ML20077S358
| ML20077S358 | |
| Person / Time | |
|---|---|
| Site: | 05000000, Crane |
| Issue date: | 12/14/1979 |
| From: | GENERAL PUBLIC UTILITIES CORP. |
| To: | |
| Shared Package | |
| ML20077S281 | List: |
| References | |
| FOIA-83-349 110, NUDOCS 8309220037 | |
| Download: ML20077S358 (10) | |
Text
-
e y,4
=
m
~ ~ + = ]-... m : m, g :,,yrL + u,s, I-p j[
gg TDR MO.
110 REVISION NO.
O
~ ~-
~..
...- ~.
~
.m TECHNICAL DATA REPORT ~
PROJECT NO.
PAGE 1
OF 7 PROJECT:
Three Mile Island Unit 1 DEPARTMENT /SECTION Eng. & Design / Mech. Camp Decay Heat Valve DH-V1 RELEASE DATE REVISION DATE DOCUMENT TITLE:
Continued Operation of TMI-l Decc.y Heat Valve DH-V1.
ORIGINATOR SIGNATURE.
DATE APPROVAL (Sp SIGNATURE.
DATE
- 19. %.
rodn
.LP/k n. G-1-n-pr 0
&' /}7ynjhw,l &l]._.
/f]M/H \\
{
g
/y APPROVAL. FOR EXTERNAL DISTRIBUTION DATE
~
\\
o.
DISTRIBUTION ABSTRACT:
R.C.
Arnold TR T1-0030:
' Review status of DH-V1 a$1d determine J.G.
Herbein whether or not permanent repair is required at D.K.
Croneberger this time.
F.S. Giacobbe J.P. Moore, Jr.
After reassembly of DH-V1 following stem replace- '
mentfin 1976, the valve developed a leak.
Although D.G.
Slear the' leakage rate was acceptable as an identified 7
D.M. Smith reactor coolant leakagg_per the TMI-l Technical R.F.
Wilson Specifications, it was decided to repa'ir the valve.
J. J.
Colitz.',
The rdpair method chosen was to furmanite the valve.
G.
P.' Miller This was done in November 1976.
M. J.
Ross D.
M.
Shovlin Furmanite, which is a nitrile rubber base compound, G. J.
Troffer contains leachable chlorides.
This raised the question of possible stress corrosion cracking of the 316 stainless steel body and bonnet of the valve.
Both the GPUSC System Laboratory and MPR Associates reviewed this issue in 1977.and con-cluded that stress corrosion cracking is unlikely to occur..The System Laboratory has also re-reviewed the issue at each subsequent refueling outage and has concluded that another year of operation would be acceptable.
Due to'the extended outage of TMI-1, the question of whether the reactor should be defueled to repair DH-V1 or whether to continue operation with DH-V1, as is,- furmanited, has to be settled.
Both options have risks associated with them.
The risks have..
been analyzed to determine which option is the most.
advantageous to the timely restart and continued operation of TMI-1.
,o y ;. y.
b b 2-8309220037 830728
,[
'...-" c T-T;'l,',1,'$' '.1 PETCHE83-349 PDR 2
r
<a
-r-
o
~.v. 4 y + d w W [ W g g n ej.,;
4.,2.,r.. a p.,.. _,..,, p s g.
,s,,,. _.,,.,,,.,;.,,,,
. g,.,.,
Mechanical. components has concluded leaving DH-V1 as is to This conclusion was reached after~
bo the most advantageous option.
discussions with the System Laboratory and MPR Associates showed that there is a small likelihood of stress corrosion cracking in DH-V1.
M::chanical' Components recommends the following:
TMI-l be operated with'DH-V1 as'is, furmanited, until the I
1.
1985 ten year inservice inspection outage with an in. spec-tion of the valve to be performed during each refueling outage as a minimum.
The Materials Technob.ogy' Group (No.ndestructive Examination 2.
and Materials Engineering Sections) be, tasked to develop an appropriate NDE technique to monitor DH-V1 to detect I
stress corrosion cracking if it should occur.
3.
DH-V1 be replaced during~'the 1985 outage.
~
~
If the reactor is defueled for other problems prior to 1985
~
4.
then and after a replacement valve is available at TMI, DH-V1 should be replaced during such an outage.
If tha reactor is defueled for other problems prior to a 5.
replacement valve fieing available at TMI, then DH-V1
. should be repaired.
O O
O 9
e D
9 O
e 6
6 e
e g
e O
E g
e
- O m.mm__n_,__
~
1 TABLE OF CONTENTS
.4*v,
em. N.,.-%
3.%e; @,*f.25N M.;padase.
,y,.
.g......,,. y,s..,g. g,,3 [ p s'.'
~
..,.,....,i... ; t,;; p r,,
.y.,,,.
..~c 1.
Purpbse and Summary 2
2 2.
Evaluation 3.
Conclusions 4
4.
Recommendations
~
6-5.
Attachments 6
j
__ o
_a
g v
.~
1.
Purpose and Summary.sp.
.%r M.C.--.44 mM,w.~ '
.... ~. ? t, & a '. -
.~."
The. purpose of this Technical Data Report is to respond to TR-T1-0030 requesting a review of the status of DH-V1 on TMI-l i
and a determination whether or not permanent repair of the valve is required at this time.
This report will discuss the history of DH-V1 and the concerns about the valve's acceptability for continued operation with the seal ring sealed with furmanite.
2.
Evaluation DH-V1 is a 12" 1500 lb gate valve with a stellited flexible wedge and a cast 316 stainless steel body manufactured by the Walworth Company, Model 5262PS, with a design pressure of 2500 p'si and a design temperature of 6500F.
DH-V1 is the first isolation valve
[
in the single decay heat drop line which is connected to the hot" leg of the reactor coolant B loop.
It is closed during all power operations since the decay heat removal system is not then in operation.
It is only opened during operation of the decay heat removal system.
The valve was disassembled during the 1976 TMI-1 refueling outage to replace a bent stem.
Also the nickel pressure seal ring gasket which forms the pressure seal between the body and bonnet was replaced.
After ap roximately one month of operation leakage was E
observed past the seal gasket.
Although this leak was acceptable as an identified reactor coolant leakage per the TMI 1 technical specifications it was decided to stop this leakage.
There were three alternatives discussed to stop the leak.
They were:
- 1. Defuel the reactor and repair the valve; 2. Weld on a specially designed redundant seal between the valve body and bonnet; and 3. inject a leak sealing compound into the thread relief area of the valve above the seal gasket.
The third altern-ative was the one selected to stop the leak.
In November 1976, nuclear grade furnan.te was inj'ected into the valve through four drilled' holes space.
~ apart.
The placement of these holes was such that they did not de
'te the valve pressure retaining boundary.
A stress and fatAgm
-alysis based upon the size and location of these holes was perfore..
'MPR Associates in October 1976 to ensure their acceptability..
structural and design code viewpoint.
This analysis was revi.
by GPUSC Engineering Mechanics and Engineering Mechanics' concurs.
with the MPR conclusion that~the holes are acceptable from a stress and fatigue viewpoint.
The valve has been in operati,on since it was furmanited having experienced several heatup and cooldown cycles and periodic visual observation has shown it to be leak tight.
~
0 e
e d.%'*
-. ~
.i5M. &h,:,.* a ~:.=,. :o : : ~ d : wsu.WM;-iss.'a...
- r..v w q. m. a. '.
r
~
Even 'though 'ho valve has operated without Icaking since it was furmanituu there are two arcas of concern about the valve being able to continue to operate with the furmanite until' the (prob' bly 1985 for the ten year in-service reactor is defueled a
inspection outage) so it can be repaired or replaced.
The first concern is will the furmanite continue ~to maintain the valve in a leak tight condition.
The second concern is about the possibility of stress corrosion cracking occurring due to
~
the presence of the furmanite.
~
The first concern has to do with the ability of the furmanite to withstand the radiation environment and the thermal cycling without degrading and allowing a new leak to occur.
Met-Ed e
discussed the degradati'on of the furmanite in March 1977 (Attachment B) and concluded that the furmanite is acceptable for the radiation environment and thermal cycling.
This con-i clusion was based on 1. Furmanite samples that were exposed to 105 rads and 106 rads and showed no apparent physical degra-dation or embrittlement; and 2. Discussions with Furmanite, Inc.
on.the effects of thermal cy' cling'at the valve temperatures on furmanite.
In a recent discussion, Furmanite, Inc. (Bill Rinz) stated that there should be no problem with the valve lasting until 1985 and if'the furmanited area should develop a leak it can be re-
)
injected with Furmanite.
Furmanite, Inc. has developed two new nuclear grade compounds since 1976 and both have better flow characteristics for injection than the Furmanite used in DH-V1, either of these two compounds should stop any leak that
+
could develop in the present furmanite.
Also Furmanite, Inc.
has knowledge of furmanited valves that have been in service for twelve years.
Unfortunately there is no documentation for this claim.
1 i
i The second" concern has to do with the possibility of stress corrosion cracking of the cast 316 stainless steel body due to the leachable chlorides in the nitrilarubber base of the furm-anite compound.
This concern has been reviewed by both MPR i
Associates and the GPUSC System Laboratory.
MPR Associates reviewed the stress corrosion cracking problem in April 1977 (Attachment C) ~and concluded.that the reported levels of-l chlorides are not cause for concern and that the furmanite~
can be.left. in place in the valve for an extended period.
-This' conclusion was based on 1. the service temperature being low, r
O 1AD F; 2. the duplex austenitic-ferritic solution annealed cast
+
~
316 stainless steel body and bonnet which is quite resistant -
to stress corrosion c, racking; and 3. the low chloride levels which are below those of elastomers which have had extensive successful use in similar applications.
In a recent discussioni MPR Associates (Jeff Gorman) stated that there is no new stress corrosion cracking information which would cause MPR Associates l
to change its assessment of risk in using furmanite.
~
.._._.--._..-.;;___,._,-..._._,.-....-..........,....~.,...... a.m
- m..-,..
c
_=m.
,...,,m,,..
y,
. rw :-e.L~..% 4/% %Wg ;g;gclp. o.: m. a..,..,;p...,
The GPUSC System Laboratory reviewed the problem in November 1976 (Attachment D) and several times since then (Attachments E,F,G,H and I) and concluded that there is a low probability for stress corrosion cracking..This conclusion was based on the same reasons as MPR Associates plus 1. The absence of oxygen which would inhibit the occurrence of stress corrosion cracking; 2. No leakage which means there is no evaporative mechanism'to concentrate any leachable chlorides; and 3. The stress levels which are below the levels required to have stress corrosion cracking in 304 stainless steel which is less resistant than 316 stainless steel.
The System Laboratory also recommended that the valve be periodically inspected for any signs of leakage which could change its assessment.
In a recent
, discussion with the System Laboratory (Scott Giacobbe) stated.
that there is a small likelihood of stress corrosion cracking in DH-V1 but it cannot be written off 100%, so if the valve
.is to be operated until 1985 an attempt should be made to use or develop an NDE Technique such as UT or acoustic emission to detect stress corrosion cracking.
The, alternative to leaving the valve in its present condition is to defuel TMI-l so that DH-V1 can be repaired.
The amount of repair required to restore DH-V3 depends on the amount of damage to the seal, ring seating area.
This cannot be determined until DH-V1 is dispssembled.
The amount of repair could vary
. from spot weld repair and minor machining to major weld buildup
~and machining and the use of an oversize seal ring.
Two other possibilities that also should be noted are that the valve could be damaged on reassembly as happened in 1976 and a leak develop or that the valve cannot be repaired and would have to be replaced.
If the valve did develop a new leak it could be furmanited and this would put the valve in the same condition that it is in now.
If the valve had to be replaced, a replace-ment would not be available until March 1980 at the earliest causing an unacceptable delay in returning TMI-l to operating condition.
3.
Conclusions Both alternatives, opdrate TMI-l with.DH-V1 as ' is, furmanited, or defuel TMI-1. to repair DH-V1 both involve risks.
The risks of each alternative have to be analyzed to determine which alternative is most advantageous to the timely restart and. con-tinued operation of TMI-1.
' Alternative 1, operate with DH-V1 as is, does not impact the timely restart of TMI-l but could impact the continued operation of the unit.
The two factorswhich could impact continued Q
w l
... m
=
operation are tho'valvo developing'a leak through.the furmanite M <<. -
and stress corrosion cracking causing a leak through the wall g.,'
of the valve body.
The first factor, leaking through the furminite, is minor as TMI-l can be. operated with an identified reactor coolant leakage rate that is acceptable per the unit's technical specifications, as was done in 1976, until DH-V1.can be reinjected with furmanite.
The second factor, stress corrosion cracking causing a through-wall leak, is major as the unit would have to be defueled to repair or replace DH-V1.
.The -risk involved'with the first factor is minimal as the furmanite should keep the. valve-leak tight until 1985.
- Also, if a leak should develop it would be no larger than the original leak and furmanite could be reinjected to stop the leak.
The' risk involved with the second factor is also minimal as there is a low probability that stress corrosion cracking would.
occur in DH-V1.
This conclusion was made by the GPUSC System Laboratory.and MPR Associates.
Alternative 2, repair DH-V1,~ could impact the timely re' start of TMI-1 and could also impact the continued operation of the unit.
The factors which could impact timely restart are 1.
problems encountered during the defueling and/or refueling of the reactor; 2. delays and/or problems encountered during in-line repair of DH-Vl; and 3. DH-V1 cannot be repaired in-line and will have to cut out. of ~the system for repair or replacement.
The r
- factor which could impact continued operation is that the valve could again develop a leak after reassembly as happened in 1976.
. The risk involved with the first factor is small as the plant personnel have had experience in fuel handling, but the reli-ability of the fuel handling and transfer equipment has been questioned by the plant staff.
Equipment breakdowns and bowed fuel assemblies could add a few days to a few weeks to the defueling/ refueling cycle.
The risk involved with the second factor is unknown because the
- amount and complexity-of the repair cannot be determined until the valve is disassembled.
Also, as with any other in-line repair, as the complexity of the repair increases so do the chances of welding or machining errors which could delay the repairs or render the valve unrepairable.
The risk involved with the third factor is large because a re-placement valve will not be available until March 1980 at the,
g earliest.
'he risk involved with the factor that could impact continued T
operation is minimal.
If a leak.does develop due to errors in repair or reassembly, the valve could be injected with.
furm.anite to stop the leak. ~
^'
{.
. / **
~*
\\
p t-Q.
ls f *}.
t-*
l
~
.Q
,. i;......v,.
w4p#wi.ab,., Aw.mp.a.'.
. g-t.. w.....# 2.
..,c;. 5 Obviously there are risks associated with both' alternatives, operate with DH-V1 as is or repair DH-V1.
Also the risks invo'lved with each alternate contain unknowns, such as the phenomenon of stress corrosion cracking is an inexact science and the amount of repair required is unknown, which complicate the final analysis.
These risks have to be evaluated and i
good engineering judgement has to be applied to determine wh'ich alternative'is the more technically acceptable for the continued safe operation of TMI-1.
Based on the above discussion of both alternatives, Mechanical Components concludes that alternative 1, operate with DH-V1 as is, is the more hechnically acceptable alternative.
There is only a small probability that stress corrosion cracking will occur in DH-V1.
Also, it is better to operate TMI-l with a component that is technically acceptabl_e than to attempt to repair the component and get involved with the problems of de-fueling / refueling, welding, machining and reassembly..
4.
Recommendations Mechanical Components recommends the following:
1.
TMI-l be operated with DH-V1 as is, furmanited, until the 1985 ten year.dinservice inspection outage with an inspec-tion of the valve to'be performed during each refueling outage as a minimum.
2.
The Materials Technology Group (N ndestructive Examination o
and Materials Engineering Sections) be tasked'to develop an. appropriate NDE technique to monitor DH-V1 to detect stress corrosion cracking if it should occur.
~~
3.
DH-V1 be replaced during the 1985 outage.
4.
If the reactor is defueled for other problems prior to 1985~
and^after a replacement valve is available at TMI, then DH-V1 should be replaced during such an outage.
5.
If the reactor.is defueled for other problems prior to a replacement valve being available at TMI, then DH-V1 should be repaired.
~
5.,
Attachments A.
Cross-Section o,f furmanited area of DH-V1.
B.
Memo, Generation Engineering to J.J. Colitz, " Continued operation of DH-V1 Utilizing Furmanite", GEM 0715 dated l
March 11, 1977.
O b
p ll
.=.:.
-l-
=--
, = -.
- p -:..:;;..~,d. a 4 4%&s<15.%.; WXQ*;:. f. >~.6. y.:..:
,a;;a.a4s
.c....:. a ;.......
C.
Letter, MPR As.sociates to S.H. Bailey, "Use of Furmanite in DH-V1", dated April' 15, 1977.
D.
Memo, GPUSC to S.H. Bailey, "Use o'f Furmanite in the TMI-l Primary System", dated November 4, 1976.
E.
Memo, GPUSC to S.H. Bailey, " Considerations Relating to the Extended Use of Furmanite in DH-V1 at TMI-1",.
dated March 1, 1977.
F.
Memo, GPUSC to S.H. Bailey, " Continued Use of Furmanite
.in DH-V1 at TMI-1", dated February 6, 1978 G.
Memo, F.S. Giacobbe to E.G. Skuchas, " Continued Operation of TMI-1, DH-Vl", dated December 21; 1978.
.... 2' -
H.
Memo,'F.S. Giacobbe to R.M. Klingaman, " Response to R.M.
Klingaman's Inemo of 12/26/78 regarding continued operation of the TMI-l DH-Vl", dated January.3, 1979.
Giacobbe to 5:.G. Skuchas, "DH-V1 Ultrasonic I.
Memo, F.S.
Testing", dated January 17, 1979
.,t t
a j
l 1
..g-.
-.gn.
r.
pr,w.es.c+.,--
g er,-.e e#e
- r' P.9"*
- s++e+-
.r
-4 4-
- <*,-<n
a.
y
. - r. -
-l P
mm ay
.v
,,,y :
%m.,s..
+, f.s h. *,
.,. 4.. ;.
n as;- s W.c.. n %ei. c.
r
---)'ord*
.s
,..s
.s 6
l l.-
..q o
l.'
. r.
a.
i
...~
.. )
. s
)
\\
s
~.:
..:t t
, e...
f
..g..
- ~y. %. }
BONMT.
u...*-
.. i '.
.2 r /
Q 3
3 e.
3 s
t.
.Ct. AMP.. '
BING '
-wcw sounr amaws.
' ~.
'. 4.
.. g l
y i
r s
.S s
.s s
4 s.
u 1
f l
.,. I,g.
........ s 4
. ~...
. m T. OFT-I
- - l!
. s.
s l
ym ye. :.
l ll l
c i..._..
..-).-..
.u
.,.. ii
...:. v...;
~s. :
a.
1
.?
r ll,
I
,s.
gg e
gl pfgg
y>
l,{
--- - a;1 J
3 1
4'.P_ _ ___.____t
['.,
IL
..l
,s L. _ _. _ _... s f lt.
l
..4.
e_,r. _ -
r -- - --.
gg g,
J
=,
y
- r e
1 s
~
k
' --.SFAl. GASMET.
'. g7py&fy. yy
=
(TYP 4 /2 4CESQe'At-;
.l
.r.
,. ~
- '.y.
~+-.
.fetc uNc Tnos d '.r.p u p
, a. : -
s....
.. q...
,.3, j
c
.... y
- y.....
a
.).,......
a.
s..n.
.s.3..
.i t
,~.
r
- v...
'.n.... c. :,...?..
.%.. s.
-.s :
,,,. p
- i.,...
.j :.,,,,.:,';. -
' '.y:T.,,..
e.
..jpk.c,.... g.. p..:;4 s7tM
)
' ~ - a w, s o o,
.. p w :.
. r.
n r.,
e
=
- ...,. ~.
. ; ~.
- L 7.
?
.j e
I
.-t.
,g.,,
- 5
- ;f ;.:,.,
.g 4.
i..ge;,i f
- n..,.
..j*e. ;,.,
- s s.,.'.
- m sq e.
,s y
,, e,..q;;,...
....,. q.;*. &.,.,...,.a..:,m
...A.;,.
, m ;*
, cos, oy _
..*...g.
=...s
...pE.-.e,,.,,..* ; a t,s
.. -A..3 1
s.,
..,e
......,s....
t.,;.*
s-
..a.
, g..
. r..
..y y2.
- ga,:.
+...
....,. tr**W"4f. g.m.J.'
.~.e, g-
,.,3
, i.,,.... - c...*......i.-..,4.,.> :~.,.f,ae: g;.,c;, ~
g r...
,g 9. <. r;,s,.,
,~y
.. e, 4.. '..s'."Y.,.* Q.:s.'.:,.w.
aq. t. e....p......
?....
]
t.Y. %..
.e
--. ~..
. m.
.v
.m.
.e w
,., :...,
- s.*.',d 4.-
....., ; l...
s.r a........<.v tiry-:.Q*iW gp'.-/;.,.g..t w
.v...,. e ~,
,,z. r -
.. f,.
. l..,a'* f $..*..h.*.t~;j r'.'4.,..
wa
-re.4
,:..r, y,;. y.f....c..afs ;?
.. !,rr..,.,
F.. -
i
.v'.e '..
- s, w.~, n.3.s m... m....'.%.g *;.(.i..* 7.y ;,u E.*.
4
..y..
u.
- 5' 4 s,... _ y. r.w' A.=h..i.l.r'e,:. s :. l l.'.?.,h~'. R;.,'.~s..~.-.....
q, s ;q,:m:z 3 r.;.g',g.%.zit;fM,. w'sn
.M$a.
.Q n o y a,.w
.o..sr..
,.~; e. g.,.. n...
- - a r
.c.).,
. :c r a.p...v.:.,... m.
,.y,.g Q a
- s.
.e
- t g a. --
y 4.w g
.