ENGINEERING RESEARCH INSTITUTE THE UNIVERSITY OF MICHIGAN ANN ARBOR Semiannual Progress Report No. 6 SHOCK ON ELECTRICAL COMPONENTS IN TRACK-LAYING AND WHEEIED VEHICLES June 1, 1956 to November 30, 1956 ~.... Project 2145 DETROIT ARSENAL, DEPARTMENT OF THE ARMY CONTRACT NO. DA-20-089-ORD-36543 CENTER LINE, MICHIGAN December 1956

~'~ \ r_~ ~ I} t'-A..,e -;A u ~ li`-. I No t 4$

The University of Michigan ~ Engineering Research Institute FOREWORD This is the sixth semiannual progress report on a research program being.carried on in the Department of Electrical Engineering of The University of Michigan. Most of the material reported here represents the endeavor of Harris Olson, who has devoted his full time to this program. The project has benefited from the counsel of Professor Jesse Ormondroyd of the Department of Engineering Mechanics, and in addition, the SAE Ordnance Lamp Subcommittee and the companies represented by its members have generously cooperated. ii

The University of Michigan * Engineering Research Institute TABLE OF CONTENTS Page LIST OF TABLES iv LIST OF FIGURES v ABSTRACT vi OBJECT vi CONCLUSIONS AND RECOMMENDATIONS vii I. CONTINUED STUDY OF THE ROTARY-DRUM IMPACT TESTER 1 A. Correlation Study of Two Similar Machines 1 B. Improved Components for the Rotary-Drum Tester 4 II. COMPARISON OF FIELD TESTS WITH LABORATORY TESTS ON 3-cp LAMPS 7 A. Test Background 7 B. Laboratory Test Results 7 1. Unsupported Type 7 2. Double-Anchored Type 9 3. Standard Type 1251 9 C. Field Test Program lO 1. General Operating Data 10 2. Lamp-Mounting Plans 10 3. Performance Data on the Standard 1251 Lamp 11 4. Performance Data on the Unsupported Type 11 5. Performance Data on the Double-Anchored Lamps 12 6. General Comments on Tests with Wheeled Vehicles 12 7. Special Comments on Tests with Tire Fleet 12 8. Special Comments on Tests with Lubrication Fleet 15 9. Performance Data on Tracked Vehicles 15 D. Summary of Field-Test Results 18 1. Operating Voltage 18 2. Absence of "Standard" Cycling 21 III. STUDIES OF TYPE 623 LAMP 22 iii

The University of Michigan * Engineering Research Institute LIST OF TABLES N~o. Page I. Summary of Analysis of Variance for the Two Rotary-Drum Impact Testers Using 3-cp Lamps 3 II. Installation, Failure, and Accumulated Mileage Record of Lamps Operated Hot on the Tire-Test Fleet 13 III. Description of Lamp Failures on the Tire-Test Fleet 14 IV. Installation, Failure, and Accumulated Mileage Record of Lamps Operated Hot on the Wheeled Lubrication-Test Fleet 16 V. Description of Lamp Failures on the Lubrication Fleet 17 VI. Lamp Installation and Replacement Record of Tracked Vehicle 9A6991 19 VII. Lamp Installation and Replacement Record of Tracked Vehicle 9A7829 20 iv

The University of Michigan * Engineering Research Institute LIST OF FIGURES No. Page 1. Overall view of the second rotary-drum tester. 1 2. Mortality results of type 1251 lamps from the second impact tester compared to the composite result of equal tests performed on the original machine. 2 3. Mortality results of 3-cp lamps from the second impact tester compared to the composite result of equal tests performed on the original machine. 5 4. Initial mortality results of 3-cp lamps operated on the second impact tester with a Masonite drum compared to a composite result of equal tests performed on the same machine using a wooden drum. 5 5. Mortality results of lamps held by flat spring clips and with a.075" cam offset compared to a composite result of the same lamp held by soldering and with a.0631" cam offset. 6 6. Modified type 1251, 3-cp lamp with double-anchored filament. 8 7. Modified type 1251, 3-cp lamp with unsupported filament. 8 8. Mortality results of the second series of unsupported 3-cp lamps compared to the composite result of the original group. 8 9. Comparative mortality results of double-anchored, 3-cp lamps and the original series equally tested on the rotary-drum tester. 9 10. Modified type 623 lamp with an unsupported filament. 22 11. Mortality results of modified type 623 lamps tested on the rotarydrum impact tester. 23 12. Comparative mortality results of type 623 and 1251 lamps with unsupported filament and type 623 standard lamps tested equally on the rotary-drum impact tester. - 23 vy

The University of Michigan * Engineering Research Institute ABSTRACT This is the sixth semiannual progress report covering the period from June 1, 1956, to November 30, 1956. The accomplishments may be summarized as follows: 1. A second rotary-drum impact tester has been completed and limited numbers of mortality tests indicate that the machine gives results comparable to those of the original machine. 2. The cam and the lamp-holding devices of the rotary-drum tester have been improved by new materials and design. 3. Field and laboratory tests using special and standard type 1251 lamps have been completed. 4. A group of modified type 623 lamps have been impact tested in the laboratory with the results indicating evidence of life improvement. OBJECT The objects of the research project are: 1. to determine practicable means of increasing the operating life of incandescent lamps used on tactical military vehicles, particularly with reference to their resistance to mechanical shock and vibration; 2. to study presently accepted methods of impact-testing vehicular lamps and to determine specifications for governing such tests; and 3. to design, study, and evaluate a new miniature lamp impact tester which may possibly supplant the tester now being used..vi

The University of Michigan ~ Engineering Research Institute CONCLUSIONS AND RECOMMENDATIONS 1. A duplicate of the original rotary-drum impact tester has been assembled and tested. It appears, on the basis of limited comparative tests, that the two machines give equally consistent and uniform mortality results. 2. Improvements have been made in the materials and design details of several tester components which tend to reduce maintenance problems and make lamp mounting easier. 3. A set of comparative field and laboratory tests of three types of |-cp lamps has been terminated with inconclusive results, in that the order of relative strength of the three types appearing from the field tests did not agree with the order determined from laboratory tests. 4. The results of some laboratory tests on a type 623 lamp modified in structure to resemble the unsupported type 1251 indicate a significant improvement in shock resistance. It is recommended that the possibility of adopting this structural change in lamps intended for rough service be investigated further. 5. It is recommended that effort be made to determine if the presently accepted procedure with regard to cycling in laboratory tests should be modified to conform more closely to field-service conditions. 6. It appears kquite evident that the rotary-drum tester in its present form should be considered as an acceptable and practical form of impacttesting device. vii

The University of Michigan ~ Engineering Research Institute I. CONTINUED STUDY OF THE ROTARY-DRUM IMPACT TESTER A. CORRELATION STUDY OF TWO SIMILAR MACHINES The rotary-drum tester for miniature incandescent lamps, which was described in the third semiannual report (pp. 18-20) for this project, has now been in operation for more than 18 months. During this time, 120 tests totaling more than 750 hours have been completed on this machine, and the only changes made have been in the lamp-holding devices. The other basic components are still in their original form and show no sign of wear or need for replacement. Since the original machine has thus demonstrated its physical durability and has given consistent mortality-test results (see Reports No. 4 and No. 5), it seemed appropriate to determine if a second test machine of this type would be equally durable and consistent in performance. It was determined therefore to build a second machine, duplicating all the structural features of the original model except overall length, which was increased to accommodate 20 lamps with either G-6 or S-8 bulbs. See Fig. 1. Fig. 1. Overall view of the second rotary-drum tester. After this machine was assembled and inspected, preparations were made for a series of mortality tests to be conducted in accord with the procedure described on p. 49 of the last report. The results of five such tests, using type 1251 lamps, are compared in Fig. 2 with a composite curve obtained from 1

The University of Michigan ~ Engineering Research Institute 100 SECOND MACHINE. l -— FIRST TEST -\ - | — SECOND TEST ----— THIRD TEST n- -- FOURTH TEST 0. --- FIFTH TEST _J FIRST MACHINE uJ 0 ( l: j COMPOSITE RESULT 4.. 0 40 %\.\% | _ _ _ 60 C 20 64( 0 6 0 8 O 9O TIME IN MINUTES Fig. 2. Mortality results of type 1251 lamps from the second impact tester compared to the composite result of equal tests performed on the original machine. data taken on the original machine. It is quite evident, on the basis of this limited comparison, that consistency and good correlation have been secured. When the composite results from both testers were compared, the mortality rate for the new machine was consistently higher by a very slight amount. It is not unrealistic to think that if additional tests were completed, this relationship could easily equalize or reverse. A second series of 3-cp lamps was tested in a manner similar to that just described. The relationship of the four tests and the composite curve for an equal number of tests from the original machine are shown in Fig. 5. It can be noted that all the tests on this second machine appear to have a higher mortality rate. The spread of this second series of tests for the new machine was greater than expected; so statistical analysis was employed to determine if these tests deviated significantly from the expected distribution. The X2 (ChiSquare) test was applied to compare the frequency distribution of lamp failure for each test with the average frequency distribution for the four tests. The test indicated that the machine was consistent in each of the four tests, since the probability of this distribution ranged from.08 to.75 for the four tests. Further analysis of this second series of tests was continued by the use of a three-way analysis of variance similar to that described on pp. 1-8 of the last semiannual report. With this analysis, the following hypotheses were tested: 2

The University of Michigan ~ Engineering Research Institute SECOND MACHINE — FIRST TEST --- - ~SECOND TEST ________ _ --- - X | |THIRD TEST a. -- -FOURTH TEST <4 i t \\ s\ | l }FIRST MACHINE -J0 _ |_COMPOSITE RESULT >60 ~ -- t4 z C, C' 2 ) 40 60 860 960 TIME IN MINUTES Fig. 3. Mortality results of 3-cp lamps from the second impact tester compared to the composite result of equal tests performed on the original machine. The two machines are consistent; i.e., repeated tests on the same machine give similar results. The machines are uniform; i.e., similar tests on the two machines give similar results. The time periods are uniform; i.e., there is no select time period for failure. TABLE I SUMMARY OF ANALYSIS OF VARIANCE FOR THE TWO ROTARY-DRUM IMPACT TESTERS USING 3-cp LAMPS (All figures rounded to integers) Source N ZX ZX 2 L or I DF L or I F Significance.L or I _F DF Level1 i 2 91 4136 10 1 10 j 4 91 2081 60 3 20.. k 8 91 1290 2054 7 294 1.96 -- ij 8 91 10553 92 3 31 ik 16 91 714 1097 7 157 1.05 - jk 32 91 396 2282 21 109. __ ijk 64 91 266 3139 21 149. __ 3

The University of Michigan * Engineering Research Institute Table I summarizes the computations in the analysis of variance for this series of tests. None of the ratios of the two estimates of variance are significant, i.e., all the hypotheses may be accepted. These two series of tests give at least a tentative indication of the results that may be expected when two or more machines of this type are compared. The results of the first series of tests were so closely related that no statistical analysis was performed. The second series, although not as closely related, showed the machines consistent and uniform, It is expected that these results will be duplicated when an extensive test program is completed involving these machines in their finished form. B. IMPROVED COMPONENTS FOR THE ROTARY-DRUM TESTER The drum of the original rotary-drum tester was made of four segments of hard maple, chosen because of its availability, cost, weight, and desirable physical qualities. This original drum, as was pointed out earlier, had served well in this application and a drum of similar material was specified for the second tester. After the new tester had been in service for a short period, a number of problems that had not previously been encountered arose with regard to the drum. It became evident that the effect of change of moisture content and the ability of craftsmen to fabricate hard maple were bringing in unanticipated operating variables. Several replacement materials were considered, and Masonite die stock, made from exploded wood fiber, was chosen. Some of the characteristics of Masonite are: Good dimensional stability when subject to thermal and moisture variations Low coefficient of friction (0.24 on steel) Lightweight (specific gravity of 1.41) High physical strength Ease of fabrication Product availability The cam quarters were redesigned for the use of this material. These changes allow the cam to be adjusted more easily and accurately, and the use of this material should give the drum a much longer life. Several impact tests, similar to those previously described, were conducted on the machine with the masonite drum to determine the effect of the new material on lamp mortality. The results of this series are shown in Fig. 4 where they are compared to the composite curve of similar lamps tested on the wooden cam of this same machine. It is obvious that the two types of drum give very similar results. It should be pointed out that a review of the mortality schedules indicated the same failure characteristics of previously observedtests.

The University of Michigan * Engineering Research Institute MASONITE DRUM xo --- -----— FIRST TEST L- - -SECOND TEST \ coJ8 —----- THIRD TEST. --— FOURTH TEST Q \\\J \WOODEN DRUM COMPOSITE RESULT >60 640 9. ~20 aC ~2( 4 0 600 80 960 TIME IN MINUTES Fig. 4. Initial mortality results of 3-cp lamps operated on the second impact tester with a masonite drum compared to a composite result of equal tests performed on the same machine using a wooden drum. Early experience with the masonite drum indicates it to be quite satisfactory. This material can be fabricated to closer tolerances and is very stable when subjected to moisture variations. Mortality rates on early tests seem to indicate that the new drum will have failure rates consistent with the original tester and drum material. In the last semiannual report for this project (pp. 25-27), a flat spring-clip lamp holder for the rotary-drum tester was described. Further improvement of this holder arrangement has been made whereby single contact lamps can be installed in the tester with no soldering and in a minimum of time. This has been accomplished by slightly modifying the spring clip and by adding a flat lightweight beryllium-copper spring contact for electrical connections at the lamp base. Electrical connections are made to the pivoted unit by the use of lightweight flexible insulated cable connected to binding posts on a redesigned terminal board. The new pivoted unit was subjected to several impact tests to determine if the entire unit would be practical to operate and show indication of durability. The tests were operated in a manner normal for this tester, except the cam offset was increased to.075" to compensate for the expected energy loss for this type of holder. Three tests were completed, using test lamps with a known mortality rate. The results of these tests are shown in Fig. 3 where they are compared to a normal curve for this same lamp type. On the basis of this limited experience, it appears that the holders 2

The University of Michigan ~ Engineering Research Institute K00 FLAT SPRING CLIPS -I I I F FIRST TEST --- SECOND TEST 80 -I ----— THIRD TEST 3EQ.I~~~~~~~~ ~SOLDERED U_ I_\___ ICOMPOSITE RESULT >60'E -0.... TIME IN MINUTES z l20. 2( 40 0660 800 960 TIME IN MINUTES Fig. 5. Mortality results of lamps held by flat spring clips and with a.075" cam offset compared to a composite result of the same lamp held by soldering and with a.063" cam offset. weight and energy loss. satisfactorily, but some corrosion developed rather promptly, either due to appears to be essential. A rotary-drum tester, incorporating all the changes in materials and lamp-holding devices which our operating experience has shown to be necessary, is presently being assembled. With these modifications, it should prove to be lamp types now used on ordnance tactical vehicles.

The University of Michigan ~ Engineering Research Institute II. COMPARISON OF FIELD TESTS WITH LABORATORY TESTS ON 3-cp LAMPS A. TEST BACKGROUND As soon as it was learned that arrangements might be made to mount a group of lamps on a fleet of military vehicles about to be operated at the Yuma Test Station for certain tire and lubrication tests, it seemed desirable to utilize this opportunity as a means of determining the relative durability of several types of 3-cp lamps under field-service conditions. At the same time, the test data thus secured in the field under conditions subject to some control might be compared with test data obtained on the same types of lamps in the laboratory to determine if any correlation existed between the two types of tests. It was also evident that the experience gained from these field tests might be helpful if additional field tests at some future date should appear to be desirable. Three types of 3-ep lamps were selected for the proposed tests - a standard type 1251, a lamp with a double-anchored filament (Fig. 6), and a lamp with an unsupported filament (Fig. 7). Through the cooperation of two lamp manufacturers, sufficient quantities of these types of lamps were secured to permit each special type to be divided into two groups. It was planned to subject one of these groups to routine laboratory tests on the rotary-drum tester and the other group would be shipped to Yuma for installation of the vehicles. B. LABORATORY TEST RESULTS 1. Unsupported Type.-Upon conducting the usual visual inspection, numerous lamps were found to have abnormal filament mounting, possibly occurring during the sealing of the mount. Other routine examinations revealed no differences between this group and the original lot. Figure 8 shows mortality curves obtained in four impact-test runs compared to a composite curve representing an earlier series of tests on the same type of lamp (see Semiannual Report No. 5, pp. 40-44, for further details). Statistical analysis of these four runs indicated that the results were consistent, with a very high probability of repetition. A total of 52 failures occurred, with only 5 during a hot cycle and most of these during the first cycle. Initial filament distortion was common, but ith apparently little adverse effect on shock resistance. Figures 28 and 30 of the fifth report show the type of distortion and failure prevailing. Since the composite mortality curves of the two test series had some ifferences, a three-way analysis of variance was undertaken to determine if eac eries was consistent, the two lamp series were uniform, and whether failure rate 7

The University of Michigan ~ Engineering Research Institute Fig. 6. Modified type 1251, 3-cp lamp with double-anchored filament. Fig. 7. Modified type 1251, 3-cp lamp with unsupported filament. SECOND SERIES -— FIRST TEST _ — SECOND TEST -- --— THIRD TEST IL. -- — FOURTH TEST l- 8 \ ORIGINAL SERIES6 COMPOSITE RESULT >60 4 pared to the composite result of the original group. zIt ~~~~8

The University of Michigan * Engineering Research Institute were uniform. The results of this analysis indicated (a) that each series was consistent; (b) that the two series were uniform; (c) that the failure rate was not uniform, with significance at the 5-percent level, doubtless due to the high mortality rate during the early part of the second series. 2. Double-Anchored Type.-Five impact-test runs were conducted on this type of lamp, with the results shown in Fig. 9 in comparison to a composite curve of previous test runs on the original samples of this lamp. The spread of these five runs was so narrow that it seemed unnecessary to subject them to statistical analysis. SECOND SERIES FIRST TEST l~- -| SSECOND TEST ----—.THIRD TEST rn80 --- FOURTH TEST | a80l. hi. l | -— X FIFTH TEST.J | i \ | | | ORIGINAL SERIES oW6 COMPOSITE RESULT >60 14. 20 0 0TIME IN MINUTES TIME IN MINUTES Fig. 9. Comparative mortality results of double-anchored, 3-cp lamps and the original series equally tested on the rotary-drum tester. In line with previous experience, there were almost no hot failures; most of the fractures occurred near the supports; no shorting of filament segments was observed. The excessive blackening, noticed in the previous test runs was absent. 3. Standard Type 1251.-A considerable number of lamps of this type were available for test purposes. Some of these were remnants of a homogeneous group from which considerable laboratory mortality experience had been accumulated. Others were part of a new shipment of very recent manufacture. Preliminary comparative tests of lamps from these two lots showed only very slight dif ferences. 9

The University of Michigan ~ Engineering Research Institute C. FIELD TEST PROGRAM 1. General Operating Data. —Arrangements were made, in cooperation with research and development groups at the Detroit Arsenal, to equip the military vehicles of a tire-test fleet and a lubrication-test fleet with 3-cp lamps taken from the three groups just discussed. The types of vehicles participating in each test are listed below: Tire-Test Fleet Lubrication-Test Fleet 4 - 5-ton M51 dump trucks 6 - M211 2-1/2-ton cargo trucks 4 - 2-1/2-ton M135 cargo trucks 10 - M37 3/4-ton trucks 3 - 2-1/2-ton M211 cargo trucks 2 - M48 tanks (3 of the 5-ton and 3 of the 1 - M41 tracked vehicle 2-1/2-ton trucks to be oper- 1 - M8E2 tracked vehicle ated at all times; others to be spares) The operating schedule for the tire-test fleet was as indicated below: Cross-country mileage, 15 percent at 10 mph, average. Gravel mileage, 15 percent at 30 mph, average. Pavement mileage, 70 percent at 45 mph average. Each vehicle was operated 150 miles each day on this schedule, with rated loads. In the lubrication fleet each vehicle was operated 10,000 miles on pavement, at 45 mph, with rated payload and rated towed loads. 2. Lamp-Mounting Plans.-Each wheeled vehicle on either fleet could accommodate 8 lamps of the 1251 type. A short was made in the lighting circuit so that 5 lamps would be energized at all times and the other 5 lamps would always operate cold except when the lamps were checked for failures. The location of the lamp housings and the designation of the lamp at each location was as follows: A16853 C168 E1685 G B I F H Right Front Left Front Left Tail Right Tail (Lamp positions A, C, and G cold during vehicle operation.) 10

The University of Michigan ~ Engineering Research Institute For hot operation, each vehic.le was assigned one standard 1251 lamp, two unsupported lamps, and two double-anchored lamps. By methodical rotation of lamp position, each lamp type was equslly distributed in the five locations. One lamp of each type was assigned to a socket that was to operate cold on the vehicle, and a similar rotational plan was followed. The lamps were operated (except the 3 cold lamps) whenever the vehicles were in motion, with checks for failures made at periodic intervals during operation, and any failures were replaced by a similar lamp. 3. Performance Data on the Standard 1251 Lamp. —The actual distribution of this group of lamps to various lamp locations on the wheeled vehicles of both fleets was as follows: Lamp location A B C D E F G H Total No. of lamps installed 8 5 7 8 6 4 9 6 53 The inequitable distribution is mainly due to the proportion of standard and test lamps designated to operate hot or cold. These 53 lamps, mounted on 27 separate vehicles, varying from 3/4 ton to 5 tons in size, had four failures that could be attributed to the filament. All four lamps were operated at hot locations. Seven other lamps became inoperative for some other reason, i.e., glass loosening from base, accidental breakage, etc. Five other lamps apparently had filament fractures that had become rewelded. The lamps were inspected when returned to note the test effects upon the filament. An arbitrary rating of excellent, good, fair, poor, or completely broken was used to describe the condition of the filament. The 53 lamps were rated as follows: Filament condition: Excellent Good Fair Poor Completely Broken Operated hot 10 12 2 1 5 Operated cold 19 3 1 0 0 Total 29 15 3 1 5 As expected, the hot-operated lamps seemed to be more affected, many of them showing evidence of coil stretching, unhooked filaments, excessive deformation, etc. 4. Performance Data on the Unsupported Type. —The distribution of the lamps of this group, by socket locations on the wheeled vehicles, was as follows: Lamp location A B C D E F G H Total No. of lamps installed 7 10 9 7 13 11 5 9 71 The inequitable distribution of lamps is due to the planned proportions of hot and cold operation, replacement lamps, and error in installation. Twelve of the 11

The University of Michigan ~ Engineering Research Institute 71 lamps had filament failure during the test, with most of the failures apparently due to filament segments shorting together. The other failed lamps had fractures at random points on the filament coil. Most of the returned lamps showed distorted filaments similar to the results of the laboratory tests described on pp. 42 and 43 of the last report. Only one of the surviving group of lamps indicated a fracture that had rewelded. With hardly any exception, the filaments of all survivors were noted to be in good to excellent condition with very little stretching of filament segments or shorting of individual turns whether they were operated hot or cold. 5. Performance Data on the Double-Anchored Lamps. -Eighty-two doubleanchored lamps were used during the tests on the wheeled vehicles. The distribution of these lamps at the various locations was the following: Lamp location A B C D E F G H Total No. of lamps installed 10 8 6 11 12 12 11 12 82 Here again, the inequitable distribution is due to the hot and cold operation, replacements, and error in installation. Of the 82 lamps, seven had filament failures, one was broken accidently, and two had possible filament fractures that had rewelded. Most filament fractures occurred near or at the supports, with evidence that the segments had shorted either before or after the fractures None of the survivors' filaments showed much evidence of being subjected to any great forces. That is, all seemed to be in relatively good condition upon inspection at the completion of the test. The only evidence of strain upon the filaments was some stretching of the coils at the supports. 6. General Comments on Tests with Wheeled Vehicles.-The total number of lamp failures was too small to obtain significant mortality curves. It is of interest, however, that there were no failures among the cold lamps and most of these showed very little filament distortion. Of the lamps operated hot, the standard type 1251 seemed to show more severely distorted filaments than the other two types, although the number of actual failures was small. Almost 80 percent of the lamp failures occurred at the left tail-lamp position; the size or weight of the vehicle did not seem to influence the failure rate. 7. Special Comments on Tests with Tire Fleet.-Detailed information on the performance of all lamps operated hot on the tire fleet is shown in Table II. It will be noted that the ratio of one standard lamp to two unsupported and two double-anchored lamps was adhered to quite closely, and that the ratio of failures followed practically the same pattern. Only 2 vehicles, with 23 percent of the total mileage, had no lamp failures. The tire fleet accumulated a total mileage of 26,291 and had 18 lamp failures, or 78 percent of all failures on all wheeled vehicles of both fleets. A summary of the failure schedule given in Table III is as follows: Location B D E F H Number failing O 1 9 6 2 12

TABLE II INSTALLATION, FAILURE, AND ACCUMULATED MILEAGE RECORD OF LAMPS OPERATED HOT ON THE TIRE-TEST FLEET Standard Lamps Unsupported Lamps Double-Anchored Lmps La 0 0) a) a I I e I o I ~~~H IO bD d e, 1 a~ e l Cd~~~~~~~i rd l)rd Ca(I rd a()bO bD O ( 1 O O HH H I I w I H ii C])I H H H Q 4o H H o H H 4 C) cii H'' 1 U 1 4 5 27.0 4,294 1 0 4,294 3 2 7,494 2 0 8 5 2-1/2 25.7,29:.3!,247 1 1 934 2 1 2,181 4 1 2 ~810 ='a 6 2-1/2 25.7 0 7 2-1/2 24.9 1,694 1 0 1,694 2 0 3,388 2 0 3388 8* 2-1/2 26.9 1,855 1i~ 1,855 2 0 1,710 270 03,710 8a -/2 28.3 994 0 994 1,92 1 88 0 H9 49 42-14/2~27.03 32504 1 0 32504 4 2 7,8 2 0 7,8 10 2-1/2 26.7 3,004 2 2 2,303 3 3 5,04 4 2 0 6,008 15 5-t/ 27.1 1,7944 1 0 1,7494 2 0 35,488 2 5 1,7 8* 2-1/2 26.9 115 1 0 1,1855 2 0 15,710 2 01,1 8a* 2-1/2 28.15 994 0 994 0 1,19881 198 9 2-1/2 27.,3 15504 1 0 15,504 4 2 7,08 20 7,8 10 2-1/2 26.7 15,004 2 2 2,1015 3 s5,o44 206o8 Ibtal 26,291 10 3 25,277 22 8 50,211 24 7 52,684 *Lamps transferred from truck 8 to truck 8a at 1855 miles. |*Lamps transferred from truck 8 to truck 8a at 1855 miles. (

TABLE III. DESCRIPTION OF LAMP FAILURES ON THE TIRE-TEST FLEET = Lamp Type of Vehicle Socket No. Miles Until No..Type of Lampl Installed on Location Lamp Failure 14 Standard M1355 E 934 One support and one lead burned off-bulb very black. 26 Standard M211 F 339 Broken at lead —bulb very black. T' 54 Standard M211 F 1964 Very little filament, supports, or leads left — bulb black. X 72 Unsupported M51 E 1369 Both filaments broken near coil ends-bulb very black. 81 Unsupported M135 F 934 Filament broken near coil end —bulb very black. 107 Unsupported M211 H 888 Tangled mess —bulb very black. 109 Unsupported M211 E 2699 Tangled mess-bulb very black. m 110 Unsupported M211 F 765 Both filaments broken —random locations on coilo 167 Unsupported M1211 H 1457 Tangled mess —bulb very black. 168 Unsupported M211 F 965 Tangled mess —bulb very black.. 169 Unsupported M51 E 1831 Break in center of coil-second filament tangled. m 211 Double Anchor >1L35 E 158 Unknown, 257 Double Anchor M135 E 994 Broken at supports -second filament tangled and broken.: 287 Double Anchor M51 F 964 Broken at supports -second filament O.K, 288 Double Anchor M51 E 964 Tangled mess -bulb blackened. 289 Double Anchor 1M51 E 2210 Broken at supports. 290 Double Anchor M51 D 165 Broken at support-second filament operative. 302 Double Anchor M51 E 673 Broken at supports..;~~~~~~~~~~~~~~~~~~~

The University of Michigan * Engineering Research Institute The location showing the greatest number of failures was the left tail lamp, which appears to have significance, since the probability of this distribution is less than.01. There is no obvious reason for this location to produce such a mortality. From Table III it can be noted that the average miles per failure of all failed lamps was 1129, ranging from 158 to 2699. For the standard type 1251, the average miles per failure was 1079; the unsupported lamp, 1363; and the double-anchored lamp, 875. No data are available as to what type of road caused the failures. The lighter vehicles experienced more lamp failures than the heavy 5ton M51. The M51 operated 2000 miles per failures the 2-1/2-ton M135, 1500; and the 2-1/2-ton M211, 1200. 8. Special Comments on Tests with Lubrication Fleet, The lubrication fleet of 16 vehicles was operated only on paved roads for a total of 160,056 miles. During this operation only five failures were observed; one standard lamp and four unsupported lamps (see Tables IV and V), which were all located at hot-lamp locations. In other words, this fleet operated 32,000 miles for every failure in comparison to 685 miles for the tire fleet. The most logical explanation seems to be that a truck operated only on paved roads does not provide sufficient excitation to cause either failure by fatigue or ultimate failure by plastic deformation of the filament. It should be noted that four of the five failures appeared to be the result of fatigue and also that three of these were at the left tail lampo 9. Performance Data on Tracked Vehicles. —Of the four tracked vehicles assigned to the lubrication fleet, the only lamp data received from the Yuma Test Station concerned the two M48 tankso The tracked vehicles had a lamp arrangement as shown in the diagram. A C Right Front Left Front Left Tail Right Tail Lamps E and G were designated to operate cold and the other positions were to be operated hot at all times the vehicle was in motion Lamp assignments were made in a manner similar to that described for the wheeled veheicles. The average 15

TABLE IV INSTALLATION, FAILURE, AND ACCUMULATED MILEAGE RECORD OF -I LAMPS OPERATED HOT ON TEE WHEELED LUBRICATION-TEST FT.ET Standard Unsupported Double-Anchored - Vehicle Vehicle Vehicle Totamp lp Lamp No Rating, Vehicle Lam p No. tons Voltage* Miles Lamps No. of Lamps No. of Lamps No. of. Installed Failures Installed Failures Installed. Failures < 11 2-1/2 26.8,27.6 10,011 1 0 2 0 2 0 12 2-1/2 26.9,26.3 10,003 1 0 2 1 2 o 13 2-1/2 26.4,27.7 10,000 2 0 2 1 2 0 14 2-1/2 25.7,25.9 10,005 1 0 2 0 2 0 15 2-1/2 26.7,26.7 10,000 1 0 2 0 2 0 16 2-1/2 27.1,26.5 10,005 1 0 2 0 2 0 17 3/4 25.6,26.7 10,001 2 1 2 0 2 0 aY 18 3/4 25.2,25.2 10,000 1 0 3 0 2 0* 19 3/4 27.6,27.6 10,006 1 0 3 1 2 0 m 20 3/4 26.3,25.8 10,013 1 0 2 0 2 0 21 3/4 26.6,26.7 10,003 1 0 2 0 2 0 22 3/1 26.8,25.7 10,001 1 0 2 0 2 0 23 3/4 28.1,27.0 10,o002 1 0 2 0 2 0 = 24 3/4 28.7,26.1 10,000 1 0 2 0 2 0 25 3/4 27.4,24.9 10,006 1 0 3 1 2 0 26 3/4 27 1,26.8 10,000 1 0 2 0 2 0 Total 160,056 18 1 35 4 32 0 First voltage reading observed at beginning of tests second voltage reading observed at test completion. One lamp smashed, replaced by an aunsupported lamp.

TABLE V DESCRIPTION OF LAMP FAILURES ON THE LUBRICATION FLEET Lamp Type of Type of Vehicle Socket No. Miles Until Description of No, Lamp Lamp Installed on Location Lamp Failure Returned Lamp 41 Standard M37 D 478 Filament twisted. H 93 Unsupported M211 E 3006 Tangled mess. 98 Unsupported M211 E 7386 Filament broken near coil end-second filament stretched. 122 Unsupported M37 H 3907 Filament broken near lead-second operative.: 126 Unsupported M37 F 411 Unknown.

The University of Michigan ~ Engineering Research Institute tracked-vehicle speeds for cross country, gravel, and paved road were 10 mph, 20 mph, and 25 mph, respectively. The two tail lamps of one M48 (No. 9A7829) had double-bubble shock mounts, i.e., the lamps were isolated from the lamp housing by rubber spacers. The record sheets from these two vehicles are reproduced as Tables VI and VII. Since one vehicle was responsible for about 2/3 of the failures, they will be discussed separately. The first M48 (No. 9A6991), which was described in the field report as "subject to extreme vibration," had twenty lamps installed during the course of the test, with none surviving. The vehicle did not seem to discriminate against lamps whether operated hot or cold. Those lamps that failed cold usually had a segment or the entire filament broken off the leads and/or supports. Hot failures usually had the filaments entangled to create a short that quite often burned off a lead or support. Most of the hot failures had very little left of the filament portion of the lamp. The two front marker lamps, which are the farthest from the engine, seemed to survive the longest of any of the lamp locations. All but two failures occurred while the tank was traveling on paved roads. The second M48 had fewer lamp failures and each lamp accrued far greater mileage before failure. Possibly some of this difference can be attributed to the fact that the rear lamps, which are located nearest the engine, were shock mounted. Two lamps mounted in these housings (location E and G) and operated cold were each able to survive about 2300 miles of operation. The two front marker lamps, which were not shock mounted, also showed a relatively long life in comparison to those mounted in a similar position on the other vehicle. The inspection of the returned lamps of this vehicle showed that the failures resembled those previously described. Data obtained from tests on only two tracked vehicles seem insufficient to justify any attempt to classify failures by lamp types. It is obvious, however, that lamp maintenance on tracked vehicles is a real problem, probably calling for improved shock mounts at all locations. D. SUMMARY OF FIELD-TEST RESULTS As might be expected in any field test, there were numerous variables in the test conditions in spite of the conscientious efforts of the test-station ersonnel to control all operating conditions and to report the results faithfully and accurately. Some of these variables doubtless affected all three types of test lamps equally; others may have operated in favor of one type. Two will e mentioned briefly. 1. Operating Voltage.-The voltage of each vehicle was much lower tha xpected and conditions did not permit correcting this situation. Voltage read18

TABLE VI LAMP INSTALLATION AND REPLACEMEIT RECORD OF TRACKED VEHICLE 9A6991 Cd Test Miles Cxl -p -4 02 W 0 rd Description of Lamp U krj.4 0 0 0 0 P 0 I 0 E F-4 Cd P4 a ~r rd 4 Hrd r4 r_4 ~~ H 02 Ct) E= ~~S P 02 () Cxl Hl Cxl 02 0 cii:Z 4) 0 0 _ _ _ _ A X 0 332 2692 3024 Filament breaks at anchors. W C X 0 0 277 277 Filament, leads, and supports gone-bulb black. E X 0 0 69 69 One support and filament broken off. F X 0 0 69 69 Filaments broken at lead. G X 0 0 69 69 Filament broken at point of attachment. m H X 0 0 9 69 Filament breaks at anchors-bulb Replacement Bulb Data G X 0 0 72 72 Filament broken at point of attachment to leads. H X 0 0 72 72 Filament fractures at anchors-bulb black. F X 0 0 72 72 Filaments tangled-broken at random locations. E X 0 0 72 72 Filament broken off at leads. G X 0 0 82 82 Filaments broken at leads. H X 0 0 156 156 Unknown-bulb darkened. F X 0 0 82 82 Filaments broken at leads. E X 0 0 82 82 Filament broken off at leads. C X 0 0 76 76 Tangled filament segments-bulb black. E X 0 0 51 51 Fracture near lead attachment. F X 0 0 51 51 Fractures at random locations-entanglement. G X 0 0 51 51 Filaments broken at attachment points. C X 0 0 854 854 Filament broken near supports. C X 0 301 0 301 Filament tangled-bulb very black.

TABLE VII LAMP INSTALLATION AND REPLACEMENT RECORD OF TRACKED VEHICLE 9A7829 e1 o Test Miles -| ~~ 0 o o Description of Lamp m Cd r d -P I o 0 0 o C X 0 99 99 Fractures at supports-bulb very black E* X 791 981 954 2726 No failure. F* X 0 183 183 Fractures at random locationsG* X 0 87 87 Fracture midway in coil segment.0 H* X 7 981 954273 273 Filament broken at anchor-stretching and entanglement. C X 0 0 979 959 Fractures at supports-bulb very black. EG* X 791 981 954 2726 No failure. = F* X 0 0 185 185 Fractures at random locations-bulb black. CG x 0 0 87 87 Fracture midway in coil segments., H* X 120 105 275 41 27 Filament breaks at anchors-tangled. Replacement Bulb Data F* X 0 728 767 1495 Filament fractures at leads-bulb very black. H* X 0 0 666 666 Tangled-bulb very black. G* X 791 981 525 2297 No failure. C X 557 981 15 1533 Tangled filament segments. H* X 120 1055 41 1216 Filament broken at supports. *Shock mounted.

The University of Michigan * Engineering Research Institute ings, taken at the voltage regulator after about 30 minutes of motor operation, ranged from 24.9 to 29.0, with an average of 26.7 volts. This low voltage would tend to increase lamp life. 2. Absence of "Standard" Cycling. —During field tests, lamps at certain locations on the vehicles were burned continuously while the vehicles were operating; at other locations the lamps were off continuously. It is our belief that the absence of standard cycling tends to produce a significant increase in lamp life as affected by shock and vibration. In laboratory testing, using the standard cycle of 25 minutes "on" and 5 minutes "off," the majority of lamp failures occurred at the beginning of the cold or "off" portion of the cycle. Apparently the cooling of the filament caused it to lose its ductility and made it more susceptible to shock. It is not surprising therefore to find little or no correlation between the results of the field tests and the laboratory tests on these three types of lamps. Field failures were small in number and seemed to indicate that the standard 1251 and the double-anchored lamp were the strongest and the unsupported was the weakest. Laboratory tests reversed this order and indicated the unsupported type to be the strongest. In the field tests, all the standard-lamp failures seemed to be caused by plastic deformation and/or entanglement of the filament segments. Eight of the twelve unsupported filament failures were of this same nature. Of the seven failures in the double-anchored type only two seemed due to the above-mentioned cause; the others appeared to be due to sawing at the supports. The operation of wheeled vehicles on paved roads does not seem to provide enough shock excitation to cause significant lamp failure rates. Failure rates are much higher when these same vehicles are operated on rough roads, with the left tail-lamp position appearing to be the most severe. The operation of tracked vehicles provided extremely severe shock excitation, with some evidence to support the use of shock mounting. In fact, improved shock mounts at all sockets would appear to be needed. 21

The University of Michigan ~ Engineering Research Institute III. STUDIES OF TYPE 623 LAMP In the last report for this project, a special unsupported type 1251 lamp was described and the results obtained with it in laboratory tests were reported. Its apparent resistance to shock showed such an encouraging improvement over the standard design that the manufacturer was requested to supply the project with a group of type 623 lamps (rated at 6 cp) in which the same structural features were to be embodied. Figure 10 shows a type 623 lamp with an unsupported filament. X~i~ i Fig. 10. Modified type 623 lamp with an unsupported filament. Inspection of the individual lamps in this shipment showed some with irregular filament mountings, distorted filaments, and displaced mounts, probably as the result of sealing the mounts in the envelope. The natural frequency of the mount was observed to be in the range of 570 to 700 cps, and of the filament segment, 210 to 220 cps. The lamps of this group were then tested on the rotary-drum impact tester, following the standard procedure for such tests, and the results are shown in Fig. 11. The spread of the five tests seems quite large, but when the test data were studied statistically the machine was found to be consistent in each test. The failure schedules show that filament failure occurred chiefly during the cold portion of the cycle, with the fractures appearing at random locations. Filament entanglement was almost completely eliminated. Initial distortion occurred, but did not grow noticeably worse as the test continued. Figure 12 compares the performance of this group with that of some t 22

The University of Michigan * Engineering Research Institute 00.? -— FIRST TEST 100 ___ a- - 1I- -SECOND TEST ______ _ —-. -.....~THIRD TEST u)80 a. --- FOURTH TEST 4'~.~~~ ~- ~~FIFTH TEST >6.. 0 40........0 6.0 800 960 TIME IN MINUTES Fig 11. Mortality results of modified type 623 lamps tested on the rotary-drum impact tester. 0 - -— 623 UNSUP ORTED LKN _ _ _\ +I I1 —- 1251 UNSUPPORTED n80. Oa. X'\ — |623 STANDARD (.3 %20 200 4.00' 6 0 8'00 96 TIME IN MINUTES Fig. 12. Comparative mortality results of type 623 and 1251 lamps with unsupported filament and type 623 standard l- s tested equally on the rotary-drum impact tester. 23 25

The University of Michigan ~ Engineering Research Institute 623 lamps of standard design and with the unsupported 1251 lamp. It is quite evident that the unsupported type 623 seems stronger than the standard 623 and also stronger than the unsupported 1251. It is therefore recommended that methods of improving this lamp, by structural changes similar to those contained in the test samples, be studied and that consideration be given to the possibility of substituting this modified 623 lamp for the 1251 in extremely rough service applications. 24

UNIVERSITY OF MICHIGAN 3111911111111151 0208611 1111111111111 3 9015 02086 6417