/ % ,--i>:*< a* «v^?*/ V'—"> < V^\/ V •"••• a* ** 
 
 /.•afcX .0"*.iaat%.. y.-^itX .0»*.5^;.%.. >*,•-■ 
 
 •bv* 
 
 ♦♦*% ■ 
 
 a, ^ ♦'TV. 1 ,6 V ^o 
 
 
 "b^ 
 
 
 
 
 ^ 
 # %/'^« # ^ ^ %* 
 
 
 X *^a^ <^ 
 ^ <^ 'Asm,. <^ ^ *SlS^ a ^ «^ ••i 
 > *v *^s •iclfgf* «^ ^, ovJIX^* a^ "^ • 
 
 A> - « - , ^- 
 
 
 
 > ^ '.' 
 
 •• ^ 
 
 
 4V ^» 
 
 a °<v : 
 
 
 
 
 r oV 
 
 
 ^7^ 
 
 - a v *^ : 
 
 G v ^ -i".7* A 
 
 
 HOft 
 
 
 
 4* ^ 
 
 <5°* 
 
 
 
 ^^ 
 
 
 
 r > 
 
 v t ^k* ^ aP 
 
 ^o x 
 
 
 0^ t «^f*/ 
 
 
 sP-nK 
 
 ^°^ v 
 
 ^ 
 
> ^ ,V si,/*???* o- 
 
 A c o « o „ V 
 
 A« a-*, o»»- ,A' ->-j. -v . s - r.- 
 
 
 
 
 •*wE- /\ °Wws /\ \llf** /\ ^™* : *?% llpy «/% 
 
 •^t.v .***£&>* ,s,<j^-:«+' \°*,^k-\ &£&&. v,^. 
 
 &*■ n o » o 
 
 
 c5°^ 
 
 i0v\ 
 
 
 A°* 
 
 N 
 
 
 
 •^••> 4 /.-^<X c°.^^,-> ,/.-^,t\ *°.i^: ^ / -^i- 
 
 
 
 
 
 
BUREAU OF MINES 
 INFORMATION CIRCULAR/1988 
 
 it 
 
 Rollover Protective Structure (ROPS) 
 Performance Criteria for Large 
 Mobile Mining Equipment 
 
 By Stephen A. Swan 
 
 UNITED STATES DEPARTMENT OF THE INTERIOR 
 
Information Circular 9209 
 
 Rollover Protective Structure (ROPS) 
 Performance Criteria for Large 
 Mobile Mining Equipment 
 
 By Stephen A. Swan 
 
 UNITED STATES DEPARTMENT OF THE INTERIOR 
 Donald Paul Hodel, Secretary 
 
 BUREAU OF MINES 
 T S Ary, Director 
 

 * 
 
 Library of Congress Cataloging-in-Publication Data 
 
 Swan, Stephen A. 
 
 Rollover protective structure (ROPS) performance criteria for large 
 mobile mining equipment. 
 
 (Information circular / United States Department of the Interior, Bureau of Mines ;9209) 
 
 Includes bibliographical references. 
 
 Supt. of Docs, no.: I 28.27:9209 
 
 1 . Mining machinery — Rollover protective structures. 2. Loaders (Machines) — Safety appli- 
 ances. I. Title. II. Series: Information circular (United States. Bureau of Mines) ;9209. 
 
 TN295.U4 [TN345] 622s [622'. 8] 88-600283 
 
CONTENTS 
 
 Page 
 
 Abstract 1 
 
 Introduction 2 
 
 Test procedures 2 
 
 Roll hill characteristics 2 
 
 Roll hill slope and vertical drop height 2 
 
 Roll hill length 3 
 
 Roll test procedures 3 
 
 Method of release 3 
 
 Launch pad 3 
 
 Bucket position 4 
 
 Articulation joint 4 
 
 Instrumentation and measurement 
 
 parameters 4 
 
 Data link and sensor types 4 
 
 Strain 4 
 
 Deflection 5 
 
 Acceleration 5 
 
 Roll rate 5 
 
 Time 5 
 
 Page 
 
 Tests with 52,000-lb-GVW, wheeled, 
 
 front-end loader 5 
 
 Roll test 1 5 
 
 Roll test 2 6 
 
 Tests with 390,000-lb-GVW, wheeled, 
 
 front-end loader 8 
 
 Roll test 3 8 
 
 Roll test 4 9 
 
 Static test of four-post ROPS 10 
 
 Test with 286,000-lb-GVW, wheeled, 
 
 front-end loader 13 
 
 Conclusions 14 
 
 ILLUSTRATIONS 
 
 1. Equivalent rollover, 720° 3 
 
 2. Tilt table 4 
 
 3. Instrumentation schematic 5 
 
 4. ROPS before and after rollover, 52,000-lb-GVW, wheeled, front-end loader 7 
 
 5. First 390,000-lb-GVW, wheeled, front-end loader after rollover 9 
 
 6. Mount failure, 390,000-lb-GVW, wheeled front-end loader 9 
 
 7. Second 390,000-lb-GVW, wheeled front-end loader before and after rollover 10 
 
 8. Static test side load deflection curve, 390,000-lb-GVW, wheeled, front-end loader 11 
 
 9. Static test vertical deflection curve, 390,000-lb-GVW, wheeled, front-end loader 11 
 
 10. Static test longitudinal load deflection curve, 390,000-lb-GVW, wheeled front-end loader 12 
 
 11. Roll test 5, 286,000-lb-GVW, wheeled front-end loader 13 
 
 TABLES 
 
 1. Parameter measuring sensors 4 
 
 2. First roll test summary 6 
 
 3. Second roll test summary 6 
 
 4. Third roll test summary 8 
 
 5. Fourth roll test summary 10 
 
 6. ROPS static test side and longitudinal load energy 12 
 
 7. Fifth roll test summary 14 
 

 UNIT OF MEASURE ABBREVIATIONS USED IN THIS REPORT 
 
 °/s 
 
 degree per second lb pound 
 
 ft 
 
 foot pet percent 
 
 g 
 
 acceleration of gravity psi pound per square inch 
 
 in 
 
 inch s second 
 
 in'lb 
 
 inch-pound 
 
ROLLOVER PROTECTIVE STRUCTURE (ROPS) PERFORMANCE CRITERIA 
 FOR LARGE MOBILE MINING EQUIPMENT 
 
 By Stephen A. Swan 
 
 ABSTRACT 
 
 Certain types of mobile surface mining equipment are required to be equipped with 
 rollover protective structures (ROPS) capable of protecting the operator in case of a rollover 
 accident. The Bureau of Mines has developed ROPS performance criteria for wheeled 
 front-end loaders. This report reviews the development of rollover test procedures and 
 subsequent testing of ROPS for 52,000-, 286,000-, and 390,000-lb gross vehicle weight 
 (GVW) loaders. The program consisted of both dynamic (roll) and static tests to assist in the 
 development of the ROPS performance criteria. Each test included extensive instrumentation 
 and photographic documentation to fully evaluate the performance of the ROPS. Data 
 summarizing side, longitudinal, and vertical force and deflections are included. The data 
 indicate that for loaders in excess of 240,000 lb, GVW, a side force-to-mass ratio of 1.0 and 
 an energy-to-mass ratio of 6.0 appear to be adequate criteria to protect an operator in the 
 event of a 540° rollover. For smaller machines, Society of Automotive Engineers (SAE) J 1040 
 performance criteria provide adequate protection. Longitudinal loading, though not ad- 
 dressed in previous ROPS performance criteria, was shown to be a potentially significant 
 factor in ROPS failure. Longitudinal loading and energy criteria equal to 80 pet of side 
 loading criteria are recommended. 
 
 1 Mining engineer, T\vin Cities Research Center, Bureau of Mines, Minneapolis, MN. 
 
INTRODUCTION 
 
 Mine Safety and Health Administration (MSHA) and 
 Occupational Safety and Health Administration (OSHA) reg- 
 ulations require that certain types of mining, construction, 
 earthmoving, agricultural, and forestry equipment be 
 equipped with a rollover protective structure (ROPS). The 
 types of mobile machines that must be equipped with ROPS 
 include crawler tractors and crawler loaders, motor graders, 
 wheeled loaders and tractors, skid-steer loaders, and the 
 tractor portion of tractor-scrapers. Although not mandated by 
 regulation, ROPS are sometimes installed on off- highway 
 haulage trucks and water trucks. 
 
 The requirement of ROPS regulations in the United 
 States, Canada, and other countries is due to the significant 
 injury potential represented by accidental rollovers of mobile 
 equipment during field use. The use of ROPS and seatbelts on 
 these types of equipment can greatly reduce the number and 
 severity of injuries resulting from rollover accidents. Although 
 the regulations require that the employer who owns the 
 machine equip it with ROPS to provide a safer work environ- 
 ment for employees, manufacturers of the machines typically 
 install ROPS on the machines before ownership is transferred 
 to the employer. 
 
 The Society of Automotive Engineers (SAE), through 
 Subcommittee 12 — Off Road Machinery Technical Committee 
 (ORMTC), develops ROPS structural performance and test 
 method criteria for use by industry in the design and perform- 
 ance certification of ROPS used on construction and mining 
 machines. The ROPS regulations promulgated by OSHA and 
 other regulatory bodies base ROPS structural performance 
 capability on criteria developed by SAE. 
 
 In 1972, ORMTC began development of SAE Recom- 
 mended Practice J 1040 "Performance Criteria for Roll Over 
 Protective Structures (ROPS) for Construction, Earthmoving, 
 Forestry, and Mining Machines." However, because of the 
 
 absence of rollover test data, the force-energy criteria for 
 machines exceeding 132,000 lb gross vehicle weight (GVW) 
 were projected based on the collective judgment of the com- 
 mittee. The committee reasoned that ROPS force-energy cri- 
 teria, expressed as a function of total machine size and weight, 
 could be less for large machines than for small machines. The 
 lower requirement was possible because the operator compart- 
 ment on a larger machine is smaller, in proportion to total 
 machine size, than for a small machine. As a result, when a 
 large machine rolls, a greater proportion of the energy gener- 
 ated by the roll would be absorbed by the machine body, tires, 
 frame, etc., than would be the case for a small machine. 
 
 In 1977, rollover test data were produced indicating that 
 the force and energy requirements contained in the SAE J 1040 
 recommended practice were not sufficient for large machines. 
 A machine manufacturer performed ROPS rollover tests on a 
 front-end loader and crawler tractor, each weighing approxi- 
 mately 200,000 lb. Although both ROPS exceeded SAE J 1040 
 performance criteria, neither survived the roll test. SAE sub- 
 sequently informed OSHA that the validity of the force and 
 energy requirements delineated in J 1040 was in doubt and 
 recommended that new guidelines be developed for larger 
 construction and mining machines. An interim guideline, SAE 
 J1040c, was established by ORMTC, however it, like its 
 predecessor, was based on subjective judgments and extrapo- 
 lation from available test data. As larger and heavier front-end 
 loaders were introduced (up to 390,000 lb), the adequacy of 
 these judgments and extrapolations was seriously questioned. 
 This report documents research to establish ROPS perform- 
 ance criteria for large front-end wheeled loaders through a 
 systematic program of dynamic (roll) and static testing. The 
 test program involved trials with loaders weighing 52,000, 
 286,000, and 390,000 lb. The research was performed by 
 Woodward Associates, Inc., under contract to the Bureau. 2 
 
 TEST PROCEDURES 
 
 Because the objective of the program was the establish- 
 ment of ROPS performance criteria for large front-end 
 wheeled loaders without reliance on subjective judgments and 
 extrapolation from tests of smaller machines, emphasis was 
 placed on full-scale testing of ROPS under realistic, yet 
 controlled and reproducible conditions. Roll testing, wherein a 
 host machine equipped with a fully instrumented ROPS is 
 caused to roll down an embankment, is the most realistic test 
 possible, as it nearly duplicates field conditions. However, not 
 all test variables can be easily controlled. Thus, "identical" 
 rolls may produce a range of data values that require consid- 
 erable interpretation to provide meaningful results. Static 
 testing, wherein the ROPS is installed in a load frame that 
 applies measured loads of known magnitude and direction, 
 produces results which are more reproducible and there is less 
 uncertainty in the data. However, the magnitude and direction 
 of the loads that must be applied in order to predict ROPS 
 performance under field conditions are not always known. 
 Roll testing is also significantly more costly than static testing. 
 
 As a result of these considerations, a test plan was devised 
 that incorporated both roll testing and static testing. ROPS 
 performance criteria would be based on the combined results 
 of the two procedures. Depending on the degree of correlation 
 between results of the roll and static testing, it was also hoped 
 that greater reliance could be placed on static testing in the 
 
 future, perhaps in conjunction with computer simulation, as a 
 way of reducing the cost of assessing the performance capa- 
 bility of ROPS structures. 
 
 The following sections describe the characteristics of the 
 test fixtures and equipment and the procedures utilized during 
 the test program. 
 
 ROLL HILL CHARACTERISTICS 
 
 Roll Hill Slope and Vertical Drop Height 
 
 Two considerations govern selection of roll hill slope: side 
 loading of the ROPS at initial impact and continuation of the 
 roll to achieve at least 360° machine rotation. The critical 
 design load for a ROPS is generally in the side direction 
 because of high bending stresses in the support members. 
 Thus, shallower slopes that maximize side loading of the 
 ROPS at initial impact are favored by some equipment man- 
 ufacturers. SAE J 1040c stipulates a 30° maximum slope to 
 
 2 Dahle, J. L., and G. R. Gavan. Development of Rollover Protective 
 Structures (ROPS) Performance Criteria for Large Mobile Mining Equipment 
 (contract H0292020, Woodward Associates, Inc.). BuMines OFR 57-86, 1985, 
 277 pp.; NTIS PB 86-216066. 
 
insure high initial side loads. However, if the slope is too 
 shallow, the machine simply tips over and slides down the 
 slope without rolling. The manufacturers with the most roll 
 testing experience favored 30° to 40° slopes because they 
 found the machines would reliably roll only on steeper slopes. 
 It was concluded that a compromise roll hill slope of 35° 
 would produce acceptable initial side loading while insuring at 
 least 360° roll rotation. 
 
 Another test parameter related to roll hill slope is the 
 vertical drop height from the launch pad to the slope. Low 
 vertical drop heights, like shallow roll hill slopes, introduce 
 high initial side loads, which is desirable. However, higher 
 drop heights induce greater angular momentum in the ma- 
 chine, which is necessary to cause rotation and successful 
 rolling. Analysis indicated a 30-in vertical drop height would 
 produce the desired results. 
 
 Roll Hill Length 
 
 There was general agreement among the manufacturers 
 consulted that the length of the roll hill should be based on two 
 factors — the perimeter of the machine to be rolled and the 
 number of expected rolls. It was recommended that an addi- 
 tional 30 pet of hill length would be required to account for 
 any machine slippage during the roll. Therefore, it was deter- 
 mined that a length of 120 ft would accommodate a minimum 
 of 720° of roll from the smallest to the largest machine to be 
 tested (fig. 1). 
 
 ROLL-TEST PROCEDURES 
 
 All manufacturers consulted agreed that the machine 
 should be positioned at the top of the slope with the outermost 
 wheels at the edge of the slope. The machine would then be 
 tipped to the side to initiate roll down the hill. This method, as 
 opposed to the machine having a forward motion when tipped 
 over the edge of the hill, would provide for the best repeat- 
 ability of test conditions. 
 
 Method of Release 
 
 The machine was positioned at the top edge of the slope 
 on a tilt table. The side of the platform away outboard from 
 the edge of the roll hill was elevated (fig. 2) and the machine 
 rolled off the platform when the center of gravity passed 
 beyond the inboard tires. The advantage of using a tilt table, in 
 addition to providing repeatable release conditions, was that it 
 could be reused for other testing. 
 
 Launch Pad 
 
 A launch pad for the machines was necessary in order to 
 provide the method of release as described above. The inboard 
 side of the tilt table had to be pinned to the structure in order 
 to rotate. In addition, as the machine is being tipped all of its 
 
 Roll initiation 
 
 Initial side load contact of ROPS first impact 
 
 Second side load contact 
 
 Final contact of ROPS 
 
 Figure 1. — Equivalent rollover, 720* 
 
Outboard 
 
 Lift point 
 
 Inboard 
 .: tire 
 
 no,.- 
 
 
 °°:ol 
 
 
 & 
 
 OflpV 
 
 
 
 40' 
 
 Concrete 
 pad 
 
 Figure 2. — Tilt table. 
 
 weight is on the two inboard tires. A launch pad or footing was 
 necessary to prevent the edge of the hill from slipping or falling 
 away (fig. 2). 
 
 Bucket Position 
 
 During all tests the bucket was locked and welded in the 
 carry position as defined by SAE J732c. This definition 
 requires the bucket to be tipped backward to a 15° approach 
 angle. 
 
 Articulation Joint 
 
 The articulation on all three machines was locked by the 
 transportation locking bars and additional supports were 
 welded in place. Again this was to provide for a repeatable test. 
 
 the best data link. The recording system and components were 
 based on the use of this method. 
 
 Once the measurement parameters were determined, as 
 described in the following sections, it was possible to select the 
 type of sensors necessary to meet the program objectives. The 
 sensors selected to measure the parameters are listed in table 1 . 
 
 Table 1.— Parameter measuring sensors 
 
 Instrumentation 
 
 Accelerometer 
 
 Potentiometer 
 
 Rate gyroscope 
 
 Strain gauge 
 
 Time generator 
 
 Umb 
 
 ilical 
 
 Front 
 
 Rear 
 
 3 
 
 3 
 
 3 
 
 3 
 
 1 
 
 1 
 
 10 
 
 10 
 
 1 
 
 1 
 
 Channels 
 
 6 
 6 
 2 
 20 
 2 
 
 INSTRUMENTATION AND MEASUREMENT 
 PARAMETERS 
 
 Data Link and Sensor Types 
 
 The type of data link was an important factor in estab- 
 lishing the other components of the instrumentation system. It 
 was necessary to define the data link early in the planning of 
 the instrumentation system (fig. 3). The types considered were 
 the umbilical (hard-line) type or a telemetry system with 
 AM-FM transmission. The umbilical system was selected as 
 
 Strain 
 
 Strain was considered to be the most important parameter. 
 Of the 36 channels to be recorded, 20 were strain measurement. 
 
 Some structural columns were instrumented with strain 
 gauges on opposite sides of the column. These opposing 
 gauges were wired into a single bridge, which permitted the 
 measurement of bending with a single output. This method 
 maximized the information gained from a smaller number of 
 recorded channels. It was necessary to locate the gauges in 
 areas that would not be subjected to local yielding. All strain 
 
Accelerometer 
 (3 each) 
 
 Potentiometer 
 (3 each) 
 
 Rate gyro 
 (I each) 
 
 Strain gauge 
 (10 each) 
 
 o 
 
 | Bridge 
 
 
 Umbilical 
 
 completion 
 
 Oscillograph 
 
 To camera -^, 
 timing * ^> 
 circuits 
 
 Time 
 generator 
 
 Figure 3. — Instrumentation schematic. 
 
 gauges were wired with a three-wire hookup, and precision 
 bridge-completion resistors were located on the machine near 
 the strain gauge locations. All gauges were 1-in uniaxial 
 gauges. 
 
 Deflection 
 
 Six channels were utilized to measure deflection of par- 
 ticular locations on the ROPS. The measurements were made 
 with resistive, linear position transducers with a 20-in stroke. 
 The diagonal spans across the ROPS columns were measured 
 in three planes to resolve the relative motions of the structure. 
 
 Acceleration 
 
 Six channels of acceleration were measured and recorded. 
 The accelerometers were mounted in a triaxial configuration at 
 two locations on the machine. The accelerometers had a range 
 of ±20 g. 
 
 Roll Rate 
 
 This measurement was a direct result of the manufactur- 
 er's survey information. One manufacturer was recording roll 
 rates routinely on all roll tests to provide correlation from one 
 test to the next. Roll rate is used to determine differences in roll 
 test results. Acceleration measurements can also be used in this 
 manner, but roll rate is a more positive comparison method. 
 Two channels were utilized to record roll rate. 
 
 Time 
 
 Each oscillograph recorder was equipped with an internal 
 time base that was recorded on each individual record. In 
 addition, a time base generator for the cameras was recorded 
 on the film edge, as well as on the oscillographic records. 
 Flashbulbs in the view of all cameras were activated at the 
 beginning of the roll to correlate a starting time. 
 
 TESTS WITH 52,000-lb-GVW, WHEELED, FRONT-END LOADER 
 
 ROLL TEST 1 
 
 The rollover test of a 52,000-lb-GVW, wheeled, front-end 
 loader equipped with a four-post ROPS that had a 106- 
 pct-GVW (55,120-lb) side load capability successfully met the 
 objectives of the roll test. The combination of the tilt table and 
 the relative location of the launch pad and the roll hill 
 provided an excellent method for roll initiation, which resulted 
 in a relatively high side load on initial impact. The ROPS was 
 subjected to three impacts during the test. Two impacts 
 occurred on the roll hill and one impact occurred on the bench 
 at the base of the hill. 
 
 Table 2 summarizes the data obtained from the test. The 
 data indicate that the structure was subjected to successively 
 greater loads with each impact. The strain gauge data indicate 
 vertical loadings of approximately 200 to 400 pet GVW, 300 to 
 500 pet GVW, and 700 pet GVW for the first, second, and 
 bench impacts, respectively. The vertical loading estimated 
 
 from the soil resistance and accelerometers measurements 
 indicates even higher vertical loads. 
 
 The strain gauge data indicate a side load of 58,000 and 
 60,800 lb for the first and the second impacts. The side load 
 could not be determined for the bench impact. The ROPS had 
 a side load capability of approximately 55,100 lb, as deter- 
 mined by a static test conducted by the manufacturers of a 
 similar unit. The measured side loads were thus higher than the 
 estimated capacity of the ROPS. The discrepancies are prob- 
 ably caused by measurement inaccuracies during the static and 
 roll tests, and differences in the cross-sectional geometry and 
 yield strengths of the two ROPS. The side loads for the first 
 two impacts determined from the side deflections of the ROPS 
 were approximately 50,800 and 52,800 lb. 
 
 Estimated vertical and side loads, and estimated side 
 deflection, are also given in table 2. These estimates are based 
 on the data obtained from the test, considerations of the 
 reliability and accuracy of the particular measurements, the 
 
capacity of the mounting bolts, and the observations of the 
 high-speed film. 
 
 The evaluation of the data obtained from the roll test 
 indicated that the ROPS experienced vertical and side loads 
 that exceeded the requirements of SAE J1040c. For compari- 
 son, SAE J 1040c requires the following loads and deflections 
 during static testing for the machine and ROPS that were 
 subjected to the roll test: Side load, 73 pet GVW; side 
 deflection, 13 in; and vertical load, 200 pet GVW. 
 
 The static test required by SAE J 1040c is to provide 
 protection for a 360° roll on a 30° slope. The ROPS that was 
 subjected to the roll test meets these requirements. 
 
 From the data and review of the impacts of the test, major 
 structural damage to the ROPS appears to have occurred as a 
 result of the third (bench) impact. The damage included tensile 
 failure of the two 7/8-in, grade 8 mounting bolts at each rear 
 post of the ROPS; tensile failure of the three 3/4-in, grade 8 
 mounting bolts at the right-front post (the three 3/4-in bolts of 
 the left-front post were not fractured); buckling of the rear 
 plate (between the two rear posts) in the area where an access 
 hole was cut; fracture of the top front crossmember and top 
 plate; yielding of the complete cross section at the top of all 
 four posts just below the gussets; and weld failure at the right 
 mounting bracket for the instrument panel-ROPS attachment. 
 Figure 4 shows the ROPS configuration before and after the 
 roll test. 
 
 Table 2.— First roll test summary 
 
 (52,000-lb-GVW, wheeled, front-end loader) 
 
 1st impact 2d impact Bench impact 
 
 MEASURED 
 Vertical load, 10 3 lb: 
 
 ROPS strain gauges 
 
 Soil resistance 
 
 Acceleration 
 
 Side load, 10 3 lb: 
 
 ROPS strain gauges 
 
 Soil resistance 
 
 ROPS deflection 1 
 
 Longitudinal load, 10 3 lb: 
 
 ROPS strain gauges 
 
 Side deflection, 2 in: 
 
 Potentiometers 
 
 Physical measurement.. 
 Roll rate, s °/s: 
 
 Accelerometers 
 
 Gyroscopes 
 
 Ground resistance, psi: 
 
 Penetrometer 
 
 Max ROPS penetration, in: 
 
 Field measurement 
 
 ESTIMATED 
 Vertical load: 
 
 Force 10 3 lb.... 
 
 Force pet GVW.... 
 
 Side load: 
 
 Force 10 3 lb.... 
 
 Force pet GVW.... 
 
 Side deflection in.... 
 
 115-203 
 
 159 
 
 208-260 
 
 58 
 
 40 
 
 50.8 
 
 15.14 
 
 3 4.43 
 3.25 
 
 75 
 106 
 
 142 
 
 24 
 
 160 
 310 
 
 51 
 98 
 4.4 
 
 147-237 
 
 192 
 
 156-312 
 
 60.8 
 44.3 
 52.8 
 
 9.78 
 
 377 
 
 3,400 
 
 520 
 
 ND 
 ND 
 ND 
 
 ND 
 
 5.25 
 
 3 4 >24 
 
 3.80 
 
 >24 
 
 90 
 
 160 
 
 120 
 
 325 
 
 157 
 
 3,000 
 
 19 
 
 ND 
 
 210 
 
 >377 
 
 400 
 
 730 
 
 53 
 
 >55 
 
 102 
 
 106 
 
 5.3 
 
 24 
 
 The operators platform to which the ROPS was mounted 
 was damaged only in the area of the mounting bolt holes. 
 Other machine damage was minimal. However, the left-rear 
 tire rim seal was damaged and loss of pressure occurred. 
 
 As indicated by the ground resistance and ROPS penetra- 
 tion data in table 1 , the roll hill was very "soft" and the bench 
 was extremely compacted. 
 
 ROLL TEST 2 
 
 The second rollover test of the 52,000-lb, wheeled, front- 
 end loader and a four-post ROPS with a 120-pct-GVW side 
 load capability successfully met the objectives of the roll test. 
 Similar to the first test, the ROPS was subjected to two 
 impacts during the test on the roll hill. The ROPS mounts did 
 not fail during the test. 
 
 Table 3 summarizes the data obtained from the second 
 test. The data indicate that the ROPS was subjected to 
 successively greater loads with each impact. The strain gauge 
 data indicated vertical loadings of approximately 170 and 180 
 pet GVW for the first and second impact, respectively. The 
 vertical loadings estimated from the accelerometer measure- 
 ments for each impact indicated vertical loads of 1.4 and 4.2 
 pet GVW, respectively. Table 3 also contains estimated vertical 
 and side loads, and estimated side deflection, based on data 
 collected during the test. 
 
 The strain gauge data indicated a side load of 57,700 lb 
 (1 10 pet GVW) and 62,800 lb (120 pet GVW) for the first and 
 second impacts. The measured side load during the second 
 impact is higher than the estimated capacity of the ROPS. The 
 structure had a side capability of approximately 120 pet GVW 
 
 Table 3.— Second roll test summary 
 
 (52,000-lb-GVW, wheeled, front-end loader) 
 
 1st impact 
 
 2d impact 
 
 ND Not determined. 
 
 1 Static test curve. 
 
 2 Plastic 3d impact method. 
 
 3 Maximum. 
 
 4 Estimated, sensors reached maximum at 12 pet of full scale. 
 
 5 Measured using average strain gauges and instantaneous pictures. 
 
 MEASURED 
 Vertical load, 10 3 lb: 
 
 ROPS strain gauges 
 
 Accelerometer 
 
 Side load, 10 3 lb: 
 
 ROPS strain gauges 
 
 ROPS deflection 
 
 Longitudinal load, 10 3 lb: 
 
 ROPS strain gauges 
 
 Roll rate, °/s: 
 Before impact: 
 
 Gyroscope 
 
 Film 
 
 After impact: 
 
 Gyroscope 
 
 Film 
 
 Max ROPS penetration, in: 
 
 Field measurement 
 
 ESTIMATED 
 Vertical load: 
 
 Force 10 3 lb.... 
 
 Force pet GVW.... 
 
 Side load: 
 
 Force 10 3 lb.... 
 
 Force pet GVW... . 
 
 Side deflection in.... 
 
 ND Not determined. 
 1 Off scale. 
 
 87.45 
 72.8 
 
 57.7 
 49 
 
 10.55 
 
 102 
 113 
 
 75 
 87 
 
 18 
 
 87.5 
 170 
 
 57.7 
 
 110 
 
 >4.5 
 
 94 
 218 
 
 62.8 
 ND 
 
 O 
 318 
 
 C) 
 273 
 
 16 
 
 94 
 180 
 
 62.8 
 120 
 >13 
 
as determined by the static test results of a similar unit, 
 corrected by the material strength allowables established after 
 the roll test. This estimate is also based on the fact that the 
 mounting bolts that failed during the static test were replaced 
 with standard production bolts. 
 
 The side load for the first impact as determined from the 
 side deflections of the structure was approximately 49,000 lb. 
 This side force was determined from the deflections measured 
 from the high-speed film. The displacement transducers ap- 
 peared to give inaccurate deflection measurements (approxi- 
 mately 1 in) because some of the linear potentiometer mount- 
 ing studs were bent. Therefore, the side forces determined from 
 the deflections are not as reliable as those determined from the 
 strain gauge data. 
 
 Changing the roll hill penetration resistance from 150 psi 
 for the first test to 1 ,800 psi for the second test appears to have 
 affected the dynamics of the roll more than any other factor. It 
 was expected that an increase in penetration resistance of the 
 hill would have caused an increase in the vertical loads on the 
 structure. In fact, however the vertical loads on the harder hill 
 were only about one-half the loads that resulted from rolling 
 the machine on the softer hill. The harder hill allowed the 
 machine to gain angular momentum from the first to second 
 impact thereby reducing the vertical loads. The angular mo- 
 mentum entering the second impact was 2.8 times greater on 
 the harder hill than on the softer hill. This was observed by 
 comparing the high-speed film of the second impact for both 
 tests. 
 
 
 Figure 4. — ROPS before (top) and after (bottom) rollover, 52,000-lb-GVW, wheeled, front-end loader. 
 
TESTS WITH 390,000-lb-GVW, WHEELED, FRONT-END LOADER 
 
 ROLL TEST 3 
 
 The first rollover test of the 390,000-lb-GVW, wheeled, 
 front-end loader with a four-post ROPS met the objectives of 
 the test. The side of the machine contacted the roll hill before 
 the ROPS did, which introduced maximum side loading of the 
 structure. The machine rolled two complete revolutions on the 
 roll hill, subjecting the structure to two impacts, and landed 
 upright at the base of the hill. The data needed to analyze the 
 behavior of the ROPS during the roll test were obtained and a 
 summary of the results is given in table 4. 
 
 The results indicate that the first impact imposed approx- 
 imately a 110-pct-GVW side load (448,000 lb) on the ROPS. 
 This is supported by the strain gauge data and also from the 
 measured deflections. The strain gauge data show an equal 
 distribution of the side load on the front and rear of the 
 structure. A load-deflection curve generated from the com- 
 puter analysis provided a means to correlate side deflection to 
 applied load. 
 
 The strain gauge data also show approximately a 190- 
 pct-GVW vertical load (727,000 lb) for the first impact. The 
 accelerometer data show a 4.9-g (1,911,000-lb) loading. This 
 loading is predictably higher because only part of the machine 
 weight contributed to the vertical loading. In this test the 
 reduced vertical loading is mainly due to the top corner of the 
 machine's frame absorbing part of the vertical force because it 
 impacted the roll hill first and was still in contact with the 
 ground when the ROPS hit. 
 
 The longitudinal load during the first impact, as deter- 
 mined from the strain gauge data, indicates about a 30- 
 pct-GVW rear load (130,000 lb), consistent with previous tests. 
 
 Table 4.— Third roll test summary 
 
 (390,000-lb-GVW, wheeled, front-end loader) 
 
 1st impact 
 
 2d impact 
 
 MEASURED 
 Vertical load, 10 3 lb: 
 
 ROPS strain gauges 
 
 Accelerometer 
 
 Side load, 10 3 lb: 
 
 ROPS strain gauges 
 
 ROPS deflection 1 
 
 Longitudinal load, 10 3 lb: 
 
 ROPS strain gauges 
 
 ROPS mount failure 
 
 Roll rate, °/s: 
 
 Before impact: Gyroscope 
 
 After impact: Gyroscope 
 
 Max ROPS penetration, in: 
 
 Field measurement 
 
 ESTIMATED 
 Vertical load: 
 
 Force 10 3 lb.... 
 
 Force pet GVW.... 
 
 Side load: 
 
 Force 10 3 lb.... 
 
 Force pet GVW.... 
 
 Longitudinal load: 
 
 Force 10 3 lb.... 
 
 Force pet GVW.... 
 
 1 Computer load deflection curve. 
 
 727 
 
 1,014 
 
 1,911 
 
 1,755 
 
 448 
 
 398 
 
 429 
 
 526 
 
 58 
 
 390 
 
 ND 
 
 367 
 
 108 
 
 204 
 
 42 
 
 73 
 
 47 
 
 70 
 
 727 
 
 1,028 
 
 186 
 
 264 
 
 448 
 
 390 
 
 115 
 
 100 
 
 130 
 
 390 
 
 30 
 
 100 
 
 Between the first and second impacts the machine devel- 
 oped an angle of about 45° to the vertical centerline of the hill, 
 measured from the impact impressions on the hill. Thus, when 
 the ROPS impacted the hill it was subjected to equal longitu- 
 dinal and side loads, which is confirmed by the strain gauge 
 data. The longitudinal and vertical loads were at approxi- 
 mately 100-pct-GVW (390,000 lb). The vertical load was 
 determined to be about 260-pct-GVW (1,028,000 lb) from the 
 strain gauges. 
 
 The ROPS was designed for the following loads: Side 
 load, 175 pet GVW; and longitudinal load, at one-third point 
 on top 99 pet GVW crossmember. 
 
 Although the ROPS survived the 360° roll, it did not 
 survive the 720° rollover (fig. 5). The large longitudinal load 
 and the manner in which it was applied to the ROPS during the 
 second impact caused failure of the mount at the left rear post 
 (fig. 6). The loads determined for the second impact are just 
 prior to the failure of this mount. Despite the mount failure, 
 the data indicate the structure was loaded to approximately the 
 designed longitudinal load level. However, the mount failure 
 prevented the ROPS from absorbing most of the energy 
 associated with plastic deformation. 
 
 Analysis of the test data and film showed that the 
 longitudinal load for the second impact of this test was not 
 applied in a manner consistent with current design philosophy. 
 It appears that for large machines the ROPS penetration into 
 the soil is extremely deep (70 in for the second impact). The 
 deep penetration in combination with the machine developing 
 a 45° angle to the vertical centerline of the hill in effect caused 
 the ROPS to "plow" through the soil. 
 
 The plowing action effectively distributed the longitudinal 
 load along the post and to a lesser extent across the top 
 crossmember of the ROPS. This effectively lowers the height 
 of the resultant load, with the following results: 
 
 For loads applied at the top of the ROPS from the rear, 
 the reactions at the bottom of the posts are evenly distributed 
 front and rear. 
 
 When the applied load is effectively shifted to one side, 
 the reactions at the bottom of the posts increase on the side 
 that is loaded. 
 
 When the applied load is effectively lowered, the reactions 
 at the bottom of the post are redistributed with the horizontal 
 reactions on the rear posts increasing and the horizontal 
 reactions of the front post reduced. 
 
 The change in the load application point from the top 
 crossmember (at the one-third or one-quarter point) to along 
 the post does not change the overall capability of the structure. 
 
 A failure analysis showed that the mounts had the capa- 
 bility of supporting a 150-pct-GVW longitudinal load if the 
 load was applied at the top crossmember of the structure. That 
 is, the mounts were 50 pet stronger than the ROPS. However, 
 for a load distributed on the ROPS post, simulating the test, 
 the capability of mounts (94 pet GVW) is slightly less than that 
 of the ROPS. The strain gauge data indicated that the ROPS 
 was subjected to a 100-pct-GVW longitudinal load before the 
 mount failed. 
 
 The results of this rollover test confirmed a design philos- 
 ophy that ROPS mounts must have the structural capability to 
 allow large plastic deformations in order to absorb the impact 
 energy. However, the location of the longitudinal impact load 
 on the ROPS did not occur as anticipated. The deep penetra- 
 tion into the soil, in combination with the machine developing 
 a 45° angle, caused the longitudinal load to be distributed 
 

 Figure 5. — First 390,000-lb-GVW, wheeled, front-end loader after rollover. 
 
 Figure 6. — Mount failure, 390,000-lb-GVW, wheeled, front- 
 end loader. 
 
 along the post instead of along the top crossmember. This 
 effectively concentrated more of the load on one post instead 
 of a more even distribution on all four posts. 
 
 ROLL TEST 4 
 
 The second roll test of the 390,000-lb-GVW, wheeled, 
 front-end loader met the objectives of the test. Data were 
 obtained to determine the magnitudes and direction of loading 
 on the ROPS (table 5). The machine rolled two complete 
 revolutions and subjected the ROPS to two impacts on the roll 
 hill. 
 
 The machine dynamics were very similar to those of the 
 previous test of the same machine. The major difference was 
 the fact that the longitudinal centerline of the machine only 
 changed about 19° before the second impact as opposed to 45° 
 in the previous test. This meant that there was a smaller 
 longitudinal load on the ROPS during the second impact. 
 Figure 7 shows the ROPS before and after the test. 
 
 The strain gauges provided impact load data for the 
 vertical, side, and longitudinal directions. The displacement 
 potentiometers provided data to determine loads for only the 
 
10 
 
 Table 5.— Fourth roll test summary 
 
 (390,000-lb-GVW, wheeled, front-end loader) 
 
 1st Impact 
 
 2d Impact 
 
 811 
 
 1,015 
 
 801 
 
 1,158 
 
 604 
 
 616 
 
 508 
 
 576 
 
 50 
 
 118 
 40 
 
 57 
 
 189 
 
 218 
 67 
 
 84 
 
 811 
 
 1,092 
 
 210 
 
 280 
 
 604 
 
 616 
 
 155 
 
 158 
 
 50 
 
 189 
 
 13 
 
 48 
 
 1.5 
 
 3.0 
 
 MEASURED 
 Vertical load, 10 3 lb: 
 
 ROPS strain gauges 
 
 Accelerometer 
 
 Side load, 10 3 lb: 
 
 ROPS strain gauges 
 
 ROPS deflection 1 
 
 Longitudinal load, 10 3 lb: 
 
 ROPS strain gauges 
 
 Roll rate, °/s: 
 
 Before impact: Gyroscope 
 
 After impact: Gyroscope 
 
 Max ROPS penetration, in: 
 
 Field measurement 
 
 ESTIMATED 
 Vertical load: 
 
 Force 10 3 lb.... 
 
 Force pctGVW.... 
 
 Side load: 
 
 Force 10 3 lb.... 
 
 Force pctGVW.... 
 
 Longitudinal load: 
 
 Force 10 3 lb.... 
 
 Force pctGVW.... 
 
 Av side deflection in.... 
 
 1 Static test curve. 
 
 side direction and accelerometer data provided loads for the 
 vertical direction. The strain gauges provided a means to 
 determine the loads by a standard method of data reduction. 
 However, the reduction of the accelerometer data and the 
 displacement potentiometer data required consideration of the 
 machine dynamics in order to obtain meaningful results. 
 
 In using the accelerometer data to calculate a vertical 
 load, an effective mass of the machine acting on the ROPS 
 during the impacts had to be assumed. Correlating the vertical 
 loads and accelerometer data from previous tests indicates that 
 for this test approximately 40 pet of the machine weight 
 contributed to the vertical load on the ROPS. Therefore, 40 
 pet of the machine mass times the measured vertical accelera- 
 tion of the machine was used to calculate the vertical load on 
 the ROPS. This correlated closely with the strain gauge data. 
 
 The deflection data are difficult to use in determining the 
 applied side load. The location of the side load during a roll 
 test and static test will invariably be different. Thus, the 
 deflections measured during a roll test cannot be directly used 
 to correlate loads using the static test curve. However, an 
 attempt was made to estimate the effects of the different load 
 locations. In addition, the deflections were adjusted because 
 of the geometry effects caused by potentiometer locations. 
 
 STATIC TEST OF FOUR-POST ROPS 
 
 The static testing of the four-post ROPS for the 390,000- 
 lb-GVW, wheeled, front-end loader met the objectives of the 
 
 ^675 
 
 ; ; ; " 5 
 
 f • 4 1-^ 
 
 Figure 7. — Second 390,000-lb-GVW, wheeled, front-end 
 loader before (top) and after (bottom) rollover. 
 
 test. Load deflection curves were obtained for the side, verti- 
 cal, and longitudinal directions. The energy absorbed by the 
 ROPS was determined for the side and the longitudinal load 
 tests. The loading sequence was as follows: 
 
 Side-load. — On the side opposite the roll test loads to 
 minimize the effects of possible frame damage. 
 
 Vertical load. 
 
 Longitudinal load. — Rear load at the one-quarter point 
 on the same side of the ROPS as the side load. 
 
 During the side load test the ROPS was subjected to a 
 load of 685,100 lb. The ROPS deflected a total of 8.5 in and 
 absorbed 4,179,000 in*lb of energy. The load deflection curve 
 for the side load test is shown in figure 8. The frame at the 
 ROPS-machine innerface deflected a maximum of 0.85 in and 
 had a permanent deflection of 0.26 in. Table 6 shows the side 
 load energy for each displacement. 
 
 A load of 1,567,000 lb was applied to the ROPS during 
 the vertical load test. As shown in figure 9, this resulted in a 
 vertical deflection of 0.24 in. 
 
11 
 
 800 
 
 700 
 
 600 - 
 
 500 - 
 
 to 
 O 
 
 
 d" 
 
 400 
 
 < 
 
 
 o 
 
 
 _i 
 
 
 UJ 
 
 
 Q 
 
 300 
 
 c/> 
 
 200 
 
 100 
 
 
 
 I 2 3 4 5 6 7 8 £ 
 
 DEFLECTION, in 
 
 Figure 8. — Static test side load deflection curve, 390,000-lb-GVW, wheeled, front-end loader. 
 
 10 
 
 400 
 
 VERTICAL LOAD, 10 lb 
 Figure 9. — Static test vertical deflection curve, 390,000-lb-GVW, wheeled, front-end loader. 
 
12 
 
 During the longitudinal load test a load of 659,000 lb was 
 applied to the ROPS. The ROPS deflected >3 in and it 
 absorbed 1,176,000 in*lb of energy. The load deflection curve 
 for the rear load is shown in figure 10. The rear load energy is 
 shown in table 6. The capacity of the hydraulic system was 
 reached during the test, which prevented any higher loads from 
 being reached. 
 
 Table 6.— ROPS static test side and longitudinal load energy 
 
 < 
 o 
 
 700 
 
 600 - 
 
 500 - 
 
 400 - 
 
 300 - 
 
 200 - 
 
 I 2 3 
 DEFLECTION AT LOAD POINT, in 
 
 Figure 10. — Static test longitudinal load deflection curve, 
 390,000-lb-GVW, wheeled, front-end loader. 
 
 Load, 
 10 3 lb 
 
 Displace- 
 ment, in 
 
 Av load, 
 10 3 lb 
 
 Energy, 10 3 in-lb 
 
 Incremental 
 
 Cumulative 
 
 SIDE LOAD, 6.03-in FINAL DISPLACEMENT 
 
 
 
 } 
 
 43.19 
 
 21.60 
 
 21.60 
 
 86.37 
 
 .50 
 
 
 
 
 86.37 
 
 .50 } 
 
 141.88 
 
 71.65 
 
 93.24 
 
 197.40 
 
 1.01 
 
 
 
 
 197.40 
 
 1.01 } 
 
 345.58 
 
 694.27 
 
 787.51 
 
 493.77 
 
 3.01 
 
 
 
 
 493.77 
 
 3.01 } 
 
 513.74 
 
 252.76 
 
 1,040.27 
 
 533.70 
 
 3.51 
 
 
 
 
 533.70 
 
 3.51 } 
 
 551 .60 
 
 273.59 
 
 1,313.87 
 
 569.50 
 
 4.00 
 
 
 
 
 569.50 
 
 4.00 } 
 
 581 .98 
 
 293.32 
 
 1,607.19 
 
 594.47 
 
 4.51 
 
 
 
 
 594.47 
 
 4.51 } 
 
 599.84 
 
 299.32 
 
 1,906.51 
 
 605.20 
 
 5.01 
 
 
 
 
 605.20 
 
 5.01 } 
 
 612.30 
 
 307.99 
 
 2,214.49 
 
 619.40 
 
 5.51 
 
 
 
 
 619.40 
 
 5.51 } 
 
 626.18 
 
 311.21 
 
 2,525.70 
 
 632.97 
 
 6.01 
 
 
 
 
 632.97 
 
 6.01 } 
 
 638.44 
 
 321.14 
 
 2,846.84 
 
 643.90 
 
 6.51 
 
 
 
 
 643.90 
 
 6.51 } 
 
 648.04 
 
 305.87 
 
 3,152.71 
 
 652.17 
 
 6.98 
 
 
 
 
 652.17 
 
 6.98 } 
 
 658.48 
 
 346.36 
 
 3,499.07 
 
 664.80 
 
 7.51 
 
 
 
 
 664.80 
 
 7.51 } 
 
 672.54 
 
 334.93 
 
 3,834.00 
 
 680.27 
 
 8.00 
 
 
 
 
 680.27 
 
 8.00 } 
 
 682.68 
 
 345.44 
 
 4,179.44 
 
 685.10 
 
 8.51 
 
 
 
 
 685.10 
 
 8.51 } 
 
 342.55 
 
 847.81 
 
 3,331.63 
 
 
 
 6.04 
 
 
 
 
 
 LONGITUDINAL LOAD 
 
 1.18-in FINAL DISPLACEMENT 
 
 
 
 } 
 
 39.14 
 
 18.55 
 
 18.55 
 
 78.27 
 
 0.47 
 
 
 
 
 78.27 
 
 0.47 } 
 
 156.07 
 
 76.47 
 
 95.02 
 
 233.87 
 
 .96 
 
 
 
 
 233.87 
 
 .96 } 
 
 308.78 
 
 159.95 
 
 254.97 
 
 383.70 
 
 1.48 
 
 
 
 
 383.70 
 
 1.48 } 
 
 441.26 
 
 217.98 
 
 472.96 
 
 498.83 
 
 1.98 
 
 
 
 
 498.83 
 
 1.98 } 
 
 544.23 
 
 269.94 
 
 742.90 
 
 589.63 
 
 2.47 
 
 
 
 
 589.63 
 
 2.47 } 
 
 621 .33 
 
 311.29 
 
 1,054.18 
 
 653.03 
 
 2.97 
 
 
 
 
 653.03 
 
 2.97 } 
 
 656.20 
 
 122.05 
 
 1,176.23 
 
 659.37 
 
 3.16 
 
 
 
 
 659.37 
 
 3.16 } 
 
 329.68 
 
 -651.12 
 
 525.12 
 
13 
 
 TEST WITH 286,000-lb-GVW, WHEELED, FRONT-END LOADER 
 
 The rollover test of the 286,000-lb-GVW, wheeled, front- 
 end loader (test 5) met the objectives of the test. Data were 
 obtained to determine the magnitudes and directions of load- 
 ing on the ROPS. The machine rolled two complete revolu- 
 tions and subjected the ROPS to two impacts on the roll hill. 
 An additional ROPS impact occurred when the machine came 
 to rest on its side on the road at the bottom of the hill as shown 
 in figure 1 1 . This impact was softened because the road had 
 been ripped prior to the roll. In addition, the tires had blown 
 out on the machine, which caused a large decrease in its roll 
 rate prior to impact. The only damage to the machine frame or 
 ROPS mounts that was observed was the blown tires and bent 
 rims on the left side of the machine. 
 
 During the roll the longitudinal centerline of the machine 
 remained perpendicular to the vertical centerline of the hill. 
 
 This resulted in low longitudinal loading on the ROPS. A 
 summary of the ROPS loading is given in table 7. 
 
 The displacement potentiometers provided data to deter- 
 mine loads for the side and longitudinal direction. Accelerom- 
 eter data were used to obtain loads for the vertical direction. 
 The strain gauges provided impact load data for the vertical, 
 side, and longitudinal directions. The strain gauges provided a 
 means to determine the loads at a standard method of data 
 reduction. The reduction of the accelerometer data required 
 consideration of the machine dynamics in order to obtain 
 meaningful results. The potentiometer data were corrected for 
 the geometery effects in their readings caused by the location 
 of the potentiometers. 
 
 The ROPS side loading during the first two impacts on 
 the roll hill was much lower than expected. In order to obtain 
 
 Figure 11. — Roll test 5, 286,000-lb-GVW, wheeled, front-end loader. Note that an anthropometric dummy was strapped in the cab 
 to evaluate the Bureau's Vest Restraint System. 
 
14 
 
 an understanding of the low side loads a quantative analysis 
 was conducted of the following factors that influence the 
 dynamics of a roll: 
 
 Machine mass. 
 
 Radius of gyration. 
 
 Location of the center of gravity. 
 
 Distance of the ROPS from the center of gravity. 
 
 Change in roll rates during impacts. 
 
 Translational velocity of the machine. 
 
 Penetration resistance of the roll hill. 
 
 A review of these factors does not indicate any gross 
 difference with the previous roll tests that would explain the 
 low loads. It appears that the major difference in the factors is 
 the location of the center of gravity. The 286,000-lb-GVW 
 machine appears to have a much lower center of gravity. The 
 machines previously tested rolled off the tilt table at approxi- 
 mately a 50° angle. However, this machine rolled off at 
 approximately a 65° angle and entered the first impact at a 
 slightly higher rotational velocity. From an analysis of the 
 films it appears that the machine was in a more vertical 
 position when the first impact occurred. 
 
 An examination of the overall configuration of the ROPS 
 and the ROPS-machine configuration does suggest some 
 differences which may account for the low loads. For example, 
 the ROPS on the 286,000-lb-GVW machine differs from the 
 390,000-lb machine in overall dimensions such as follows: 
 
 Smaller plan view area because of slanted posts and the 
 location of the operator within the ROPS. 
 
 Smaller side profile area (no crossmember and shorter 
 length) because of the plate construction. 
 
 Also, overall machine parameters, such as ROPS-machine 
 width ratio and ROPS-machine height, differed between the 
 machines tested. 
 
 A qualitative comparison of these factors for various 
 machines presently available may indicate that there are sig- 
 nificant differences between the 286,000-lb-GVW machine 
 with the present ROPS configuration and the other machines. 
 
 A review of the motion picture coverage of the test 
 indicates that the generally smaller profile of this ROPS, 
 especially the thin (1 in) top plate, may have resulted in the 
 
 Table 7.— Fifth roll test summary 
 
 (286,000-lb-GVW, wheeled, front-end loader) 
 
 1st impact 2d impact Bench impact 
 
 MEASURED 
 
 Vertical load, 10 3 lb: 
 
 ROPS strain gauges 562 559 NAp 
 
 Accelerometer 313 627 NAp 
 
 Side load, 10 3 lb: 
 
 ROPS strain gauges 226 192 NAp 
 
 ROPS deflection 1 180 180 NAp 
 
 Longitudinal load, 10 3 lb: 
 ROPS strain gauges 65 90 NAp 
 
 Roll rate, °/s: 
 
 Before impact: Gyroscope 125 241 NAp 
 
 After impact: Gyroscope 63 165 NAp 
 
 Max ROPS penetration, in: 
 
 Field measurement 22 38 NAp 
 
 ESTIMATED 2 
 
 Vertical load: 
 
 Force 10 3 lb.... 562 559 <7.3 
 
 Force pet GVW.... =197 196 <3 
 
 Side load: 
 
 Force 10 3 lb.... 226 192 <52 
 
 Force pet GVW.... 79 67 <18 
 
 Longitudinal load: 
 
 Force 10 3 lb... 65 90 <12.8 
 
 Force pet GVW.... 23 32 <8 
 
 Side deflection in.... 1 1 NAp 
 
 Longitudinal deflection in.... 0.6 ND NAp 
 
 NAp Not applicable. ND Not determined. 
 
 1 Computer load deflection curve. 
 
 2 Data are rounded. 
 
 ROPS slicing through the soil and offering very little lateral 
 area to develop a side force. On the first impact the ROPS 
 posts left a clear imprint in the hill indicating the side forces 
 developed lower on the top plate. This is clearly seen for the 
 first 0.05 s where little side load was experienced. For this time 
 period only the top plate was in contact with the ground. 
 
 CONCLUSIONS 
 
 It should be recognized that ROPS performance criteria 
 do not guarantee protection of the operator in all cases. From 
 the testing conducted during this program for example, it 
 appears that when machines lose contact with the hill and 
 become airborne, even larger forces and energy demands are 
 imposed upon the ROPS structure. Thus, care must be taken 
 to ensure that test conditions and/or field usage do not 
 invalidate performance criteria. 
 
 A longitudinal force and energy capability has never been 
 specifically addressed in ROPS performance criteria. Al- 
 though the tests conducted for this program were designed to 
 provide maximum side loading and only minimum longitudi- 
 nal loading, longitudinal loads were recorded for each test. 
 With the exception of one ROPS impact, the longitudinal 
 loads were relatively low in comparison to the side load. The 
 second ROPS impact of test 3 did cause the ROPS to fail 
 because of the location of the longitudinal load resulting from 
 the orientation of the machine relative to the downward slope 
 of the roll hill. 
 
 Larger longitudinal loads than were observed during the 
 roll tests are possible in an actual field rollover. For example, if 
 prior to a rollover a machine was being operated in one of the 
 following situations larger longitudinal loads could result: 
 
 1 . In a position pointing slightly uphill or downhill. 
 
 2. If an operator abruptly changed steering direction. 
 
 3. If the machine was traveling at an appreciable forward 
 or aft velocity. 
 
 4. If a small machine was suddenly accelerated causing 
 rear upset because of rotation about the powered axle. 
 
 5. In an accident during loading or unloading onto a 
 lowboy truck. 
 
 In any case, it would be anticipated that the principal roll 
 axis would rapidly revert to the longitudinal axis of the 
 machine, thus subjecting the ROPS to greater side loads than 
 longitudinal loads. Therefore, the longitudinal loading and 
 energy criteria should be equal to or less than the side load 
 capability. It is believed that a longitudinal force that is 80 pet 
 
15 
 
 of the side criteria would be adequate for 540° rolls (two top result of the high vertical stiffness and load carrying capability 
 
 and two side loads). required to meet load and energy criteria in the critical side 
 
 The other major performance factor to be considered is direction. A ROPS generally has a vertical capability in excess 
 
 the vertical load. SAE J1040c criteria currently require a of 200 pet GVW and maybe as high as 1,000 pet GVW even 
 
 minimum 200-pct-GVW vertical load for all machine types though the roll test data indicate actual maximum vertical 
 
 and masses. This requirement is usually met or exceeded as a loads of only 200 to 400 pet GVW. 
 
 INT.-BU.OF MINES,PGH.,PA. 28843 
 
 U.S. GOVERNMENT PRINTING OFFICE: 611-012/00,013 
 
m 
 
 z 
 
 "0 ID 
 
 O 
 
 T3(DC 
 
 £P 
 
 o 
 o 
 
 3" 
 
 O 
 Q. 
 
 CD • 
 
 ?o 
 
 .3 O 
 
 CO 
 
 C 
 
 o 
 
 c CD 
 
 <3 
 
 
 o' 
 
 3 
 
 O T3 
 
 So 
 
 8070 
 PA 15 
 
 S 
 
 0° 
 
 1 3 
 
 IALI 
 
 RPR 
 
 DO 
 o 
 
 D 
 to" 
 
 CD <J> 
 
 o 
 
 > C 
 
 ro 
 
 CO 
 
 Q. 
 
 C 
 
 3- 
 
 H (/> 
 
 
 
 o" 
 
 CD 
 
 25 
 
 
 
 3 
 
 5" 
 
 Si m 
 
 
 
 
 CD 
 
 i 
 
 CO 
 
 o 
 o 
 
 0) 
 
 m 
 D 
 
 c 
 > 
 
 i - 
 
 o 
 
 "D 
 O 
 3D 
 
 H 
 C 
 Z 
 
 m 
 
 O 
 
 -< 
 m 
 
 DO 
 
 C 91 89 ** 
 

 
 * 
 
 
 
 "o V 
 
 
 •TV *2» 
 
 
 
 
 
 ?•%. '• 
 
 
 
 / 
 
 ^°A 
 
 
 
 
 L^ ^ . 
 
 ^0 
 
 
 c *.-iife-"* < ? 
 
 
 ° ^P ,^ 
 
 
 
 
 
 
 
 
 
 
 S? ^ 
 
 A V 
 
 
 
 
 
 A V ^ 
 ** ^ 
 
 r oV 
 
 '* o 
 
 •-• ^ v 7 ^* */ %^-\^ %.-^\# %'tt % '4r 
 
 % / :H' : V -'W' V* -lit V -*W" V* : ll6 V 1 
 
 ^ 
 
 
 ^ 
 
 
 ^ A^ 
 
 
 "<>**" "-*.»** <\- H . <* 
 
 v oV* 
 
 JP^ A 
 
 
 
 
 
 
 
 
 4- V V 
 
 J ^ 
 
 ^"V 
 V ^ 
 
 
 .^i^, ^° ^. 
 
 40. 
 
 ^°^ v 
 
 
 
\v _ „ _ #T, 
 
 **% 3 
 
 V* -Mm-. %s :28fc \/ -Mix %s A \/ 
 
 • ">?„ A 
 
 
 C.VP 
 
 6 «A V ^ - 
 
 
 
 
 
 ^d* 
 
 .4 0. 
 
 ^ * 
 
 
 4> e, " » , 
 
 ,0 
 
 A, 
 
 .0 ^* 
 
 
 
 
 
 K^^> 4 °> & *" 
 
 ^ ^ 
 
 <. *^.«* ,G 
 
 ^ ^ 
 
 
 H EC KM AN IXJ 
 BINDERY INC. |e| 
 
 #FEB 89 
 N. MANCHESTER, 
 INDIANA 46962 
 
 
 
 
 
 
 
 
 
 
 *°«* »S»»' JP VI 
 
 ^^ . 
 
 4 0t 
 
 ■^