ML20214Q890

MO-7068 Thermal Degradation Equivalency Evaluation

ML20214Q890
Person / Time
Site:	Big Rock Point File:Consumers Energy icon.png
Issue date:	09/19/1986
From:	CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.)
To:
Shared Package
ML20214Q878	List: ML20214Q875 ML20214Q888 ML20214Q890
References
NUDOCS 8609290006
Download: ML20214Q890 (25)

Text

--

(_ . ,

ATTACHMENT 2 Consumers Power Company Big Rock Point Plant Docket 50-155 MO-7068 THERMAL DEGRADATION EQUIVALENCY EVALUATION September 19, 1986 8609290006 860919 PDR ADOCK 05000155 P PDR OC0986-3034A-BP01 24 Pages

MO-7068 Thermal Degradation Equivalency Evaluation This evaluation is to provide an assessment of the relative degradation that is postulated to have occurred to the MO-7068 motor operator and its non-metallic parts during the 13 years of installed operation, the design basis event simulation, and the subsequent operation for the past 11 years. In addition the analysis is used to estimate the potential degradation that the operator's non-metallic parts may experience during a second postulated design basis event plus a 30 day post accident operability period.

Reference 1 provides information from Limitorque Corporation relating to the types of non-metallic materials that were used in the 1950-1960 vintage SMA-Limitorque valve actuators.

Reference 2 provides actual thermal aging results of accelerated aging tests of Buna-N. The results of this test indicate that Buna-N has a qualified life of 65 years in 90 F ambient. The activation energy value used for Buna-N is equal to 0.86 ev. The activation energy is the amount of energy required to start a reaction. Utilizing the Arrhenius Methodology for thermal aging, one can predict the amount of degradation that is likely to occur when a specimen is subjected to a specific thermal stress. The Arrhenius Methodology is a single reaction event. That is, the Arrhenius Methodology only models the aging process induced by thermal stresses.

Since Reference 1 states that Buna-N was the material known to be used as seals during the time period in which the operators were supplied, we can assume that it is also the weakest link material within the component. Other materials known to be used are presented below along with the activation energy values. The values provided were obtained from industry data sources such as EPRI-1558, "A Review of Equipment Aging Theory and Technology".

Material Activation Energy

• Buna-N 0.86 ev

• Phenolic (general purpose) 0.96 ev

• TW Wire (thermoset plastic 0.99 ev PVC)

Formvar or Formex (class B 1.04 ev motor insulation)

Silicon Rubber 0.94 ev Polyimide 0.93 ev Polyester 1.00 ev Polyethylene 1.11 ev

• These materials are known to be in the MO-7068 valve operator. Additional materials presented are typical of other materials and their activation energies that could be used.

Utilizing the Buna-N as the weakest link material, the analysis provided in Reference 3 provides a step by step evaluation, using the Arrhenius Methodology, MIO986-3028A-BP01

2 of the amount of degradation that has been postulated to have occurred because of being installed / tested and a prediction of what is reasonable to expect given a postulated event.

Although not an exact method of predicting age degradation, this analysis can be used to estimate what can be expected. As can be seen through the analysis after both operability periods (24 years), both accidents (one actually simulated and one postulated), and a postulated post accident period of 30 days, the results indicate that approximately 11 years of life is expected to remain on the weakest link material. Although Consumers Power Company does not intend to use this as a bonafide, qualified life, we do conclude that the actual and predicted rate of degradation is reasonable and provides reasonable assurance that the valve operator will remain operable throughout a short usage period of a matter of months until replacement can be effected. On this basis, Consumers Power Company concludes that the MO-7068 motor operator is operable.

l O

MIO986-3028A-BP01 I

I L___.._...,-.__._-~~._._.__-_.._._...-.__.._._.._. . _ _ . _ _ . _ . - _ _ . _ . _ _ _ . _ . _

LIMITORQUE CORPORATION 1;WE E' .

5114 Woodall Road . P. O. Box 11318 Lyncroorg. Virginia 24506 M .1 Telephone-804-528-4400 - Telex-82-9448 Re(eme. 1 -

February 8,1984 Consumers Power Company 212 West Michigan Avenue Jackson, MI 49201 At tention : Mr. Leon F. Murray - Staff Engineer - Electric Transmission Engineering and Construction Department Gentlemen:

Subject:

Big Rock Point Qualification Information Regarding response.

your letter of February 3, we are pleased to provide the following The SMB Limitorque valve actuator is the next generation valve actuator following the SPk model.

The HBC is an additional gear box converting rotary input motion to 90 degrees output rotation for use with butterfly valves and other similar type valves.

The SMA Limitorque valve actuators were supplied in the late 1950's and early 1960's. Due to this fact, we are addressing the actuators in a general manner.

A.

The switch insulation materials used in the SMA actuators were phenolic.

We have no additional specification further defining the phenolic used.

However, since phenolics are currently used with success in the Nuclear Industry, tnis information would probably be of use in*your qualification analysis.

B. The seals used in the actuators during this period of time were made of Buna N. In isolated cases, other seal materials might have been used, however, our records are not suf ficiently complete to permit us to determine where this might have occurred. Generally, even though des.gn configurations are slightly different, the application of seals and gaskets was very similar to that in use in SMS actuators currently being supplied which mignt permit generalization predicated on current qualification testing.

Regarding lubrication and maintenance of the SMA Limitorque valve actuators, we would recomend following the procedures shown on Pages 11,12, and 13 of our Bulletin SMBI-82A "Limitorque Type SMB Instruction and Maintenance Manual."

Automated Valve Actuators and Jacks for Industry

LIMITORQUE CORPORATION Sheet No. 2 Please contact us should you have any questions.

Very truly yours, LIMITORQUE CORPORATION 2 9e rwk J. B $ ab Special Projects Engineer jhb

- . . ... . s.. . . . _ . . . . . . . . . ,

- - y;t+ -y- -

,

• 9AL.ts Ao es q g x-A n :

m 7 < .>

s' {..it'ca g i

l #( '

k. fk -w. -

l

&h. l 6.t4

. w. c 9~ cN84 ,

e-: . : 4-n y C.*.. . R.s.cC-TR.47 420, vor - it part Iof% 3

.' PW.

%,*.s.A 1

5N@

ev d (N-ie .

i F.inal Rapart m- i.

I I@g rI.@$

}.? d.. $,..DM, O ~

. . .c t-~ $- A y .a=.:.< & ..,

R, ecenenxe. 2.

! '..:-5.E ACCELERATED TESTir4G TECHNOLOGY t a. ru VolunM: il Handbuck of Accelerated Life Testing Methods

f. . .pt-

. ,'~.t

- a,t -

,t UC .q, W. Y::rkowsky E-99,jd et al Nghes Aircroft Cempeny tg ' :.*:.E].^ '.

' ' ?*, 1 Ip; 0.~ . 5

• TECHNICAL REFORT NO. RA00-TR-67-420

.a . :)

s . r. ..':~n# Note:nber 1967

. <1 '

.-. k:$ # -

• c .: . ~ )

• This document it nabisct to special

..;. f . j -

expert controls and each Isansmittal t.,".*q .,a- . ,.. y to foreign g;verraer.ts, foiring ns.

f Tia. ? ; tio.* sis or re;tesentat: .es tti reto may

, . ; *.' s be pia'e only with prior 9 proval of

. f,._;' ? ~ J R ADC (E*.tEi1N), CAFB, N.Y.

* -(fp The distributiori of e. Sis 62cvment is 1-,C limited beca'.se the informatics as ~~

<.--*.P . estarlto-d un6:* the Department of

,g -a,A " tate ITIARs.

.-, . s1

.7:3 ,

-. 1 M.'il 2 C ' i d-M .R

. , . a ? . ..

~ ~ ~ i. l

. i. ~

' u ', .m

.:. is

'?*..l

. *M':.1 .. ,.

.,..y ..,

- " - Rome Air Oeveisoma,$t Center C he Air Ferce Systems Ccemend

's A - Criffiss M. Force Bess, New Yc k I BECHTEL POWER CORPORATION I~ CENTRAL LIBRARY

^

- " ,3 777 E. Eisanhower Parkway c.1 1

. Ann Arb.or, ..

MI 48104 V

j , ;- , .,. . ,. . . i .^ _ .r- -s

_ a., f C.Stafford' - .

e I

a

,. ; .. g I. , ,i1 .-

3

_ _ _ - - c l

..- , 4.8 PARf NAME AND ECRIFfr0N: Meches. M Mustfon j

ical Seals (0-Rings) Assurrptiens Generally Accepted I y Anti 2275-ll; OD-3A", ID-9/lfi", Empirical l'nmf X '

)

. c mss sectional vidth 3/32"; Algorithm X

' anterial: Buaa-3 Rubber; applied as a flanged cover recessed gasket Physical Wele! 6 l

seal against hydraulic pmssum. ~

SotDC3: " Final Report Tor Accel.2rsted Reliability Test Methods For Mechanicsl arri Electressechanical Parts", RADC R-654, 1 January 1965, Refannce so. 298.

)

9 PJRPOSE OF TEST: To develcp an AIT method beaed on time trsnsforma- j tions of the distribution functicas of parts tested  :! .

to thilure st overstress conditions to yield p anti- e' tative eatimatas of rel.iability chancteristics at j

] ra:e4 stress levels. s 1.

t1 crc cr VAumTIc'r $

1.

Validation of this ALT is not cz:uete, because the schedule of the l. '

. study progma qan which it is based did not permit life testa of the 3-0-rAngs at nonal co?*.itions of te=perstum. Therefore, of the thme (*

proposed m2.els developed in the study, which 4.-eindad ALT's for other .'

typea of parts, it was not possible to deter::ine the modal applicaole to the 0-rings tested.

') .

ECRIPTIO?! 0F N fc1TDD n.

f.

, 1. Definitten of hilum In the application as a flanged cover recessed gasket seal assinst hydraulic oil atstic pressure, the definition of 0-ring failure is ,

that thr.m wt'.1 be no Whase for three test pressuro cycles of '

one air.ute daration esch, during which the pressure on the test j fixtum is raised to 1500 PCIG and held for one minute a:4 returned j ;.

to O PIIO. The mode of failum is radial f:seture or cracks and/or  ?

' circus *erential cracks ia the 0-ring

• O

2. Method of App 1viru Ac ele sting Stresses

. a. E m s Selection ,

g 4

- Very little infor:stion relating archanieml life to environ- )

.asntal stresses was a m ilaole. An 1.vestigation of pctentially i:

usable accelerating stresset it:11cated the fouwing general  % :,

Properties g

.4-

'N

. W 149

?:t d-

~~" -

i

..-g, t.

= ..: .- - _n

- ** ^ ^

e* .

, , _ , ,_ ,.,,,,,,--y,, , , _ . ., .,,,_,,.__-___,,.,,m,,,,,,.-,,_,c , , .-_,_-.mm ,_9,,a,_ e em. ,._-,_r -

- g B

- I Tes.r resistance fair ,

Ahrasion resistance good Aging (trltavinist) fair l Oxidation (Czone) fair

{

Resistance to compression set good 011 and gasoline resistance excellent -

Acid resistance good  !

Coldresistan:e(freezirg) gooi t'

Eant resistance good l Permeability to paes meditna '

Electrical resistivity Icv Resistance to cutting good .

, Kesistance to vater swelling excellent An Wnation of the abeve properties resulted in the elimina- i tion of tbooe that veuld be difficult to re,est with confidence. I Also eliminated were stresses that would require re::crval of l the rings frtza their test block, because handling vtnld be a  ;

difitcult-to-ecstrol cocdition.. Since a' combined envircr=sent  :

test tsa an objective cf the project, knavn ncn-cce:patible I teste were rul.M out.  !

I This resulted in an initial accelersting stress selec+. ion of:

f f

Aging (Ultn Molet) l 1

0xidation (0 zone) i Esat Resistance i a

i While conducting tests to determine strese Irvels to be util-ized, it vna found that total bramirdawn of the ozene ocr:urred .

' at tempe.atures above POO"T. This resulted in elimination of

• otor.e ss.an accelerating stress. !;eferences indicated that Puns N rubber ccarpounds are relatively insensitive to daange ,

• at eerperatuns g below 250'3F, therefore samples were tested s.t 300 7 and it was found that fatim occurred in less than 88 hours0.00102 days <br />0.0244 hours <br />1.455026e-4 weeks <br />3.3484e-5 months <br />, .

I 153 i

I

• 1

.. .  : l

- -. l

. i ee -

, ,- , -. , . , . , y a --

,g.

f . . .

U:i

_ . am _% y_

=un-er -

i It was felt that measuneent error :esulting fresa the ec:sbina-tio.s of beating nort ecoling cycles involved in the teating g

]

proced re vould be too great with this short length cf life j a - spea. Subsequent teatr and 1nformation led to the ope *:ifica- as tion of 275 F as the upper limit 200*F as tne low? limit and n sa inter:sediate condition of 250 . All availahle infer:ntion d indicated that a lowr temperatun than 200*F world not produce $

failures within the contractual period. The level of the ultra-  ?

violet exposure was deteratined fra's a sts:dy of marraracturers data on the light sources.

a A sicumry of the overall stress etlection is: l a

4 Auplied Stress 1 Not sal - Inter =ediate Mari-sm -

2 Te=merstu'1 200*F 2SO*F 275*F I1 Ultraviolet Exposure None 0.1 vatt/ft2 a

0.2 vatt/ft*- t

b. Stress Apoliestion  !

s A2 3 factorial statistical experinent with 10 replications ve.s &

(

• selected fer the study progrsa on 0-rings. This was to perinit e eval:.ation of tra main effects of temperstve and ultraviolet as well as their interar.tions vtan applied as cczabtred stnsses. (. ,

i Table 4.8.1 describes the different co=bitattacs of test conii- i tions included in the test program. f:

k As previously stated, it was impossible to operste the parts at i j '.

temperatun s lover than 200 F and still be able to observe a sufficient ta:nber of failures during the contractual psrioi to [

perform an analysis. However, ranges of ten:perature and ultra- E:

violet exposun were included as test conditions which would

• yieli a general ides as to the mancer in which life varied ess stress levels van varied. l-f:

TAEI2 L.8.1. FAC:0 RIAL FXPERDO.Tf FOR 0-RUCS 9

Ultraviolet Exposure Temperature (*F) None

- .1 Wett/Ft.2 .2 Watt /Ft.2 L .

1 e 200 (1) (2) (3) [j l .

250 i t

. (4) (3) (6) Hj t $-

275 (7) (8) (9) [, l l

?"'r

3 1

f (1 151 ;1 ,

- -_ s L -

1 1

4

- - - - - . ~ -.-~

-. , -- . . . _ . , . _ _ _ _ . - - - - . .. , ~ . . . -

e..-~. . , ..,3 : ~ , .- .-

~

9

. )

1

• I I

I I

i 1

e

. Essa:

• no :sasbers in parenthesee are designations used throuatmt

• l tas study to iden.ity smch of the nine test runs. For = = p1.* Test

• I l Ft 5 consists Each test ofrun 10 con 0-rings *ained which were tested at 2$0*F and .1 Vatt/

10 0-rings.

I l

c. Tsst Equitanent and Physical Test Method g

Figun 4.8.1 illustratae the asthod of environmental stress i application. The outline reprseents the tempersture centrolled ,

even, which saa set for the zwquired tampawtiu s stress level.  :

The ebamber vna ret up to s".maltane.ausly supply the three  ;

levels of ultrsviolat exposure by using an opaque shield for

' ' - the "no ultraviglat" leval, setting the bulb-to-o-ring distance d

{

far 0.2 vatt/ft , and providing g W. distance-frge-bulb ,

relationship for the 0.1 vatt/ft and0.2vatt/ftd levels.

• The bulb-to-o-ring distance was pededically reduced to cas-  !

pensate for the light outpat decrease associated with bulb ,

a612g. .

i

'3ae 0-rings were ipstalled in individual test blocks ani vere .

not re.1 red until failure was evident. This MMet ed the possibility of damage by mishandS.ing or from installation and rencval. The test blocks were suunted on the pressure test * '

, fixturs illustrated by Figun k.8.2. A hydraulic pressure of 1500 psi was applied for a period of 1 mi::ute and then re-

• 1 eased. The cycle was repeated 3 times fcr encs test cycle. ,

The 0-ring and test block were resoved as a unit and washed ,

with gasoline to removs the hydraulic cil. After drying, l 0-ring and test blocks were placed in tes appropriate environ-ment chsder. Periodically the test specimens were removed  ;

fr:m the environment chamber, allcwd to cool, pressure tested, '

cleaned and returned to the envizcament e' amher. This was

• reposted until failure occurred.

i The following items of test equipmant were used in the tecting .!

procedure:

3 Ites Function l N Tenny Oven Temperature environment k 520-011528 I f

Dispatch Oven Temperature envirossent serial f52517 S I

, U1.travialet Isap Ultraviolet ersvironment

. 0?f5 Crosby Modal AIN Indicate test pressurs 1 90 pas sage

• Blacibewt Model F151 Rydraulie Test Pressure Source Eydraulic Pump 152 e

. . e , g e.

- - - - - - . , , , - , . ~ . , -

_. , , . . . .-,,.m__.,.,_ -_._,.-.,._,.n_,..,

- . - , , , , , . _ _ _ _ _..v_. _ _ _ _ - , , . . , a.,_.--_ , , - _ _ - . , ,

. t f  ;

. .- e -

i ,

M .D

?

L f

't i

% 1 3

.1 '

SUMMf 0F RMTS -I The study to find an accelerated reliabilfty test method for 0-rings 'i was performed by test ng 10 parts at each of 9 different cabir.ations

• * ]

of two t; pes of stresses: tes:ytrature ami ultraviclet exposure. Each '

. stress was apy. lied at 3 levels of severity. The mean of the tests on f individnal parts at each of the conbined stresses is abovn in Table 'I 4.8.2. 9

- ',i

1. Analysis of variance Before an analysis of varianca cou.ld be perfor.wd on the 0-ring life test d,*ta, a logsrith=1c transfomation ha.1 to be made on the data. Tea reason tor this was that t2.e variance of the longer life v.est rans was mch h3gher than those tested at accelerate'i conditions. The analysis of variance iniicated that both ter.pera-ture and ultraviolet and their interactions affected the life of '

O-rin6s. The details of tne effects an shown in Table 4.8 3 I The F Patio indicates that both temperature, ultraviolet exposure, s.nd T. heir interactions are h1.shly significant. This is evaluated by eczrparing the values in the column in Tahle 4.8 3 marted F.05 $

to the values in the F Patio column. A larger snaber it. *.he F 4 Ratio column denotes significance. The Cc=ponents of variance ~

smalysis substantiates the values of the F Ratio but poin*s ou,  !

that temperature exerta un extremely powerfJl effect on the life i of thest, parts (about 96.L% of the observed varinzr e). Regard.less j of the overwhelning temperature effect, tta effcets of ultraviolet  :

and their inte:setions are large enough to be felt. The ri6ht hand e coluz:n in Table 4.8 3 sives a czplete list of the % :entribution to total variance of all the effects. {.

k "LU G k.8.2. 0-lEDC3: 5:PJR$ TO FAIIEE  !

No 0.1 Wat*g I Te=persture (Titraviolet Per Ft.'

0.2Vatg* g Pe- Ft.  :

1 200*F I = 1751.4 I - 1395 4 I = 1297.0 250*r I = 327 9 I = 340 9 i= 347 3 I 275*F I = 148 7 I = 136.4 i= 134.8 s' I.

~  ;-

.)

L

.}

'}~

153 l

. 'I, ow _ _ __

--wm og g i .  ;

1

i

. M 5',. [ - *

• g' G- . .

. t.c .m _

l.

TAEt2 4.8 3. AE":I:3 0F YARIANCE RDCS i Source Sun Degreen of $

of of Mesa F 7,oe' Contrik tion l 7arianco Squ: no Nee &sa Squares Ratio To Varicace r,etween 15 90668 2 7 95334 h791.2e 4.00 tarperstuns 98.4 l aW> .08211 2

.041f4 24 7e 4.00 violet Isvela *5

~

Interaction .03199 4 .01600 9 6e 2 53 Between 5 Tc=parature .

and ,

(D.traviolet . I i

Residual .13h73 81 .00166 '

.6 Tc*al 16.15551 89 100.0 $

t

• Danotes significance at the F.05 1*'*1*

2. Failure Analysis of 3. neem .

Esc.h 0-ring, a' rter D4149g the presr re test, was inspec . A under

• a microsecpe to datardne the nature of the failun no.da. There -

are only two failure modes that appearei predenantly. They ven: .

1. Radia* fracture or deep radial crack 1r.c.

i

2. Circ.nferential cracks usually on the outside dienster of the 1 ring. ,

• l'

,t De circumferential cracks that appear occasionally throuahout all kt Test Run l+ vere located r= Manly along the outer surfaces.  !

They usus11y, kt not always, were cons: acted to deeper radial .t cracks. Ce cire':n.ferential cracks periodically ensand the mold

. i line on the outer disamtar of the 0-ring. The severity sai fre- I 6 'quency of bota the radial and the circuarenntial cracks did not l seem to fers a pattern that increased in severity with stress.

The lone ext'e:2 tion to this was at 275"F vhers a larger proportienal -

amount of cirramferential cracks occurred. .

3. Analysis Related to Life Eintribution Tunctions l

The Impac. failure data for 0-rings vna plotted on the Veibull probability The lines of best fit were esL.ulated on the computer by  !

the method of least squares. , The Weihil abaps (S) and scale (a)

. 154 h

v.

,e p.

y e tr- - -

4._

. 9 6

fiN 0

. parumsters were calculated and are su madzed in Table 4.8.5 3 Alpha (23) is the coded value corresponding to tha scale paraceter 3 that can % obtained frem the Vei' cull plot of each test run. Alpha )

(3,) is the uncoded value of the scale partaster. 'j SUNRI 0F VEDUU PARAMPTIK3 FE,M 0-RIM TOS3

- TABLE 4.8 5 )

i Te=9entun ultraviolet Exposun i I

(*F) c UV .1 Watt /Ft2 .2 Vatt/Ft2 .'

i 200 S - 17 75 S = 10 71 S - 11.01 3 "3 " 198930 83 - 43,424 a3 - f.2,6h6  !

ao 35,k36 x 1p53 a , - 65,324 x 1029 a, - 24,271 x lo33 2 250 g - 3.46 g - 8.12 S - 22.k6 I

a2 - ic2,7ko a2 = 25,33o a 9 ):

2 " 2314 x'10

, e, - 65,069 x 1017 a, - 43,618 x 1016 a,- 19,248 x 10 41 275 d - 14.65 g - 7 94 g - 5 17 i a a2 - 15.ks7 a 2 - ho3.h3 ,

2 k 95 {

a, = 92,426 x ic=7 a, - u9,ul x lala e, - 106,326 x lo6 ,

. o I

4. Selection of Accelere f.ed Test Method (0-Rings) f 1

It vou11 aopear tlat any of the stresses induced during the o-ring  !

testa vould result in an acceleration of the failurs of the parts. l Te:sperature exerts by far the most sightficant effect en the life i of 0-rings, but ultraviolet has demonstrated its usefulness also as an accelerating stress. The true validation of vnether or not the results found in this study are translarmhf a to the life of I 0-rings at some other ranEn of temperatures cannot be anzvervi. j The basic relationships shown bere doubtless vary for ;mrts pro-duced frasi different materials. However, the main objective of l the test has been accomplisbed - namely de=enstrating how the life of a given type of 0-ring varies vten aulajected to a given r.;,lica.

tion, under various combications of ulttariolet and temperature stresses. It has also been democatrated trat a coabined enviroe-ments test involving both ta=;eratun and ultravialet results in i I

en added reduction in life that vov1d not be in effect if the I

• estparature and ultraviolet were applied separstaly or sequentially.

I?"f!"'UCTIONS FOR CSE

1. Step-By-Step chysical Test P ocedure {

4,- :

.~, The specific steps in the o-ri=g test procedun are as foucws: .h t

e 2

a 155 0

~

l i

-arJesMussess a

--~

m-

.g.. .. , -

r. .. . .

y ,

f. *

, , y . e-1 r l

. t l

i I

i Msually inspect 0-ring for manufacturing imperfections and care- , j fully install in test block with identification dots ca the exposed surface.

' Mount 0-ring and test block on the pressure test fi:sture and raise

• nydraulic pressure to 1500 poi for approx 1:mately 1 a.inute. It  !

0-ring passes initial pressure test, remove block and ring frtan '

test fixtum, wash in paa of 74h and confully blot with a elsen cloth until dry.

~

l Rece21L envirarsmental exposure ir. formation on data sheet. Place

. 0-r.'ng and test block in oven ami ex,ase to the proper level of ultreviolet light. '

Done - Place in shield ares

• 0.1Vatt/?t 2- set block in position 14" from bulb 2

H3.2 Watt /Ft - set bulb at 9 9" above 0-ring Reduce the vertical height of tne bulb by 0.1" each 7 day perirri te compensate for bulb aging. Replace the bulb at 10 week intenals With a new taalb t'at has been " burned in' for 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br />.

0-rings exposed to the +275'3F envimament shal.1 be pressure tested at 24 hcur ===* = intervals until failu.'t. Pressun testing shall .

cc=sist of following the initial test pr> ettre except it shall be pressurized to 1500 psi for three cycles for esch test.

0-rings exposed to the e250*F envirocaent stall be pressure tected

• I+8 hour anx1=um intervals in the same manner as tte 45*F 0-rings.

0-rings exposed to the *2000 F environment saall be pressun tested '

at ur4== 4 day intez mis in the aans macner as the 275*F 0-rin s.

Vben failure is evid. nt, rm the 0-ring from the teet block, wash in ganoline, ard place in an envelo:= having the identifica-tion infomatics on the exterior.

• e Deee were the values used in the Reference 276 stady; other -

values the user. of temperature and ultraviolet exposum may be chosen by ~ , , '

. t l 2. Mathematien1 Precedure 1 e

, s. Perfir:s the accelerated tests for 3 levels of te=peratum, all

• g: vater than norma? (toom), and 3 levels of ultruvialet, includ- -

its O and 2 higher, superposed on temperature. *

b. Record hours to failure for 0-rings in a matrix in accordance I t

' * 'rith 'hble k.6.2; that is, with levels of ultraviolet far columns and levels of temparsture for rows. i 156 I.

f j o i

i g, , e e emme e e m e.m -

_. " ~ ^

-,-m-, ,, w-- -,,-,,,y,,-,\y --,--,,,-,y , - - , , , , - - - , - --y gny-g owwy -,-4 --w -

wm-y-mo m-m,-- e-,,rgm- -m- y rn-,,n es.ymm yms--

l .';

7 ,

~

OEwo.-- __ _.. .

p 4

P i<

nr.

- . c. Find the arit.tssetic mean (X) o.' each cell. p l 3

d. Perfers icgarithmic trar.sfomation of data to detemine effects M of temperature ami ultravtolet on life of 0-rirup by detax :in- 3

) '

. ing by the cr=panents of variates analysis as shcvn in Table j 4.d.3 If the F ratio ne:bers are larger than F,oS in .11 three cases, then the effects of temperature and ultraviolet }

. are s$cnificant for the levelt chosen to test. g i

g

e. Perfom the t tests to obset-te hcw the inte11teticas are :l;rtupert. {

in orcer to select those crx:rir.atit:n6 of st.ecses enuolr4 i

<11f ferences to oc:.:.r 12 :ne tran Liver, as sncvn belev in steps 1 1, 2, ar.1 3 l

1. Tabulate the lo,;nrithmic neans of each of the 9 0-ring ]

test runs as shown in 'Dthle k.8.4. {

.. 9

2. Calculate Xt-X2, the interval between the lo6 arith =le s.eans 8 required for significant differences to exist, based on t.0$ vith 81 de;;rees of freedca a:d the residu.al mesa j square fr:r.4 the analysis of variance in Table 4.0 3 (

I 3 Form tr.a logarithmic wans into 6 groups as describad by  :

the temperature and ultraviolet expos,ur_e levels, and anal- g yte the resulta vith respect to the Ty-X 2 interval faund i in 2 and the effects of the te=perature azzi citraviolet }

expcsuzi.

f. Analyze the failure nrdes (in this case there vere o'ily two: .

radial ami circumferential crack =) to insure that exposurt to . }

sczne levels of te=perature and/or ultraviolet do not e es*e n. I.

third failure mode. Data for a third tmde of failure :s:st be  !

aiscarded ani the ALT teminnted Just below that level of env12tacurnt (accelerated stress). f[

t

g. ferform the analysis relaW to life cycle distribution func- '.

tioris by cerroming the follavina stees t )

I

! 1. Plot the 0-ring failure data for each cell of the matrix

)

of Table 4.8.2 on 'deibull protability paper. For sceursey, ';

the limes of best fit are calculated on the ccuputer by u the method of least squares. .

-l 1

2. The Veitull shape (S) and scale (a) sammeters are calcu. 1

, lated and sucarized in accordance with Table 4.6 5

• ' Mattecatical M;dels 3 .

{V

'a Reference 213 proposes three anthematical codels (Models 2, 4, 5) N as p;ssible dsscriptors o' the relationship between accelersted ng it lW jl a

, LI

q e

--w.~,.

o, e __

O

= = = ' _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _

.__. . _ _ - . . , q

., ta ,L- .. . .- .

. s. .

24 .; ,,, . ,;t_ *.- - '

.a

( l i

l

. I i

l i

I and aerani conditions. A sathemattaa, andal, wttich yields repent. *

- able results, is a prime requiremmat far a use.N1 accelerated test, f, 1 - .no so.as tem m rio. for me., Veimu 3mrumeters observed at accelerated conditions to the Weibull parsa. I stars (to be) cheerved at normal conditions. Fipre 4.8 3 presents

, the models. ,

afort stely, in the e-ring -, av.iahis etnet ma da i I

not permit tests at normal conditions (room temperature); therefore, vtLich of the mala is applicable to the O-rings tested is not  !

presently det m ble. ,

j I

Ihe andals vcmild be wed to predict aj and ${ to see if the predic.

tions fell within the confidence limits calculated from already ch results at the nom conditions. 1 t'

i The user of the ALT woesi be equipped with an estimate of an old accalarated ami old mornal set of WeiW11 shape ami scale paren- t eters as datined in tts table. Es would obtain thaJe estimates .

frus an initial nor=al and accelerated teet run ande on his own  !

test eTJ13 cant and t.he parts of interest. Any time an estiante I vas rev. ired Jf the life expectsacy of the parts when operated at I

=mnufac*urers' rated conditions, the user would simply test a ,

i sa/0le of currtet proauction 3r.:ts, at accelerated stress conditions '

andobtaintheesttratesBI,andp* estima could be ese- j l binedwiththepreviousestimatesk', ,& and in the algo- , y rithm to solyw for the new estiastes and . Thess could bc .

used to construct the cunnalative failnie dist bution and 'ided for i estiantes at any desired probability of survival or missica time.  !

1 I

T'ur nsults obtained veuld represent an estimate of the life g ex 4ctancy of the pa?ts from the current pruhaction run or .1st. ,

l The estanate could be octManut by perferuing a relatituly shcrt  :

'Ast, thereby sering test time end test expences. I

-l It is assumed that the L amlk generated in the refersuced study j represent actinatas for parts produced only by those respective g annufacturers. The mathematical =odal should he translatable to e  ;

usefulness for similar parts prodi.:e1 by other sunnufacturers and i to parts in the same generic family. H ' wever, -iroof of the andel's {. l range of applicability are not contained in the study. l

. . I

. f D

l ~

l e '

158  :

g..

~ ,e , __

_. .j e

- , - - - - - - - - - - , - ,wn-, , , . ,w .--------n,,--,n--,-ca ,g--,_.,._,-,__--nn-,

. ~ . _ , , , , - ,-m,.-,m_,--

. 5. -:'r.I, 'h

. . w * -

r.------- -

ta If.;!i

,g.

. FIGURE 4.8 3. Algoritten for Calculating Life Expectar..y at Maest- e(.

r facturer'sRatedStressconditionsaj,Sg i, i ;.

maa a; .ar F; / F, #; 3,Er g i-ra t.p i

\*R)  ;

I Tr ${-1 3g-1  !

F; '

nodel 4 ag - g dj ,. (q -1) + 1  :

)[ ,

~

Bx-1 (#-lj 4 i i

(~i sa) i r

bT) A 23d EI M.g

• a5 a; g g,a 3, 3; - R + Gy - si)  ;

where: i a1 **

= estiante of Weibull scala gnrumeter for parts if they tad I been tested at manufacturer's ratad operatir.g and ecviron- t

.s.ta stmas.

l d*g= estiante of Weihil shape pcemeter for farts if they had been cested at aant.fachtrer's rated opersting and environ-mental stresses '

a.

i

'e* = esti.:sta of Weibull scale parameter obtained from a current test run at acenierated stresses .__

g""j = estiante of Weihll . shape parameter obtained fztia a cur.vnt . .l test run at accelerated stresses  :

~

g = estimate of Weibull scale parameters from a prwrious test run of parts cperated at arm.tfacturer's tuted cenditicas estiante of Weibull shape parsanter frtza a ptweisus test l k = ran of switches operated at manufacturer's .sted conditiuns

• l g-= estiaste of Weibuu scale parnaster free a previous test , l rst cf switches operated at accelerated stress couditions l , i y, = estiante of Weiwu saa,e -ter fra. . pronous te.t  :

.m of switches operated at accelerated stzwss conditions.  !

159 '

. .l '

. t' w =

t i

i l

. - _ _ - _ _ _ . - - au l . . . . . ..

l l

- . . . - - - . - - - - - . _ - - - _ -._.___n_..-_., -._-- - ---. , - -._, _ __.- - _.- _.,_.,-,,__ - - ,,,-..-- --...- ,.._-

a., *-

~

1^

. ts:tecas/nArt or .nm.tfArros ,

1. The ALT physical test methods aci analyses techniguos are gaaerally and.icable ts a wide range of sta11ar parts and different generic {

4 fa=111es. *

. 2. The ALT is unique in that it fartudra two different types of i 1

, accalarating stress, one superposed on the other - in this case 1 ultraviolet light w posu n added to

• w titre. A ulysis tech-nigaes are presented for determining the effects of each type of 2 stress, slagly and combined, on the 0-ring life. These technicas, -

or twlated ones, have a wide run6e of important applications is '

f today's products, wt.ich are subter.ted to many different types at 4

• stiti"-onvironments in aeroepe systems ami others. Ia f

3. The ALT is specifically limited in the particular instance of an I

, 0-ring of Rans-5 rubber asterial, in that a material changa to I one of silicone cr other could have profoutd effects on the failure  !

andes and causes of failurt (en-ironment), with reculting possible chioges of test equipment trjes, kirds r.f accelerating stresses, f test procedures, and pe-bape r en the m'alysis technipes ar,4 '

mathauntical spiels.

{

RD*Dn:I3: 298. .

a~

. l.

i.

1-

-1

.: L

[

' I l-I -

.I I

i i- *

- }.

. :i

. . s

. 9 i n

. 1e s

i I.

- 1

. 16o  !

i

/.

, ,s .

i

= '

_ _ _ _ _ _. _ . _ - _ - ^^

- - * - ~

e m - . o

.-,y -- --- -,- w.- --,m - - - . g

=

,s .. . ,

j ; . I ,e .

.. y, . . I \-)

. . . Is m wem j i

!i i i.

l.j.i 4

1

,,0,,.r nweenians reasses

/j smeisin , ',lI, Stet 50e448 t ..

. E e 1

. L 1

V id 3

i

.I t i

'.*annin e.smenon 1r i

es e.,h h

8 l-Figure 4.3.1. O-Ring Test Method j U,

o q.

l i ,

( :

i  !

i 1

1 I n  :-

i ..

t Ei 3 -

1 ,

"A 3 8

a,,. ,

w; E! t.

r1:

1Q. r.1

• e P '$ ,

! .'.5 J ,

~~

- _ _ . . = _ , j; .

8

. . k

_ I

_ _ _ _ __,__, - _,_ -- - - 2 -- ~- * - - A * ~ ^ ^ ^ ' ^

~ *

.' .l. . * ' , *;. ~ * -a. ~.

g- , ' , ,* . .- J -"4 '. '.f. . . ~ 1. . T , *

. 2

. - . .,..,,p .:-

... .. - . . .2 .,

. , p. . .

7 s

58089 4

.m I .

east I e ~

W

- PM tt ee t Tatti

_ c

• t' StM hh h?5lO//Aassf. a.a-j i

retssess s intf artet i l

sie nsities: -

^

o I '

.Q. Q) -

o O O TTPitat Tilf plots o

O i

f I l

~

Figure 4.8.2. O-Ring Test Fixture 4 l

. j i

i 1

. 1 f

.. 162 u

. .. g .

W e Consumm

\ E.wcwcs W NUCLEAR OPERATIONS DEPARTMENT EA 80 %-01 Engineering Analysis Work Sheet Sheet i of f Th /. i Title MD - 90 0 - La F6_61890N CAtc4.;cA71o ch Performed by '! s SM / i1C Date 9 //786.__

References An 6 hm7_e_pgs RevhMethod by: b Alternate Calculations

( h 6:47Y %g@% i ko  % Detailed Review Qualification Test 0 OcEl f 19!&#

Technical Review by Date u G 90$b P(ovice: 44ntP.;S T o Os.s ou sTe.-Te. s o. 7osa w =.s net ovE9ef ST2.L-%d:D 006 To s u-562vic& TI M E Ano ( o.c. A.

TEfbTs u(n . Tna Aunt 9tt5 citu tuow AOEou td6. M AE(nin Fo2. V w v6 cPEllnTo(7 us =46 (. h.5&O on T u6f.A m A L 4.ousat/i(mch Foe _ A 96.EioO o F- Tim & 66kuO Tud c.e.2 0 fAu 7 :wn u_6:0 0 m:a6 .

{

• N(.7aco_

k92 A64eu b m 6,%Q w as uSE_o 6,45 e:o o4 036V t> ut E cF bo uss 84 m mT(d?.> A L. ujsu, nc.7i s ceri o% MW O cG O . 2k.p . Tad. ATTaruho 9t oT AMO CALC.or ,cTeca C.o ossOER 5 Ius7N.t60 Lie 6, Lo,c_gg T6Ti%j f(.oros to L . o .C. . A. . j" $66Em 6TS O i vidiu(n 7uh P E.oCit.6 Ano f NTGf.99.ctT60 v5:ut.n AQ.G.N 64ios c.ntcucAfious .

f g Dis 6vuiem

+

1"N::. 765$ L.o.c. A P%gd is rue rie_sT excue.siou of

, g AC,. i j t .o.c A 15 7u6 Sec.mo &c v essou umn 7e j o2 Aus7m cthe L.i c.6 Fott oer ut.o . Tu6 Pt.oT f iuTMEc3 EO N

25' de t:o em Q,ELHL-ui u s M&7hoO esich c.ousi o L-4.5 )

ri-& L.o.c.c. Te5T , nd c tt ec Ti mE A-u o A El DCd L.O.t . A tws Au 6 0 0tv a c.h.bT balm %

Lite iW bas o p. \\ 16AL5. T46 n %AeS iS Aucu mve: m ne.6na Esono A %90se.o Atcio ewi l l

dob71 C '/ / M h G ca37 w ut::0 v 5 s4 (::r

• l Form 3119 343

e e s e

, . CONSUMERS POWER COMPANY BIG ROCK POINT Manufacturer: LIMITORQUE Equipment

Description:

MOTOR OPERATED VALVE Component Evaluated: WORST CASE MATERIAL

Reference:

Activation Energy: 0.860 eV

Reference:

Reference Temperature: 99.000 (*F)

STEP 1. THE LIMITORQUE AGING TEST TEMPERATURE DATA POINTS ARE CONVERTED TO DEGREES K AND TO SECONDS:

F = K Day Hrs Min Sec = Secs 275.0 408.150 0 0 0 0 0.000 275.0 408.150 0 148.7 O O 535320.000 STEP 2. THE LIMITORQUE AGING TEST TEMPERATURE DATA POINTS, AS PLOTTED ON FIGURE 1, ARE PARTITIONED INTO 1 SEGMENT AND THE THERMAL EQUIVALENCY OF EACH SEGMENT IS CALCULATED

WITH RESPECT TO THE REFERENCE TEMPERATURE:

Segment ist Temp 2nd Temp Duration Equivalency (Secs) 1 275.000 275.000 535320.000 1186144902.029 Tg-f L,(fP I .I i f.- G D~ * ?.],- OT r/,r T e ;: d-Crg' '

m e = e

. CONSUMERS POWER COMPANY BIG ROCK POINT STEP 3. THE SPECIFIED BRP COMPOSITE TEMPERATURE PROFILE DATA POINTS ARE CONVERTED TO DEGREES K AND TO SECONDS:

F = K Day Mrs Min See = Secs 241.0 389.261 O O O O O.000 241.0 389.261 0 1. 6 4. 2 0 6012.000 216.0 375.372 0 1. 6 4. 2 0 6012.000 216.0 37S.372 0 4. 6 4. 2 0 16812.000 185.0 358.150 0 4. 6 4. 2 0 16812.000 185.0 358.150 0 14.0 15.0 0 51300.000 145.0 335.928 O 14.0 15.0 0 51300.000 145.0 335.928 0 23.0 55.2 0 86112.000 126.0 325.372 0 23.0 55.2 O 86112.000 126.0 325.372 0 36.0 0 0 129600.000 235.0 385.928 0 36.0 0 0 129600.000 235.0 385.928 0 60.0 0 0 216000.000 115.0 319.261 0 60.0 0 0 216000.000 115.0 319.261 30.0 36.0 0 0 2721600.000 99.0 310.372 30.0 36.0 O O 2721600.000 99.0 310.372 4778.0 42.0 0 0 412970400.000 99.0 310.372 4778.0 42.0 0 0 412970400.000 99.0 310.372 8796.0 36.0 0 0 760104000.000 STEP 4. THE SPECIFIED BRP COMPOSITE TEMPERATURE PROFILE, ALSO PLOTTED ON FIGURE 1, IS PARTITIONED INTO 9 SEGMENT (S)

AND THE THERMAL EQUIVALENCY OF EACH SEGMENT IS CALCULATED WITH RESPECT TO THE REFERENCE TEMPERATURE:

Segment let Temp 2nd Temp Duration Equivalency (Secs) 1 241.000 241.000 6012.000 4066575.765 2 216.000 216.000 10800.000 2829042.853 3 185.000 185.000 34488.000 2515553.274 4 145 000 145.000 34812.000 401899.220 5 126.000 126.000 43488.000 191502.667 6 235.000 235.000 86400.000 46832731.471 7 115.000 115.000 2505600.000 6133788.151 8 99.000 99.000 410248800.000 410248800.000 9 99.000 99.000 347133600.000 347133600.000

$2Ed -

1 SEA

• 4 e = e

. CONSUMERS POWER COMPANY BIG ROCK POINT STEP 5. THE THERMAL EQUIVALENCY OF ALL OF THE SEGMENTS IN EACH PROFILE ARE SUMMED AND THE DIFFERENCE IS CALCULATED:

Thermal Equivalency (Secs)

AGING TEST 1186144902.029 SPECIFIED PROFILE 820353493.401 DIFFERENCE 365791408.628 THE DIFFERENCE IS A POSITIVE NUMBER, THEREFORE THE SPECIFIED BRP COMPOSITE TEMPERATURE PROFILE IS ENVELOPED BY THE LIMITORQUE AGING TEST.

l STEP 6. THE DIFFERENCE CALCULATED IN STEP 5 ABOVE IS ACTUALLY THE MARGIN BEYOND WHICH THE SPECIFIED BRP COMPOSITE PROFILE IS ENVELOPED BY THE LIMITORQUE AGING TEST. CONVERTING SECONDS TO YEARS, THE RESULTANT MARGIN IS:

11.591 Years at 99.00 F l

s s&EU 3

, 2e/. '

l L

o, l 7

CONSUMERS POWER COMPANY -

BIG ROCK POINT AGING TEST VS SPECIFIED COMPOSITE PROFILE PLOT 300 270

(( 240 - - - - - - - - -

au e l

~* .

tt; 210 ,

E D * -9 .

H 180 l i

"1 i L' 111 150

.I o-o '

ft X C U l? 120 ,

u -

-g 0 - --

+o c -A REQUIhED 90 Tl g 60

- 1EO 1Ei 1E2 1E3 1E4 1E5 1E6 1E7 1E8 1E9 SECONDS

{,p

- ,3 L_________ _ _ _ _ _ _ _ - _ _