xm9 pistol trial

[rocco490]

http://archive.gao.gov/d4t4/130439.pdf

Salt Water Corrosion Test While latitude was shown in assessing SACO’s dry mud performance, Army evaluators exercised no such latitude in assessing H&K’s performance after exposure to salt water Salt water immersion, a desirable characteristic in 198 I, was elevated to a mandatory requirement for the 1984 competition In a procedure similar to that used for the mud test, two weapons and a number of magazines were immersed in a saltwater solution of a specrfied salinity Between test firings, which took place over a period of 10 days, the weapons were placed in a hunudity-controlled chamber. Over the lo-day period, H&K experienced 55 malfunctions m 390 rounds fired compared with 2 malfunctions in 210 rounds for the 45-caliber control weapon. As table III.9 demonstrates, many of H&K’s malfunctions occurred after the 5th day of exposure-36 out of 55. The Army systems analysts concluded that for the first 3 days of the test, H&K’s performance was comparable to that of the control weapons. Overall, they found H&K’s performance acceptable because, in their opinion, the lo-day testing cycle was not realistic Their report noted that although no mission scenario is given for the salt water immersion requirement, “one might imagine that landing in the tropics might be simuIated by the 3 day finng cycle.” Table 111.9: 1984 Salt Water immersion Test Results-Percentage Successful Firings System 45 control SAC0 Beretta S&W .I_ -- - After 3 days After 5 days Overall 100 98 99 100 100 100 100 100 100 100 99 97 H&K 97 82 86 Like SACO’s performance in the mud test, H&K’s performance under salt water immersion was not as good as that of its competitors. Consequently, Army evaluators concluded that H&K had failed to meet this

 

[rocco490]

Mud Test The requirement to function reliably after exposure to adverse conditions, including mud, salt water, sand, and dust, is based on the fact that American soldiers could be tasked to operate m many different climates Page 37 GAO/NSIAD-8s122 Amy Selecta Beretta’s 4mm. Pistol 4ppendix tU malysis of 1981 and 1984 Test Resulti and environments The 45caltber pistol has the edge m adverse conditions tests Because of its loose-fitting parts, mud and other foreign matter trapped between the 45’s mating parts has less effect on its functioning than on g-mm. pistols with then tighter fitting parts. There IS a trade-off, however, smce their trght fit tends to make g-mm weapons more accurate. Even though the g-nun’s performance under adverse conditions did not match that of the 45 in 1981, these requirements were still mandatory for the 1984 competition However, m 1984, the requirement was for “comparable” performance. The mud test was conducted in two phases First, the loaded weapon and two spare magazines were immersed for 60 seconds m a mud bath of a specified viscosity The wooden plug used to prevent mud from clogging the barrel was removed and the gun and magazines hand wiped before test firing. For the second phase, mud-immersed guns were hand wiped and left to dry for 4 hours before test firmg As noted on page 29, test officials did not have a high degree of confidence in the test results because of the small sample size and the difficulty of completely controlling the test environment We found no evidence that the performance criterion for the adverse nditions tests-comparabihty to the .45-was further defined in ther the operatlonal requirements or the specifications communicated LO manufacturers. Just prior to the start of the testing, evaluarors were instructed to use professional judgment and reason m arriving at conclusions. (See pp, 5 and 23.) Such judgment was exercised m concluding that SAC0 had passed the mud test. As shown in Table III 8, the 45 control weapons had no malfunctions m either the wet or dry phase of the mud test, a performance not equaled by any other weapon. SACO’s performance m dry mud, however, was 1’7 percent less than that of its nearest competitor In other words, SAC0 was not only not equal to the 45 but also not equal to the performance of other weapons tested. Table 111.8: 1984 Mud Test ResuttsPercentage Successful Firmgs 45 control SAC0 --_ Wet 100 98 Qry 100 79 Average ___ 100 88 S&W 98 96 97 ~~- _~ Beretta 97 98 97 __ -.--__ ~- H&K 99 100 99

 

[rocco490]

Significance of High Reliability While the high rehablllty scores achieved by SAC0 and Beretta were lmpresslve, the systems analysts suggested that these scores might not be all that slgmflcant. They pointed out that the JSOR goal of a desired rehabllity of 495 was mtended to provide a g&percent probability of successfully completing a lo-round mlsslon (1 fully loaded magazine), This goal 1s based on the operational assessment that the personal defense weapon 1s a weapon of last resort and will not be used very often, when it is used, very few rounds will be expended. Finally, by the nature of Its most likely use, the penalty paid for a failure to function ~111 be very high The test results for all weapons demonstrates a high degree of confidence that they can engage for short mission lengths without Interruption. Table 111.7: Probabtllty of Completing a Mission Without Interruption Percentage System/magazine capacity SACO/i 5 rds Beretta/l5 rds S&W/l 4 rds 45 controV7rds 7-round 1 O-Round 15Round 30-Round mlssion mission misslon mtssion 99 99 99 98 99 99 99 98 98 97 96 92 95 94 90 82 H&K/l 3 rds 95 93 90 81 The benefits of Beretta’s or SACO’s hig

 

[rocco490]

Calculating Reliability The manner in which the Army calculated rehabrlity indicates that the performance of weapons changed significantly between the 1981 and 1984 tests Rehability 1s expressed by the Army as the mean round between operational mlsslon farlure Duru-tg test firmg seven different categories of malfunctions are tabulated. Each malfunction IS, m turn, categorrzed by a “class” xndicatmg the degree of seriousness. Class I malfunctions. the least serious, are clearable by the operator in less than 10 seconds. Class II are also clearable by the operator but take 10 seconds or more to resolve Fmally, class III, the most serious, are not operator clearable but require sending the pistol to mamtenance for repair. The total number of malfunctions LS divided into the total number of rounds fired m order to calculate the mean round between operational mission failure. Table III.4 demonstrates the marked nnprovement in the rellabrlity of Beretta Lnd SAC0 g-nun. pistols between 1981 and 1984 2Apparently, a large percentage of the current mventory of 45cahber pistols had been fired mfrequently Only weapons dlstnbuted to UN@,, such as those mvolved m trammg, xe hkely to be fired often The Army estunates that on the average a pistol LS fved only 200 times a year Page 34 GAO,‘NSL4D85122 Army Selects Berettab 9-m. Pistol Appendix III Analysis of 1981 and 1984 Test Resulte Table 111.4: Rehability-Mean Round Between Operational Mission Fatlure System 1981 1984 -- 45 control 165 162 Beretta 158 1.750 H&K 169 158 SAC0 209 2,877 - S&W 293 434 Alternate Method to Express Reliability Another way of expressmg reliability IS to show performance-either malfunctions or satisfactory firings-as a percentage of the total rounds fired. This methodology is simple to understand because perfect performance equates to 100 percent Table III 5 uses this methodology This table shows a less sigmficant performance difference between the two competrtions and the individual competitors. While the mean round between operational mlsslon failure emphasizes the differences between competitors, lt tends to obscure the fact that all the weapons tested were highly reliable. Table 111.5: Reliability-Percentage of Successful Fwtngs System 45 control Beretta 198t 1984 _ _--- 99 39 99 38 9937 99 94 H&K .~ SAC0 9941 9937 99 52 99 97 -- S&W 99 66 9977 The hrgh reliability of all the pistols tested is further indicated when one compares the seriousness of the malfunctions that occurred dunng testing. Table III 6 sun-una~zes malfunctions according to their seriousness, class I bemg the least serrous and class III the most. Table 111.6: 1984 Malfunctions Classlfled by Seriousness System Class I 45 control 180 Beretta 10 Class II 11 1 Class III Total 25 216 9 20 -__ H&K 208 0 14 222 SAC0 11 0 1 12 S&W 60 0 16 76 Page 36 GA0,‘TVSIAD-W122 hy Selecta Beretta’s 9-m Pistol Appendix III Analyels of 1981 and 1984 Test Results As shown by the table, the maJority of malfunctions were class I, minor Class II malfunctions were generally not a problem. Army systems analysts noted that none of the class III malfunction rates was hrgh considering that about 35,000 rounds had been fired on each system &-my systems analysts noted that because even minor malfl!?ctions are zounted in calculatmg reliability scores, two systems could ‘e the >nme reliability score and yet be very different due to the h rlty of the malfunctions. A detailed exammation of the test data sugge, L$ that a -capon can have a poor reliability score and yet be comparable to a eapon demonstrating higher rebabillty because tht poor score is based dn mmor malfunctions.

 

[rocco490]

Reliability Test All four contestants failed the rehabrlity test m 1981 because the highest score was less than half the mandatory requirement of 800. However, the reliability of three out of four pistols exceeded that of the .45 control weapons Thus for the 1984 test, the rehability goal was revised to requn-e superiority to the control weapons, which were to be put through the same serves of tests Only H&K, whose rehablhty was 4 points less than the control weapon, drd not meet this revised requirement. Unlike the diffenng conclusions with respect to service lrfe, there was unammlty among test officmls, systems analysts, and evaluators that H&K had failed the rellabllity requirement. Calculating Reliability The manner in which the Army calculated rehabrlity indicates that the

 

[rocco490]

Table 111.1: Evaluated Pass/Fail Results for Selected 1981-84 Tests and “Raw” Test Data Evaluated Pass/Fail Results For Selected 1981-84 Tests Selected 1981-84 “Raw” Test Data FIRM Beretta H&K SAC0 S&W Beretta H&K SAC0 S&W .45 ___. Servlce life” 1981 Pass Pass Pass Pass 9,500 8,400 1984 Pass Pass Pass ‘311 ;>8;; io,ooo ‘;‘;;I‘ 7,000 I 6,000 6,125 - ~;e&bl’ityb ‘II Fall Fall III 158 169 209 293 165 1984 S Fad Pass d 1,750 158 2,877 434 162 MudC Wet 1981 Pass Pass Fall Pass 77% 88% 67% 79% 75% 1984 Pass Pass Pass Pass 97% 99% 98% 98% 100% - Dry 1981 Fad 1984 Pass Pass Pass Fad Fad Fad Pass %f 98% 72% 82% 93% Oo 100% 79% 96% 100% Fail Pass Pass Pass Fall Pass Fail Pass 73% 93% 70% 81% 04% 98% 100% 89% 97% 100% Salt water corrosiorF 1981 1984 Fall Pass Pass Fall Pass Pass Pass Pass 76% 81% 71% 88% 85% 100% 86% 100% 97% 99% hr;,g pm energyd 1984 Pass Pass Pass Pass Pass Pass Fall Fail 100% ?OO% 100% 100% 100% 100% YF Oo N/A ‘Shown In rounds, based on average of weapons tested See footnote 1, p 30 fot explanation of differ ences In 1981-84 service life lestlng The 1984 7,C0Cl round averages for Beret+; and H&K do not represent maxlmum service life smce testing was stopped at 7,000 rounds bRellablilty IS calculated by dlvidlng the number of malfunctions Into total rounds fired ‘The percentage of successful flnngs after exposure to this adverse condttlOn dThe percentage of weapons passing the test Analysis of the officml results and the raw data reveals that some firms which passed tests m 1981 failed the same tests in 1984. In 1984, S&W failed the service life test and H&K failed the salt water corrosion test, both of which they had previously passed. The analysis of the raw data shows that all firms improved their performance m at least one of these selected tests and some m several. However, the raw scores for service life and reliability were dramatically different SACO, S&W, and the .45- caliber control weapons all showed a decreasea service life. For example, S&W’s expected service hfe fell from .J 500 in 1981 to 6,000 in 1984. Both Beretta and SAC0 showed notable :’ AIlability increases-1 1 and 14 times better, respectively, than their 11 y 1 reliablllty scores.

 

[usmcbuckwild]

I still hate the M9. The grip is like holding onto a Coke can. If you don’t have large hands, very difficult to manipulate and fire 1 handed.