: ^h the particular rivet pitches investigated therein, the carves of figure 3 show only small increases in panel stren,T 'b above values of of 1.25. O .; The rivet pitch required to develop i given stress in panels of the proportions used herein rr. ?y be determined from figure 3« Fc>r example, if it is desi-'sd to develop a stress equal to the average yield stress of the m.aterial (in this case )4.3 • 7 ksi) figure 3 indicates that at a rivet diameter I.5 times the over-all (sheet plus atiffener) thickness a rivet pitch of less than 7/0 in "/a Is required for each of the five values of t-r/tq invest.'-.rated. CONCLUDING REMi^ES Results are Dresonted of tests to determine the effect of varying rivet diam.eter and ri^ret pitch on the compressive strength of 2i|.S-T alimiinum- alloy panels with Z-section stiffeners. The panels v/ere selected on the basis of available design charts, and the panel proportions were lim.ited to the region of these charts in mhich the panels have the closest stlffener spaclngs and the smallest values of width-to-thicknsss ratio for the webs of the stiffeners and have such length that failure occur.s by local buckling. The results showed that the compressive strengths increased appreciably with either an increase in the diaraoter of the rivets or a decrease in the pitch of the rivets. Data are also presented from v/hich the rivet dlametei' and pitch required to develop a given stress in the panels may be determined. Langley Memorial Aeronautical Laboratory National Advisory CoriTinittee for Aeronautics Langley Field, Va. NAOA RB No. L5GO3 REFERENCES 1. Schuette, Evan 3.: Charts for the Mininiujri-Weight Design of 2lj.S-T Aluininurn- Alloy Flat Conipression Panels with Longitudinal Z-Ssction Stlffsners. FAG A ARR Ko . L5FI5, 19ii 5 . 2. Dow, Norris F., and Eickman, William A.: Prell:ninary Investigation of the Relation of tlie Conpressive Strength of Sheet-Stiffener Par-els to the Diameter of Rivst Used for Attaching Stiff eners to Sheet. i^AGA RB IJo. 14115. 19Ui. NACA RB No. L5G03 TAa,E 1.- ■NOMINA! DIMEWSIONS OF Z-STIFFTiMED PANELS AND TEST RESULTS SHOWING EFFECTS OP VARYING RIVET PITCH AND RIVET DIAMETER tg = 0. 051 In.; bg = 1.28 In.; L = 10.50 In.; W = 7.02 In.; ■^ - 1 25- -^ - 25- -p2 = 20 Dlam. of rivets, Depth of countersink. Pitch of rivets. Average stress at falling load. Pi l/Vc d (In.) h (In.) P (In.) CTf (ksl) (ksi) 1/16 0.655 f \%' SI 144.. OOC 8J+5.750 *57.550 5U.500 I.30U I.50U 1.095 l.Olfi. ^^ 830.550 .902 1^ ^50.175 .895 5/52 .01^0 l4i|..600 ai44.825 814.1.900 57.650 3I+.900 1.529 1.557 1.217 1.119 I.0I4.U I i 852.1+70 .961 1/8 .050 i '4. I4J4..8I1.O III;. 100 39.UOO 36.100 1.551 1.278 1.175 1.078 I i ^55.575 1.069 5/32 .060 < 15/16 i4;.950 ii5.ioo 40.750 1.551 1.308 1.208 '^ ^57.1|25 1.120 i 56.125 1.086 15/16 I4.5.900 U5.300 II2.150 l.ljl8 1.512 1.255 3A6 .060 - ^h *57.850 1.152 I ^l 856.615 1.105 [i'h. I4.5.000 U2.750 1.250 1.280 lA .060 ^l 59.400 1.178 1 i 56.600 1.060 ■Average of two tests. NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS CS 8 NACA RB No. L5G03 TABLE 1.- NOMINA! DIMENSIONS OF Z-STIFPENED PANELS AND TEST RESULTS SHOWING EFFECTS OF VARYING RIVET PITCH AND RIVST DIAMETER - Continued tg = 0. 061; In.; bg = 1.60 In.; L = IO.I4.O In.; W = 8.61; In.; § = l.OO; ^ = 25; ■r^i = 20 - Dlam. of rivets, d Depth of countersink, h Pitch of rivets, P Average stress at falling load. Of PI l/Vc (In.) (In.) (In.) (ksl) (ksl) 1/16 0.035 i 5A^ < 15/16 1+5.050 I+I.I1.5O ^58.580 1.255 1.180 1.015 1.093 ^^ 29.500 .81+0 i 26.700 .768 5/52 • Ol+O [ 'f 15/16 M+.800 1+5.500 »5B.070 aJ+0.055 1.505 1.21+5 1.069 l.ll+O 4 55.1+00 .950 1 i 50.700 .891 1/8 .050 15/1 6 I+I+.600 *l+5.755 ai+1.710 3U.750 1.517 1.227 1.186 .990 1 i 32.200 .856 5/32 .060 15/16 1+5.000 1+5. 870 1+0.500 1.518 1.197 1.1U2 -4 56.100 1.052 . ^i *55.800 .975 5/16 .065 15/16 4 U5.5U0 1+1+. 700 1+0.850 57.600 1.529 1.252 1.160 1.077 [ ^i *55.800 .968 [ilt^ |;J+.U85 I+I+.I+85 1.272 1.290 lA .065 'h 58.900 I.IOI+ i 1 1^ 55.550 1.022 Average of two tests. NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS NACA RB No. L5G03 TABLE 1.- NOMINAI DIMENSIONS OF Z-STIFFENED PANELS AND TEST RESULTS SHOWING EFFECTS OP VARYING RIVET PITCH AND RIVET DIAMETER- Continued ts = O.OBl In.; bg = 2.05 In.; L = IO.O6 in.; W = 10.77 In. 1 if = 0.7,= ^ = .5, 1^ = 20 Dlam, of rivets, d (In.) Depth of countersink, h (In.) Pitch of rivets, P (In.) Average stress at falling load, (ksl) Pi L/Vc (kal) 9/52 . I52 k2.500 I4.I.50O ^37.550 1.21(7 1.209 1.074 5/32 O.OI4.O ^36.955 1.082 lii 32 2 32.900 27.610 .96I1 .807 1/8 .060 f 5/8^ 4 1^5.700 Ii2.800 859.1120 57.550 1.286 1.255 1.150 1.090 1^ 52 51+. 820 1.013 2 29.550 .858 15/52 814.5.900 I4.5.60O i+0.655 1.273 1.266 1.161 5/52 .065 ^i 58.800 1.131 •- 1^ 52 2 3I+.75O 1.018 50.800 .S92 f VA' L5.6SO I4.O.8J+5 1.528 1.175 3/16 .075 4 Uo.ioo 1.167 lis 52 I 2 55.900 831.280 1.056 .915 1 I \% 144.. 200 *l+2.750 1.327 1.228 lA .080 { ^h 81;1.1|50 1.217 1^9 52 57.200 1.089 2 53.000 .96U Average of two tests, NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS 10 NACA RB No. L5G03 TABLE 1.- NOMINAL DIMENSIONS OF Z-STTFFENED PANELS AND TEST RESULTS SHOWING EFFECTS OF VARYING RIVET PITCH AND RIVET DIAMETER - Continued Average of two tests. ■'Average of three tests. ts = 0- 102 In.; bg = 2.' ?5 in.; L = 9.I+I4. In.; W = 13.39 In. ; y = 0. ^S '3- ^ - 25- ^5. ts - ^^' |5 = 20 Dlam. of rivets, d (In.) Depth of countersink, h (In.) Pitch of rivets, P (in.) Average stress at falling load. Of (ksl) PI l/Vc (ksl) f 9/32 7/8 P7 1+2.300 39.300 38.170 1.1+12 1.288 1.218 ■ 3/32 0.050 4 35.1+00 1.158 ^^ 31+.500 1.129 2 30.000 .981+ I4.3.800 i(.0.J4.00 39.700 I.I4I+5 1.321 1.263 1/8 .060 ' ^i 37.800 1.237 1^9 32 55.500 1.167 2 3C.2I4.O .981+ 15/32 7/8 al+3.590 ak2.335 1+1.050 l.li3l 1.588 1.310 5/32 .070 4 57.850 1.236 * I 2 55.750 31.800 1.168 1.01^9 [ '.%' ^1+5.150 1^1+1.150 1.1+51 1.327 3/16 .080 * 38.800 1.263 1^ 32 38.150 1.253 I 2 51.900 l.Oi+2 [ '7;^ l+i+.050 ^1+5. 000 1.1^71 1.378 lA .090 ^^ 1+0.700 1.529 ^H 39.800 1.307 2 3I+.100 1.120 NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS NACA RB No. L5G03 11 TABLE 1.- NOMINAL DIMENSIONS OF Z-STIFPENED PANELS AND TEST RESULTS SHOWING EFFECTS OF VARYING RIVET PITCH AND RIVET DIAMETER -Ccncluded tg = 0.125 in.; bg = 3.15 in.; L = T- = 0, 51; T- = 25; 8.70 In.; W = 16.27 ^ on T- = 20 In.; Dlam. of rivets, d (In.) Depth of countersink, h (In.) Pitch of rivets, P (In.) Average stress at falling load, af (ksl) Pi L/Vc (ksl) 1/8 5/52 5/16 lA 0.070 .080 .090 .100 5/8 15/52 7/8 ii2 32 2 1^ ^32 2 ,1S U2.000 Ii.1.200 39.190 38.050 ^57.1+25 35.600 14.1.6CO kl.350 lio.150 ^38.805 56.200 31;. 500 1;2.700 li.1.180 39.U50 37.500 3U.900 al|.2 . 600 112.200 l4.1.1|.00 38.560 ^37.150 1.595 1.563 I.U63 l.lil].0 1.U09 1.260 1.595 1.558 1.U89 1.1l75 1.360 1.296 1.655 1.538 1.U88 i.Uio 1.515 1.609 1.58I+ 1.566 i.lv52 1.596 Average of two tests, NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS Fig. NACA RB No. L5G03 I I I s NACA RB No. L5G03 Fig. 2 tw +J cd CD C 0) E •H CJ & w +J CD CU Hi Fig. NACA RB No. L5G03 ^ Ib^ I Si £.'^ F/gure 3. — l/ar/at/'on in /oca/ buc/