QUALITY ASSURANCE DIRECTORATEV-4t, Metal fatlip-e c~f a clItfcal crP = =C2 as a CM MIb.-Cl-~Fiiz...
Transcript of QUALITY ASSURANCE DIRECTORATEV-4t, Metal fatlip-e c~f a clItfcal crP = =C2 as a CM MIb.-Cl-~Fiiz...
:L J
WVT- QA- 7201
FATIGUE BEHAVIOR OF THICK-WALLED CYLINDERS(7.62MM RIFLED SPECIMENS)
TECHNICAL REPORT¢X
fbe
JUNE 1972
QUALITY ASSURANCE DIRECTORATE
U.S. ARMY WEAPONS COMMANDWATERVLIET ARSENAL
0,, WATERVLIET- NEW YORK
AMCMS NO. 4931.OM.6350
PRON M7-O.P3381
DISTRIBUTION OF THIS DOCUMENT IS UNLIMITED
Reproduced byNATIONAL TECHNICALINFORMATION SERVICE
I1 S Deportment of CommerceSpringfield VA 22151
; \-- --
Ar
04U1
DISPOSITION
Destroy this report when it is ne longer needed, Do not return it
to the originator,
I
IDISCLAIMER
The findings in t,.his report are not to be construed as an officialDepartment of the Army position unless so designated by other authorized Jdocuments.
bcud DOC4ENT COKMRLDVATA. - t & D
t~~~~~~~ . W11SATIITC#)MREPORT 3ccuaITV CLaseeVICAflO04
WATEMMI~ AUE viA Th.eIaaifted
Fatigu Beairof Thick Wus4d CYlindars
Vincent T. Chodlcowski and Peter Go GOUtos
S. CPOVVT VATK PAL TOTAL 1N0. OWr PA9S A?. No. of nEit
114t. CONTRACT OR GRNTN NO. 84. 1~T0WX OW-ORT NUM11EROfl
Ais ,io. 493l.om.63501%~ PROJECTNMO- uvT - QA - 7201
- PRON H7-O-P3381 Sb. OTHER REPORT NO(S) CAW' .ffiIN840 1609 MV bO 0600i-0i
I0. OISTRISIJTIOM STATEMENT
Distribution of this documcat is unlimited.
o I I. SUPPLEMENHTARY NOTES I.SpONSOMINS MILITARY ACTIVITY
Army materials and Research centerWatertown, Masse
Thil. report deals with fatigue life testing employing 7fr62MKh rifled barrels
in order to secure information on the rel.ationship of specimen life achieved
through cyclic hydraulic pressure stregging and that achieved by firing of service
type ammnition. liaterials used inc~luded standard M14 rifle barrel stckl from
several suppliers and 175HM M113EJ. cannon tube forging material from chree suppliers.
The report also addresses the problem of "life" reliability predicticra anddewnstrates the potential usefulness of small barrel specimen testing in pro-
* viding insight to "'big gunfl research, development, manufacturing, field support
and quality engineering type problems.
I DC) ,~'.,1473 N~LACK$ D@ FORMI 1471D.IJAN 64. WHIEN Ia nlssfe
8jp09TFOR RMYUSX
14 LIMIC A UIN a LIEE3 C
R OLZ INT ROLE WT RONLE, IFT
'Fatisue FaibreIFired LifeFiring DaaSe,
Cann~on Tube Material
Rteliability Prediction/'verificationCorrelation Hydrauilic SilzlatianIFlring
LI.
Security Clamslfication
MW 2
N40LC1 MY~ I I X
Pressure Cyclinzfired Lifefiring Dzzag-
Cannon -5be MArterisl7.62M Barrel Mater~zJ.Rliability 2Pretion#erifiation
Correlatiot2 Hydraulic Siz1ationixfing
Se-CUrit ClasifiatiI
t I
WVT- QA- 7201
oFATIGUE BEHAVIOR OF THICK-WALLED CYLINDERS j(7.62MM RIFLED SPECIMENS)
TECHNICAL REPORT
BY
VI'%'CE~iT T. CHtZOK&o'.SKl Dl D~ CA'%D L
PETER G. GOUTOS S 1. 19T2
- -' °: ""-D.yJUNE 1972
QUALITY ASSURANCE DIRECTORATE
U.S. ARMY WEAPONS COMMANDWATER VLIET ARSENAL
WATERVLIET- NEW YORK
AMCMS .o. 4931 .OM. 6350 Detaf" 6f fulustraiion- ;i'P.this documerit may be better
PRON M7-O.P3381 Studied on microfiche
DISTRIBUTION OF THIS.DOCUMENT IS UNLIMITED
- . "d.. o,
FATI(ZE EMLVII2 Q-:T flI WI1M CYLT-M2?S
This report deals with fatiEie Fatigue Tailurelif-e zesting e6plong 7.62 ! Pressure Cg clim
rifled L-arrels in order to se- Fired Life
cure Inforazri n Gn the relation- Firing age;e
ship of speci en life achieved Cannorn Tube l!aterial
chrough cyclic hydrauIlc pres- 7.62!1 Barrel .tateri l
sure stressinr and that achieved Re-lability Prediction/Verification
by firing of service tyie a,=ni-Correlation Hydraulic
tion. Uaterials used included Siculation/Firing
standard '.314 rifle barrel stock
froz several suppliers and 175MC!
Mhl3AI cannon tube forging mater-
al from three suppliers.
The report also addresses the
problem of "life" reliability
prediction and demonstrates t.he
potential usefulness of small
barrel specimen testing in pro-
viding insight to "bin gun" re-
search, development, manufactur-
ing, field support and quality
engineering type problems.
~id.
2 Oijectives3 Aproacih 24 .Z-e ts a d liscssi= &5 Conclusi n 76 Pec~edztcc
APPE ICES
(PHASE i-AIPX P.)
Estabfdslsmnt of Test Terhniques
A 9-16
(PHASE II-APPXs B T-1'I Z)Correlation of lI,,draulic Pressure Q-cled Life to Fired Life
D 17-20C 21-23D 24-30E 31-36
F Application of Yalti-Leve1 Testing to 37-41Standard Cannon
(PIHASE III-APPXs G THRU L)Comparison of 1751DI Tube Forging MaterialProduced by Three Contractors X,Y,Z.
G 42-44H 45-47I 48-50. 51-54K 55-58L 59-63
M Application of 7.62MM Barrel Specimens 64-67to a Field Failure Problem
N 7.62MM Ammunition Calibration 68-69
0 Miscellaneous References 70-71
,- nf
...
Zsta _Ils!mct cf safe c=== fe t7y ce fr c! a-=Ifrc
Is C5e S:Esfr, PartIC22hry Vb deefi alr ssccuae
V-4t, Metal fatlip-e c~f a clItfcal crP = =C2 as a CM MIb.-
Cl-~Fiiz sL!_1c I55-M2I bcwftzer *=e to It-; 1.2tJe I!Ie fer a pr-h
of cr-e mdl1k4a dcl!as Is cc~s1idcre a Zesz==I= cl a
s=[ficIe=t~re ef ="-,es L-y UIf iri to ac~ iewe a szatszLcal1
51deratIon to nmeral Issces zd 7refessiotal etis. Iopicatfr=
t - f if I:; mdtbi atte~azt extremes I= tereraturcs, Fres-
sure, azd ezerq7 Iteimg develope& I inl:-Anal t!:r Is diffIcalt If
nor :poss1le. Therefore In order to SaIn a Itettr cmderstzndimg;
of lIfe testizcg, s-=11 size specim erfrIezrztn apeared as a
logical approach for gmeratIon of data I'v uich to look Into the
theoretical, statistical aad experimental =ethods now tein;. ex-
plored by =any investigators.
2, OBJECTIVES
The irediate goals of this study are:
a. To correlate a~unition firing pressure fatigue effects to
hydraulic pressuire cycling fatigue effects using thick Valled gun
barrels.
b. To determine fatigue life performance of cannon tube
material furnished by different contractors, produced to a co~on
specification but by proprietary processes,
2--- rare gr of- eis xs=37y aml sti- res1fci
Zc ~ c z mere sf maer121 tes= ctark r
E:! C =5e fc--%! t7 1Ife
3. An4 LV
ihmse z - CT~m~ ir e=z~ e'
ne 7..6-O mfe = Ls SaeIre as a zess 'wellac bt,ea.e c:!
I=w =~s= 2= a ara&z!My ef =21 rlf!,e Bazre~s. Stf-m
t!r, fat4;=e lIfe cf a ztezlA rfLe Barel was e~fLmze! Ift
Im c4=-zrLT= WWIr tl: of a trr'!= ca=n, I= =5dzs mei mm
dr~m theC ,arreI J-st or=- th cr4I ef rifl-- =c a fccz t'o sa
1-mx douxM bai-r~a Im cr--Ie cc cciic~ arlier fmAizre. MmIcI
tests Ukerc X= UIdth tarrels =ekb~ t- a 1A4 :u-A rzxI~ (=xclo CfI
ccts!Ze 1mDze t cfr rhffrI- Vrcre dilameer). A te ef fire
of six (6) rc=n3s p=r rnlIntr ns slected as en rcau--aa~e. Even
at this Ilw rate, barrel heatL.6 scrIcusly d~e~r~zd! lI-fe (Eazrels
ttzrced blune) azd a decision 'as made t3 T-mer cool the remalin--Z
1.4, 1.6, and 1..8 vfl ratio sein d~fr fring (See AJ-ptdIx
A for infornation =d data prtain-g to this pillse of =rk).
Phlase 11 - "Crrelation of H-ydraulic 'ressure C-7c'Ied: Life to
Fired Life -
The follcu-ng decisions w-ere r-ade based on Phase 1 restilts:
a. Use a standard proof fired I-l4 barrel rodified to a 1.8v-afl
xai- o In the vit~inity of the origin of rifling.
b. Fire at 6 rmuds per mimilta.
2
- - -. *-. - - -
sx-~~~- -.:c7'-y S
Mole~ -,ta -,=m=
; m fnutr. T- :± 14= m a li
S-tc 11 " afr Z-!: 41:7' - _
scf ted ee ls if cJ2me C=1I se-.i. S
-? ces s.-e--=--.Ve f
4-'4
im71W- :2. 7vt ell Z:.! c
-- A ":-Ire
7:= t'wX-vgr=r T- e =='
fps!'-"
'I T II
Ally.SM
TCsr PER TIM-
3 2VCCfl 7,416 78=,2 C, 1)' 5,5• "'
3.'a'D 2,9733LCCO 7,C46 6,373
3, 0 2,9395 4,C 2,174 20666
16cT-rD 2,5375,63 0
6 6,20 0 Avg 6,1446,4S7 0 EdS
are s=4 in AppenIx C and yields a relationsh p between hydraulic
pressure cy'ed specinen life to fired specimetn liWe of
Y(r'cIes) ' - 2.21 X(rcurds) +12,122
or 2.21 hydraulic ce are equal to one fired round.
It sh=uld be noted tbat. although large differences are e:Ident
among individual specifen lives at any ne level, the average life
at eaeb level, closely adheres to the straight line relationship.
The fired to destruction barrels (test level 6) c,nfirn the
validity of the resultant equations raver in predicting mean
fatigue or catastrophic failure.
An analysis of variance vas perforned on the six (6) levelr., Ithree (3) replicates per level and there appears to be no
significant difference among the nean equivalent lives at each
Nr.
-?7
level. See Appendix D for details.
A reliability life determination is shown in Appendix E Slong
with photographs of failed 114 rifle specimens.
ti application of the life predicticn technique, used with the
2114, to standard cannon i3 shown in Appendix F. The concept 6f re-
tiring 1000 or more safe still usable tubes to prevent a single
accident as a result of metal fatigue has in recent years become
an acceptable practice. Additional safeguards are also provided
as a result of other rearons for condemnation sach as bore wear
(origin or =uzzle), rifling damage, rust or pitting, transportation
hazards, eneny action etc.
Phase III - "Comparison of 175M Tube. Forginp M.aterial Produced
by Three Contractors X, YZ Z."
Data generated from this experiment and a graphical portrayal of
this data is shown in Appendix 11.
Comparing 14 rifle material plot to 175M-1 gun material plot
shows more variation in IL14 rifle material llfe. This may be attri-
buted to the fact that all 1751M specimens were manufactured by one
machinist on common m chinery while the 1114 rifle barrels were drawn
from stock and involve a variety of manufactures.
In both the IL14 test and the 175M material test, hydraulic pres-
sure cycling was performed at about 3000 PSI less than the average
firing pressure. This might account for the 1114 2:1 hydraulic cycles
per firing cycle but then it does not account for the 1:1 relation-
ship shown with 175MM material with specimen fired beyond the 500
6
Q ^--- I - -
round level. See Appendix N for ammunition calibration inforwation.
The 175IM! material appears to be very seriously affected by
firing damage as demonstrated on the graph with the pure (unfired)
hydraulically pressure cycled specimens versus those fired to
destruction. Also the combination fired/cycled 5 rd., 50 rd., and
100 rd. specimens show a sizable drop in life as compared with the
unfired specimens. The 1114 material did nt exhibit this effect
although proof of one round may promote sufficient firing damage
to account for this condition.
In order to perform a comparison of life data within a forging,
among forgings from a particular supplier, and among forgings from
different suppliers, data was transformed to a common base as shown
in Appendix 1. The difference among forging life data are considered
normal and acceptable fatigue behavior irrespective of forging
supplier. Material properties and quality are considered to be
effectively the same. Appendix J contains life/reliability infor-
mation. Appendix K shows life and material properties data and pots
while Appendix L contains photos of specimens fired to destruction.
Appendix M shows information on a field failure and specimen
test results. W1hile the results cover a wide range, from 1676
rounds through 7810 rounds, the three out of four low life values
indicate cause for concern about material behavior.
S. CONCLUSION
The multi-level testing technique and 7.62WM bore fatigue spec-
imens, as demonstrated by these studies, are each considered useful
tools and reasonable (cost/effective) methods for problem exploration.
7
Typical problems for possible application being:
Swage tule vs hydraulic autofrettage tube life
Rate of fire effects
Effects of pressure levels
Changes in chemistry
Changes in heat treat
Changes in forging practices
Chrome or other type plating effects
Refractory liners
Cooling effects
Aging and rest between firing effects
lydrogen _mbrittlement
Long term storage chemical effects
Straightening
* Thermal treatment effects
Dirt size and ccunt
6. RECOMIENDATYON JContinued funding of projects using specimens of the type em-
ployed in this study in order to promote knowledge for production
of superior cannon tubes and safely exploiting maxinum life from
these tubes.
8
APPENDIX A
MrFABLIS!fI23E0 OF TEST TECHN7IUES
NO. PACE
1. Firing Fixture Used for Preliminary Investigations 10
2. 1.4 Wall Ratio Specimens - Air Cooled During Firing !3. 1.4 Wall Ratio Specimens - Water Cooled During Firing 12
4. 1.6 Wall Ratio Specimens - Water Cooled During Firing 135. 1.8 Wall Ratio Specimens - Water Cooled During Firing 14
6. Graph - Wall Ratio vs Fired Life 15
7. Specimen #5 Heat Check Pattern 16
9
9 A
I..
I it
CL
:1E 0
07A:
. fikI
7-62iM Ihdified M4l1 Ria-le Barrel~s-'-4 Wall Ratio FP-tigue Specimens
Air Co-oled, Rate of Fix-e 6 RdazzlinPioJectile Travel jic*u Ammmiition LOt LO .13023
Projeot N!o. 4332-1., Negative No, 29-c4-.69
4pec~nen #Life 87 rd
'NM I '1I
Li fe 389 rds
A-2i
7.621v2 I Mdified Nib Rifle Barre'l.s1.4j 1,441 Ratio Fatigue Specimens
Water Cooled, Rate of Fire 6 Rds/MinProject No. 0332-1. Negative No. '197-2--69
:pecimen #3Life 829 rds.,rnmiition Lot LC-13023Projectile Travel 10O-2-
3pecimefl #710Life U93 rdsAzmunition Lot tNRA-227l3Projectile T1ravel. 19 "e
Znecirnen gi11
Lire-OL*1 rds
? rojecti~e Tra~vel 1914A'
A-3?
7.6221 Modified MAlJ Ri-fle Biarrels1.6 Wall Ratio Fatigue .'-pecimens
W~ater Cooled, Rlate of Fire 6 Rds/fti?Prcjectil' Travel 0" Atiminition Iot LC 13023
Project 1b. la332-l,. Negative No. 1L97-1-69
Znecimen lLife 2756 rds
Dtt ienf
S',ecizen #6
13 A-4
~A
7.62i -bdified 1h Rifle Barrels1.8 Wall Ratio Fatigue Specimens
Water Cooled, Rate of Fire 6 Rds/KinProjectile Travel 19 ', Project No. h332-1
Negative No. 197-3-69
-r
Spenimen #7Life 547h rdsArlnmnition Lot LC-13023
Sp,,cimen #8Life 5093 rdsAmmunition Lot LC-13023
Specimen #9Life 707 rdsAmmunition Lot WRA-22713
A-5 14
L 11
7-62 MH M14 Rifle BarrelSpecimens,4achined to1.4,1.6,1.8 Wall RatioFired at 6 Rds I Min.Water cooledAir Cooleu'
I I t III I l l 1 , .: 1 - .1 - t i ll7
58 7 1 RDS ImG.
11 -it -
t m I
I TT I I I
2458 RDS PANG. 1 4 1 1 1 1 1 it zi
L
t i l l i tI I I it 1022 RDS AVG.I IT lilt IVw1 . I ' I _ f i l l -iL4I f I l l f i l i ! , t i l l : f l : t I i l l 1 1 l i l t e I l i t i f l .1 I i t
: 1 1 ; 11 1 11 I t l i l l , i f f i t 1 . J ; . I t , I t - I .1 1 1 1 -i f ;1 1 1 1 4 1 1 1 1 I l l .I ' ? I ' 1 - 1 1 1 , l i l f , , I l i l I L L L III
I f f i l l i t t I f f i l ' i t i l ! .f $ I t t I I ! i I I I j jj j I ' _.L I._LL1 1 l i l t 1 1 1 1 1 1 1 t t f i l l I t 1 1 .1 ; I .! I j t I I l i l ti f l i l l i l i l l I j i f i I I I , -I II f 1 1 1 1 1 ! 1 f ; -1_ --i. I I I I I l l j l_1 1 .1 mi t I f i l -I i t I l i ! 1 1 L S ' . I f i l l ' t 1 f l i f o li t L , -t i l l 1 : I f f i l l 1 ! l l i f l ! f f ! l I f i f i l i t
- -_ I l l 'i l -i i I I . I l i f f t l i l I f i l l 1 .-LL I till
fit L tI J I ' l l f t l ' I l i f i t , l i l t f i t
f l i f t l ! I l i f f : 1 : 1 1 .1 1 1 I l l i lI I f i t i 1 I i i i ; f i t : 1 i f II ! I i l l l l -l .. I I - -I- I ! . 11111 1 1 I L I I i L
f I I I I I I I 1--w I I - It I I i f t i f ' ! f - l i f o l l fl i f f i t t l l l l l l f l I ------ v t 4 l f ff i l l I l l t i ll I l . .1 . t l l l i l
If
ifi J T_ i
jqL
A
,I
:11
It
A-7 16
C --
~.4t -!
'it ~ rr ~ IC?
*- ---- itA - 7 1 6 F
1
:r '1tr .- -. 4, if 'js
-r * - -
-,-Z - -m
2 r ill G= -- o et c c
3. E-3ra~c Pe-=z Te=Sm=
[17
-U, / ,__ _ _ _ _ _h
i .,
T4
4-0
I - III
I TA
Aft'.jo R-2
44
B- -)
(PEASE II)
C(dIILATION OF HDRt=IC HMESSIRE- GC LIFE TO FIRM- LIFE
NO. PACE
1. Graph , -vydrPMulic. Cycles vs1Rounds Fired 2i
2. Graph - Avg HydrAulic Cycles vs Rounds Fired 21
o0
-14
2
21 C
= --( _-Z I .)
] -~
TT LL 1
or=Umm cmLms MM,01 Htil
7.621WO 1.8,,WAU UT3Ds, 104 RMFLR Nuir-M smb1m
M= AT 6100NDSMN COM COOLED)MIMMM= =WAT 7 ==/=
.- Rm. loW MMU=FUND crcm
1, Htt 11 1-' 1 Ll 0 12,3914, 0 15v582
0 8s552Tit M.
4- -A-LLL I sopo 02069V 1 12000
'1,000 110552,90DO 10s4622,000 7j416
TH ,..Ooo 5,5463,006 2s973
3sCM, 7;04691 3sOOO idi -M90
lilt, 4,oDo 2$939
t 4,000 2,,474
4,000 2,,587-Ll I
-2.21 X + 12,122
CORR, COXF. - -. 830............ BrWTM FROOF
Hm4R11SSwp& 44,309 PSI -
STD. DIV. PRE33M - 1,000 PSI
MN LOT WA-2k731lvii +A4+ + F -=m-l
I-FT-1- A I I 4-H4+
liltIf
71
M A-1JJ i I+
22
APPENDIX C
(PHASE II)
CORRELATION OF HMDRAULIG PRESSURE CYCLED LIFE TO FMRIMD LIFE
NO. PAGE
1. Graph - Hydraulic Cycles vs Rounds Fired 22
2. Graph - Avg Hydraulic Cycles vs Rnunds Fired 23
21 C
.... .i ... ..i l .I..[ : . - -: I--.-- - -- , - ,- -- i - • . . . .. . .
" .!"CORRETION OF AVERAGE HhRAULIC CYCL. VanuS ROM1,)S: -; FIRE ON't 7.62M4, 1.8 WiT., RATIO, Mlh RIFLE MERREL TEST
- ........
.
i . ........
A: ... . .. .
* . .. .. t " ' " AVERJAGE
... .. ... . " - - ..... .LES PER
A: ...... .... TES L-... Egn1_:L:__: ,h0•
..................................................... ....... .... ..- ,o .
* :::::::::: 7,808
L3 Ll.4 -1-.-4 .-1_.. 6 ,1h 4 -
4- 1. --
- "L -fT z :__'._ 4 -S....4 .
-____ . -j. --- i4 e- 4 - '+ 4 4 * -I
=. .-- : . ..'i, . . ... ........ * .- L , ... t j j .. , .... .- .L..4 .. ..
-4- FIRED O. . . . ... .. . .
~.** 4.4...4i~ 444 444 -4* 4 .
.fu 1 +~ 4,I .. .. . . I,,, ,,. .L4. , ,,,.=.,.D ROUNS (THOUSANDS) -
23 ('-2
• 0
APPNDIX D
(PHIASE UI)
CORRELAIION OF )IYDRAUJLIC PRESSURE CYCLED LIFE MM LIj
NO. PNCE
1. Conversion to Equivalent Rotmds 25
2. Graphs - Average Equivalent Rounds vs Test Level 26and Frequency vs Equivalent Rounds
3. Analysis of Variance 27-3C
D 24
i-i
-- : x~(2oi) = c'e)CWMMIT 2.2311A-F~
2 3 h 5 6
AVG SO lzw AVG MJS/CrImZ-1-S AM EZ7"
CMM~ Cmw-s- 1=69_ FPXLW%, FARMS
0 12,37-1 5,600 15,4R2 12,1 7,75 5,5%0 6,552 3,867
I,oo 3.,69 hva-32 1,Oo 8,9.B 9,h7o 5,G& 5,283
1,CcrG 11,355 6,135
2,Co 10,462 6,73,C3 2.,Coo 7shi6 7,28o8 5,,354 5,531
2,600 5,51 6 b,5
3j,0:) 2,973 ls,0jh3AW ~ 7,0&6 6,373 6,186 5,n82
3,00 9,101 7,115
5 1,ooo 2,h714 2,666 5,119 5,206h,0o 0 2,58? 5,1 70
5,683 0 5,6836 6,263 0 6,lh 6,263 6,114
,8 6,187
5= 25 -
D _ _ _ _ _
TS- LWE7A 1 2 ..L....4 .. 6* -
-711O 7,0b6 5,o64s ,554 6,186 5,,119 6,263-.4
4AV AE 520 5,p283 5j,531 5,882 5,206 6,1ri :
....................L FREDf-
fa26
I
ANALYSIS OF VARIANCE (ANOVA)
In general, there are three AhOVA classifications:
(i) FIXED EFFECTS - Wihere particular levels of a factor are
studied and the result is a comparison of mean effects of THOSE
particular levels.
(ii) RAIOM EFFECTS - Where levels of each factor axe randomly
chosen from a large finite or infinite population of levels and the
resulting analysis estimates the component of variance contributed by
each factor to the total variance rather than the main effects and
interactions at particular levels of each factor.
(iii) MI
The combination of (i) and (ii) resulting in a more
complicated analyses.
The analysis of variance technique enables L- to partition
the variance of the measured variable into portions caused by the
factors, singly or in combination, and an experimental error component.
ONE FACTOR ANALYSIS
PURPOSE ...
To compare 1L" different levels of a single factor having
I'M" replicates (repeat experiments) per level.
ASSLd>TIONS ...
(i) The "1' replications for any level represent a sample
dravm at random from a normal population.
27 D-3
DI
° (ii) (a) FIXED MFES
The "L( levels are systematically chosen as the
important (or only) levels of the factor.
(b) RANDOM EFFECTS
The "L' levels are randomly chosen from a normal
population cf levels of the factor.
ANALYSIS ...
Let . Level of S!Eificance
L - Number of Levels
M Number of Replicates
N (L)(M)5. Sample Size
Xi-- Replication t of Level iit7
Then, the partitioning of the total sum of squares
(S.S.) of deiiations from the mean into its components is as follows:
Source of Variation:
Among Levels I - 1 M )2,
S.S,(AMM)1) = L !0 (9J "ij Z
... associated degrees of freedom = (L-1)
Within Levels
S.S.(WITHIN) = SoS.(TOTAL) - S.S.(AMNG)
... associated degrees of freedom =(N-L)
Total Sum of Squares:
S.S.(TOTAL) At x N... associated degrees of freedom = (N-i)
I
D-3 (Cont) 28
--
The Mean Squares (variance estimates) are formed by
dividing the Sum of Squares by their associated degrees of freedom.
If the null hypothesis (H) of no difference among
levels (equal means) is true, then the mean squares (M.S.), among and
within, are independent estimates of the same quantity and should be
approximately equal (except for iampling error) with a ratio (among
M.S. to rithin M.S.) near unity. A ratio greater than or equal to the
tabled value for (L-1, N-L) degrees of freedom would occur in rrndom
sampling with probability 0c if H were true. Therefore, we would
reject the null hypothesis and say there is a difference among levels
(ioe., averages).
if an F - ratio loss than unity is calculated, considera-
tion should be given to its reciprocal being statistically significant
rather than accepting Ho imediately. If the reciprocal is significant,
possibly the postulated model is inadequate (Reference 2).
ANDVA, FIXED EFFETS, LEVELS 6, REPLICATES 3
LEVEL 1 2 3 4 5 6EQUIVALMIE 5603 4649 6731 434j4 5329 5683ROUND 7o46 5o64 535h 6186 5119 6263REPLICATES 3867 6135 4508 7115 5170 6487
SOURCE OF SUM OF DBGREES OF MEANVARIATION SQUARES FREEDOM SQUARE
Among 1,934,A9 5 386,890Within 13 106 290 12 1,092,190Total 17
29 D-3 (Cont)
D
Resulting F ratio is .354 which is less than the table
value 3.11 and therefore there is no significant difference amozig the
mean lives of the six (6) levels considered.
Similarly the reciprocal of the F ratio, 1/F is 2.82,
which is less than the table value of 4.68, and the model can be
considered adequate. (Reference 3, tables).
References:
1. STATISTICS MANUAL by E. Crow, F. Davis, M. Maxfield, Dover Publications
2. STATISTICS IN RESEARCH by B. Ostle, Iowa State University Press
3. EIiRIN3 STATISTES by A. H. Bowker and G. J. Lieberman
D3
D-3 (Cant) '3D
APPenDIX E
(PHASE 11)
CaREIATION OF HDRAULIC PRESSURE CYCLED LZ TO FIRED LIFE
NO. PAGE
1. Graph - Reliability vs Equivalent Rounds to Failure 32
2. Photos of Failed M14 Rifle Specimens 33-36
3
4N
VF
.31 E z
IE
ALL- u, -T
- ,- %= U ,-.re - ............_-_-, .
.1 1S-f1il 41 U .~alll t7
++. M '"ILI___ W .,
r - ---sgS*..
.- ,+
_ _. - , . .. -I :
,I,4;1~,: "-" ] " ..
U~ ~ T 7 J~ -L
""".!j - -g - -. 32 -
"" " .. t - --U . -- ' " .'
5 U.... U,, ..-.,+• * * +~ a * - . - .. + + , . ': ' f . t 1 ' ..
"
9., --i,. -4.+
.I- + ' -'- -++
S . U U ."- +-. *- .- . .. ."45++ ..
++ + -: - 5- 4 - - - .- +L.
._-. .-.,a ;.+ -__J _ - U 11..... +. . : " _+ . ' , ------ :..- --. . . . :+ ..-+ --'-
F-1 3?
NEW
tied;'!-- -I
r-eq
V3
~ 3 Fie :ds -amic Cycles 15,552
z
-- lw .2; Fed Rts 7y Tyle3 7,41
'Now___________
Icert
IF G&,-A ANI LX r 111 1
'P4~ Fired Pds 0 Hydrawlic cycies 8,55,
jjI
zecL~ijl7 Fie Ed ,O ydraulic Cycles 11,355
L
Speen #27) Fired Rds 2,000 Hfydraulic Cycles 5,546
i:3 Firel ?-Is 3,7"j- Hyrdraulic CyclIes ,
Reproduced ffom (best available copy.
35 F-2 (cant)I
ITI
EJ
9-0
ut
ia E0Z r..
U.2.
~~0
M :1 l . v l4 r V4X r
F-2 (ct 36
0 C, i .
az 0
r -
,.. . o
" > F
APPENDIX F
APPLICATION OF MU;LTI-LEVEL TE-TRIi.J TO STANDARD CANNON
NO. Pl.GE
1. Cannon 155"24 Howitzer HIAI Graph -Hydraulic Cycles vs Rounds Fired 38
2. Cgnnon 15511 owitzer MIAl Graph -Ieliability Predictions 39
3. Cannon 175'P1 Guns M113 & 11113AI Graph -Hydraulic Cycles vs Rounds Fired 40
4. Cannon 175MM Ouns M113 & Ml13Al Graph -Reliability Predictions 41
37 F
.__.a ,
4'
I CAWW, 155WH )m-irm XXIM
0 -MAUIM .i-a=a V.TM VMS MU MM A73-4PS
5,006 9,
o 16,553 -
~18~- *0 16,700
0 20,"91
0 22.632
400
i- lob it* 1 'iPm RO= 4'~dsks
r,- 381
CT Ea= W-TCEE1
4T4 ti~--
L ,:i
_1 1 7
_________ _________ Zs~
-A, ki4 -=T ElA
1373 Y--- .5_%
ft-A- -
-- T. -- 6,-5 35
11-73 1 r - 'r-
*** ~ ~F~1~_____0.5-* -44 *~
I .358
Jt 1 ~~E',TAI MUMS TO.~ _____ ___________
3,Y
F
, _:= .... ..: 6-n23 (GM I t--- ELFs-s.LI s M-. .. ..... - .... .. .' .. ..... ...-- -- .nl2A, '2? -
, ......., .. .. .. .. ... .... i .. .. , . . . . .[=---:-- !!:-.:.. ~~~~...... .: ........ 7 )h l a 86...... :::: : :: : C ..S CYCS Pz3s F. .... .... :331) 9322 1330 65 ':
__38_____16213 7586
- -- ;---4-- 3~ 1~o 330:::-!
.- :r .::: -:::::: :::3:::: :: z 1270 6317 ....... : _ _= ......... .... 2~.3 V4-. 9 = a o -?a . .
A21 73-
....... ...... :.T 231 W17 I27 2 5 69 .-
- --=G... .. .. .-- - %. ...... - -- ,--'--.4... .. . .. . . ...
27. 6 ..... 1270= 63 .
.... .... 57.6 . . _. . . . . .. . . .. . . .2:1 3157 1327.. ..
_ _ _ _" 30 ... h749
256 "M3 " L615===K !,.2.XP3, ... .. -1... ..a+7978i :......... ;. : : : . . . - : .. : : :.... :.: : : : : 2
-3 40
- T1
... .
n ..... ... ... .A .... ...
0 C' 0 0 OLsC 0 -C 0 F %0 r-C %C 4"- 4 0 4 -l r -
A-
0I WLll i LC% 0%- I - n0- c . .*0 .Ta
I i C -) L %0_ _ r - % R % C% C Qf---0 C%"- C 4r c) ,co 4 0 CO I -M -
0 1i
-I C-4 -
1 ~ ~ ~ ~ ~ ~ T cn_____________
411~~11 C% I ;d2 ~ wC241 ID
0 IV C!_______III C> IT
ci0.~-4-O~r.' ;;
U'
U44
APPENDIX G
(PHASE III)
COHPARISON OF 1751.0 TUBE ZORGING MATERIAL PRODUCED
- BY THREE CONTRACTORS X, Y, Z.
NO. PAGE
1. Specimen Selection 175f HI3A1 Forging [aterial 43
2. Phase III Test Plan 44
G 4
itt
C 42
1-1-4 %C 4M* t : ,( c nI n
r-4 4 ' N JwUI N IN N IN" tn IN
E > 0HE0
CU,
cnrX :u) o-H 4=
OC4% I Ln C4e DN0 Ar4C1C4 r%0z on %D rlDUO r- %D D D rz 4r r o lr a
E- ElZ)-
w (: raE-4'-
H / 0HO .. Cz
E,() ,
It c.J .C0 ' co %0 - Coj r44 3 tnn cr N-4tITcn O C a r4C UIA'Q C14 '
Nr 4 cot %0 4MC4IH0 % ot o CIn %D x % I'D X o %0 Ixo %D% 14I D X% r n %
.cr4 > - '
_z ( Ln a'-zZ 4 -1' CA
H E-4 xQE-
z ~ r, E i. E , n1-4E-40 H 0-4 r
(f)~ 1
,
0 L) -4 %0 U)
in
43 C-
G
PHASE III
TEST PLAN
FOR 7.62MM SPECIMENS, 175MM TUBE MATERIAL
CATEGORY QTY TYPE TEST
PbTRE 9 HYDRAULIC PRESSURE CYCLED
SPECIAL 1 3 RDS/CYCLED4 COMBINATION 1 50 RDS/CYCLED
2 100 RDS/CYCLED
3 500 RDS/CYCLED3 2000 RDS/CYCLED
15 COMBINATION 3 3000 RDS/CYCLED3 4000 RDS/CYCLED3 5000 RDS/CYCLED
REAL 9 FIREDTOTAL 37
r2
C,-2 44
IK
APPEN"DIX H
(PHASE III)
V COMPARISON OF 175M11 TUBE FORCING MATERIAL PRODUCED
BY THREE CONTRAC'rORS X, Y, Z.
NO. PAGE
1. Original Data - 1751 91U3E Cannon Tube aterial 46
2. Graph - Hydraulic Cycles vs Rounds Fired175MM MU3E1 Cannon Tube Material 47
44
A
St
'A
€I
S45 1'
-
H, L=7 ff -% l Io Cr Q-ZMLZ0
D-4 -M ~~~
A U >z m-; - F-
"U tm 0w_______I ~ .2 ~- ~ ~u~ Ad1cI I
_ a
Cl zaC-~- 43 cm - - (.5 Ca~ C.1 't
I3 C
46 --
_a
__- 4oM AT .v
Zf T
-MX
I A.; zi L
_ _ _ _ _ __U
_ _ _ _ _ _ _ _ _ _ _ ~1~47
2. A&ISme~ J41e M2a ~
48
CAIe-C1- 1 I -%: 0a l .rO
.= - nc .v z% a- -uan
of " rW 0 GC 4C4 l n%
%-II cl r.1 onC - - -- -4nco =-%0 " .:ri nt 0 19c
cm~
v-a 4 r ooce0000
O r C(10 CI C 10 an ON0% 3C >C ,
0 cal
n% 0 C -) C~3r.- r- SO 0
C449
-1 -4 C~ -;T%
ADJbSTED LIFE DATA
SUPPII __z
b-4R :,..I a
E-E-
ALL 1 X1 6208 13 Yl 6956 25 Z1 7433
Fl-ED 5 2 6124 17 Y2 7233 30 Z. 68699 X3 7696 21 Y3 7274 33 Z 6918
2 X1 7855 14 Y_ 5210 26 Z1 6866ALL 6 X2 6948 19 Y2 6366 31 z2 686M4CYCLED 10O A3 8199 22 Yj 7040 34 Z 3 7365
ADJUSTEDCOfIATION 3 X1 6583 15 YI 6673 27 ZI 7352
FIRED h x 6937 16 Y, 7215 28 Z, 7243
7 x2 6240 20 Y2 6281 32 z 2 645011 X i 7314 23 y- 7507 35 Z3 6964
CYCLED 12 x, 6828 24 2 7146 36 Z3 7795
ANAYSIS OF VARIANCE
F-RATIO
ALL ADJUSTED X 2.11ALL ADJUSTED v 1.81ALL ADJUSTED .009
ALL FRED xYZ .65ALL PDJ CYCLED XYZ 3.56ALL ADJ COMBO XYZ .84
TABLE VALUE 5.14
In order to simlify, adjusted combination fired andcycled X1 Is, X Is, Y s, Y Is etc. were averaged toprovide three eplicates par c-i.
1-2 50
l''i lll" i' i :1 i:+l . i I'" T
u i i--l'tT 1 i 'i 'f" il:m :€
]I
APPENDIX J
(PHASE III)
COMPARISON OF 1751M TUBE FORGING MATERIAL PRODUCED
BY THREE CONTRACTORS X, Y, Z.
NO. PAGEC--1. Life Characteristic Comparison by Material Supplier 52
(Using Adjusted Data)
2. Life Characteristics Comparison by Test Method 53(Using Adjusted Lives)
3. Graph - Rel.'ability Prediction 54(Using All Adjusted Data)
51
N~I
Z. * T COe CO CCrmnu T% 4 - --C4 r. z-- 'I m r V.',r-C-% % 3% %3%
C*4 l cl -IT -a n gl IT 11 i %3C, % C-E n* - %< n c '-l 0 C m 0 , %T i ng . 0 c o
I-I CzO C.)) 0-U 0 K39.F-4rAA Ea C 0 ~ c r'- U D' i d n mt D co r, cr~ ~ o ' o 0 <3Ad~ E-4 0 -Z01r-. n ci C n C 1--T c)CAc 7 d l 3
1. ':4% V % o 1- o % n G - C ' ~ 1 -
., F4 0 c
f-4~i-41
E- C%' C% -8C - c 3C4Lnr %
-0 0
0 c< F4
3--0 4
-4 1-4 E-4'a -4
o .- i 0 nr
0 -:4u :0Z
0 N
AN22WUd ZALLyifl34fl
J-1 52
r:;'
r-~' - C
113v P0 I. -; j
I-i A
CO*~ co C) i n C c~~C. % 0 -
0 1 1
E- E-4
'0 C-
U
53 -
110 "~i coC * 30
C;C;o:ozWoo;-iV:ic C z~OO OOOOOO ioocw;,zE!3 04- iiC --I 12- C- 1Ww 1% 04 1 iiBr mC oc*1 _ _ _ -Z_ _ _ co_ _ en M_ _%_ -c = i~ 3 .oitC2 - C'oCaa - rii'ik i %
_ __d W _ _ =_ _ %Z_ _t nc iii.U4 6 1 I %'0' . - 7C1C %a nC
VAM C6% . 0 % D4 0% 0Iiii ~ii.r CiI.Ci . r f ii iiiC.f C
lit, 1t 41 j2T_________- - r~j t~i4
a 1-zji ar
tiiif ___ ___
1 .4 4 %
in -4c 1J N1 c i
00 0 0ac .*4r 44 r4 *) . t ft00
544
iA O c
MMM F- 717MM TS-BMR~ MUERULPZWCCI
55 K
ac =Ij~f rr4 0 .
~ ~ ~.0- .000 -r-.0. 4 &- z~
-N ci on(
I II-E 1101 -" M -§8
%2 w .% C a- A- 1 i- - . -
0%-30 0 r-rjZ-hr - -
C-0. 0;t c: 0 ,
o 0o
ioI
F-A
in C* Is
c- C C
0 C!
c 0n 0~ In o i C.
o9 co I0.~ ~~~~~~~ .~.CIC .O-~ .O~ -A.I c .VAO . C a . .. .. .. ..
.~ ~ ~ ~ I~t--C~r'~C-%-t- C- .1A--O.O0. c'%O 0 % %0.JO
=0000 0000 IlOAIl00~00- 0tCOAs 0O*o0,
ur w .~nI IAAIU 1 IIIA
E54 ( _a = - -m e
0 It N~ C
013 -I 56 0 0 1 lc tU vIC
LIrb VS. MMUANCAL PROPERTIFS
........ . FORGM M WACT!I
Life VS.Cha.-py n A-.
.--......... ... ....
.. . . . . . . . . . . . . . ... .....
11~~ ~ ~ 717-L. .. ..
Av._ A._ .: . . .
Ufa~~~"* VS-hr ........* Ag.TS~(1~3 (PI)......... ... .. ..
~~~~~~~~ .. I ..-' .I I4 .iAg Sx1 3 P
57:
K+-:+-+: : : .... : .I-.- - .... I.... + : : ,+. . . ...:
44
, -. : .. ... . . .. .. . ... .t t 1•. o
N. - - - - -.. ..
..... .. ._ ... :..
Z ~2 . .. ...
.. .. .... . .... .. . ...... ....
.+
--.... . ... .. ... .... ... ........... ..
... .. . .... . .
_ _ _. . . . .. . ... .. .
K-3 58'
DI
APPENDIX L
(PHASE III)
COMPARISON OF 175151 TUBE FORGING MATERIAL PRODUCED
BY THREE CONTRACTORS X0 Y, Z.
NO. PAGE
1. Photos of Specimens Fired to Destruction Material 60Supplier X
2. Photos of Specimens Fired to Destruction Material 61Supplier Y
3. Photos of Specimens Fired to Destruction Material 62Supplier Z
9
Ii59 L
I REDMIZ
Repcodrea Fmavagabe coPI
[I IE ' o
low-
I-mm"
Fri i;
C, IMWS
avaulag
C; ;Ixovce 19---~---
rIED*C~114 inI~L
15ICAMNO CF 7.6ZTK S-FECUCENS TO A FIELD FAIUME PrFMBfLU
1:0. PACE
1. Field Failure Motes~ 65
2. TdIre Sectli =d Specimen tocation 66
3. Failed Specinems 67
4. !%cL=IczI Pro;,erties of ForizinF 68
63I
CqrO a be COP*'lbes'
-- :46
le .%, %b --.-
2(
0101C, I9
eI
ellgee
CA')
R SDS.1
4
9BIReroucd40o
3et3val4le0oy
14 C1
u C% c a% M
z
in n 0~c
00
*- I, *0 *
0o 0 0 0 cI-4 -* *
E-4 N c~ cEn'
67 -4
N
APPENDIX 14I 7.62Mq AIMITION CALIBRATION
NO. PAGE
1z 1. Ammunition Calibration 69
N 6
I'I
\',I N 681
N
ANIMONLi~Q CALIB.7ATION
60'% 504Eu
~3o
~20
•.5 .C 1.5 0.c
TnzM(,EC- x io-3)
TYPICAL P-7 CbU'VE
7.62121 (1114 ifle M.t'l) Test !zrno Lot 1.-'-731
Calibration Date Mean Pressure Std Dev. Mean Velocity Std Dev.Original 8/67 h4,300 1000 Uot Avail. Not Avail.1st Check 11/70 47,800 1300 2701 162nd Check 2/71 47,900 1100 2740 28
7.621I (In13El Cannon 1t'l) Test A-o Lot "Rbl 22896
Calibration Date Mean Pressure Std Dev. Mean Velocity Std Dev.Original 10/68 45,3C0 1000 2751 19.71st Check 11/70 47,500 1600 2687 282nd Check 2/71 h3,300 1200 2731 19
NOTE: Check tests were performed by Frankford Arsenal.Hydraulic cycling pressure was established on original ammunitioncontractor calibrations.
Q, Nl-1
._,- -, -_ .. ,''..,,,; , . .: , ,,':, ,:,_:<: , ,, ,, , ' 1 '' ' ..... - i' '' : o ,' , ! , , u.. .-
APPENDIX 0
MISCELLANEOUS REFERENCES
NO. PAGE
1. Miscellaneous References 71
0 70
• , f]
00
MIISCELLANEOUS REFERENCES
1. "Establishment of Cannon Life Criteria" -R. E. Weigle 1967Watervliet Arsenal (Plus additional R&E laboratory test data).
2. "Fatigue Characteristics of Open-End Thick-.Walled CylinderUnder Cyclic Internal Pressure"; T. E. Davidson, R. Eisenstadt,A. N. Reiner - 1962 Watervliet Arsenal IYVT-R1-6216.
3. "175124 MIIA Gun Tube Special Test for Service Life";B. B. Brown, T. E. Davidson, D. G. Forkas, M. E. Kraut,T. Moraczewski, A. N. Reiner, P.K. Rummel, R. Soanes Jr. -
1971 Watervliet Arsenal.
10,
S71 0-]