ML20100H121
| ML20100H121 | |
| Person / Time | |
|---|---|
| Site: | Fort Saint Vrain  |
| Issue date: | 03/21/1985 |
| From: | Rao R, Thurgood B PUBLIC SERVICE CO. OF COLORADO |
| To: | |
| Shared Package | |
| ML20100H118 | List: |
| References | |
| TAC-57248, NUDOCS 8504080588 | |
| Download: ML20100H121 (57) | |
Text
' C*
, s, GA taas Cav. to/es ROLL 2011 lSSUE
SUMMARY
TITLE EVALUATION OF FORT ST. VRAIN METALLIC O R&0 5
j APPROVAL LEVEL COMPONENTS EXPOSED TO PRIMARY COOLANT DV & S DESIGN CIILOEIDE CONTWINATION 1
DISCIPLINE SYSTEM 00C. TYPE PROJECT DOCUMENT NO.
ISSUE N0/LTR.
]
M 01 RGE 1900 907875 N/C QUAUTY ASSURANCE LEVEL SAFETY CLAS$1FICAT10N SEISMIC CATEGORY ELECTRICAL CLASSIFICATION QAL-I FSV I FSV I N/A APPROVAL ISSUE PREPARED DESCREON/
ISSUE DATE FUNDING APPUCA8LE gy NBS NO.
ENGINEEhlNG QA PROJECT PROJECT p
Preliminary Assessment /2970.221
.s. c ;w~
N/C W 8 1 ISIS Proj ect L wanson "it.i
't.Rosenber[
e o M
(,
Assurance c
B.Thur
- d. ramblet.c Sst. W-j f"; gpo
/
l b ts J. 3 ss/6 CONTINUE ON OA FORM 14851 NEXT INDENTURED DOCUMENTS N5053 e*
B504000500 850326 PDR ADOCK 05000267 p
PDR l
REV SH REV SH 29 30 31 32 33 34' 35 38 37 38 39 40 41 42 43 44 45 48 7
43 Y9 50 01 52l53 54 55 58 1
"'V
.. t SH 1
2 3
4 5
6 7
8 9
10 11 12 13 14 15 18 17 18 19 20, 21 l 22 r23 24 25 28 27 28 CARPINELLI-A-130-104 (W/T)
PAGE 1 0F 57 j
.A GA
.TECBDOLOGIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUMNT NO.
907875 ISSUE NO./LTR. N/C CONTENTS
.P,agg 1.
SUlttARY..........................................................
4 2.
PURPOSE AND SCOPE................................................
4 3.-
INTRODUCTION AND BACKGROUND......................................
6 1
4.
EVALUATION OF CHLORIDE-EXPOSED COMPONENTS........................
8 t
4.1 Cavity Liner................................................
9 l.
4.1.1 Metallurgical Evaluation.............................
9 4.1.2 Engineering Evaluation...............................
10 4.2 Plenum Elements.............................................
11 4.2.1 Metallurgical Evaluation.............................
11 4.2.2 Engineering Evaluation...............................
12
{
4.3 Cor e R es trai nt D evi ces......................................
15 431 Metallurgical Evaluation.............................
15 4.3.2 Engineering Evaluation...............................
16 4.4 Helits Purification System and Hydrogen Oetter..............
21 4.4.1 Metallurgical Evaluation.............................
21 4.4.2 Engineering Evaluation...............................
21 4.5 Instrument and S ensor Lines.................................
25 4.5.1 Metallurgical Evaluation.............................
25 l
4.5.2 Engineering Evaluation...............................
26 4.6 Control Rod and Control Rod Drive Mechanisms................
29 i
4.6.1 Metallurgical Evaluation.............................
29 4.6.2 Engineering Evaluation...............................
29 4.7 Helium Circulator...........................................
35 4.7.1 Metallurgical Evaluation.............................
35 4.7.2 Engineering Evaluation...............................
37 4.8 Thermal Barrier Attachments.................................
41 4.8.1 Metallurgical Evaluation.............................
41 i
4.8.2 Engineering Ev.aluation...............................
43 5.
CONCLUSIONS......................................................
52 6.
ACKNOWLEDGMENT...................................................
52 i
s APPENDIX A............................................................
53 i
I l
l I
Page 2 l
1
k GA TECHUOLOGIES I N C.
TITLE:-
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCU M NT NO.
907875 ISSUE NO./LTR. N/C LIST OF FIGURES 4.2-1.
Top plenum and orifice valve arrangement......................
13 4.3-1.
Core restraint device locations...............................
17
=.
4.3-2.
Core res traint device ins tallation............................
18 4.3-3 Cross section of corc restraint device........................
19 4.6-1.
Control rod assembly..........................................
31 4.6-2.
Control and orificing assembly installation...................
32 4.7-1.
Helium compressor.............................................
36 4.7-2.
FSV circulator machine assembly...............................
39 4
4.8-1.
Typical thermal barrier attachment stud fixture...............
42 A-1.
Approximate relationships between hardness, strength, and stress corrosion susceptibility of carbon and low alloy steels........................................................
57 LIST OF TABLES 1-1.
Components evaluated.........................................
5 i
4-1.
Summary of possible concerns and engineering failure evaluation...................................................
46 i
e G
Page 3
=-
GA TECNNOLOGIES INC.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED i
TO PRIMARY C00LAlff CHLORIDE CONTAMINATION DOCUSENT NO. 907875 ISSIE NO./LTR. N/C 1.
SUlteRY i
An evaluation was performed on the possible effects of chloride contasination-induced corrosion failure on components in the Fort St. Vrain (FSV) HTGR plant.
The components exposed to reactor primary coolant were reviewed.
The components shown in Table 1-1 were selected because they are exposed to primary coolant, contain materials susceptible to chloride contani-nation induced corrosion failure, and the failure could conceivably have an adverse effect on safety.
The results of the study showed that the credible chloride-induced failure 1
modes that might conceivably occtir would not adversely affect public health and safety.
The failures postulated were either similar to or less severe than those already considered in the FSAR.
A metallurgical assessment of chloride-induced corrosion mechanisms and 4
their applicability to primary loop components studied is included in this report.
j 2.
PURPOSE AND SCOPE The purpose of this evaluation is to identify any implications, relative to public health and safety of chloride contamination of components exposed to the primary coolant of the PSV plant.
The scope of work reported herein was limited to the specific compnents listed in Table 1-1.
These components were selected using the following criteria:
e 1.
The components must be exposed to reactor primary coolant, a major i
transport medium for chloride and water corrosives.
I
^
Page 4 l
.m.
_ _ _ _ _, _. _ - _ _, ~, _.
~
GA TECNNOLOGIES INC.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCIDENT NO.
907875 ISSUE NO./LTR. N/C TABLE 1-1 COMPONENTS EVALUATED j
Components Reason for Selection l
1.
PCRV Cavity Liner General corrosion 2.
Plentat Elements Type 347 stainless steels 3
Core Restraint Devices High-strength steel fasteners, type 304 stainless steel
.i retainers 4.1.
Helium Purification System Filters 300 series stainless steels 4.2.
Hydrogen Getter 300 series stainless steels 5.
Instrument and Sensor Lines 300 series stainless steels 6.
Control Rods and CRD Mechanisa High-strength steel fasteners, Assembly 300 series stainless steels j
7.
Helita Circulator High-strength steel fasteners j
and components i
{
Sy sThermal Barrier Attachments Possible high hardness in heat
$l affected zones j
i 4
I s
Page 5
s QA TECNNOLOGIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION D0cUmma NO.
907875 I&ws NO./LTR. N/C i
l
.2.
Components must contain some materials susceptible to chloride-induced corrosion damage such as a) Carbon steel and low alloy steel (these materials may be subject to general corrosion and pitting):
b) Austenitic stainless steels (which may be subject to stress corrosion cracking):
c) High-strength steels (may be subject to stress corrosion / hydrogen embrittlement).
3.
The components with susceptible materials are stressed and/or subject to an occasional noisture envirorument.
3.
INTRODUCTION AND BACKGROUND Subsequent to the June 1984 shutdown of the FSV HTGR plant, stress corro-sion cracking (SCC) failures were observed in certain control rod cables. The cables were made of type 347 stainless steel (SS).
Detailed metallurgical analysis showed the presence of significant levels of chloride on the cable surfaces.
The level of chloride on the cables caused Public Service Company l
(PSC) of Colorado to request a review of the implications of the observed chloride levels on other components exposed to primary coolant.
Accordingly, GA Technologies Inc. initiated this review at PSC's request in December 1984 1
Page 6 i
-. ~... _ _.
.._...._c,_
GA TECENOLOGIED INC.
TITLE:-
EVALUATION OF FORT ST. VRAIN MTALLIC COMPONENTS EXPOSED TO PRIMARY C00LAlff CHLORIDE CONTAMINATION j
00Cuperr NO. 907875 ISsus 30./LTR. N/C i
l.
- In January 1985, while reassembling circulator C2102, one of the high-strength (AMS 6487 type M11) primary closure bolts failed during retorquing due to the presence of SCC.
Examination of this bolt and other bolts exposed to j
the primary coolant and randomly selected from the same circulator showed 1-j evidence of incipient SCC and levels of surface chloride contamination very similar to that observed on the control rod cables. This observation increased j-the importance of the chloride implication review under way at GA.
e The presence of chlorides in the primary coolant of the reactor has a j
ntmber of potentially adverse complications including enhanced elevated temper-l ature corrosion during normal reactor operation and enhanced general corrosion, t
.[
pitting, corrosion, stress corrosion cracking, and hydrogen embrittlement when
)
water is also present in the reactor. These potential mechanisms are described in more detail in Appendix A.
As described in the Appendix, these various corrosion phenomena have the potential to affect different materials in j
different ways. Thus, it is necessary to consider implications on a component-by-component basis.
Based on such a preliminary review, it was concluded that the phencuenon most likely to be important in the reactor are:
- 1..Stresa corrosion cracking of austenitio stainless steel, j
2.
Stress corrosion / hydrogen embrittlement of high-strength steel 1 and 3.
General and pitting corrosion of carbon steel and low alloy steel.
j I
j High-strength steels can experience delayed failures in some environments including those containing oblorides. Opinion is divided regarding the exact l'
mechanisms some attributing failure to stress corrosion oracking, while others believe failure is due to hydrogen embrittlement. The difference between these phenomena, while of scientific importance, is not significant from an engineer-ing viewpoint.
Both processes produce failure.
For the purpose of this f
report, the terms are used interchangeably for high-strength steel behavior.
i I
hp7
l GA TECHEOLOGIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN tETALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLONIDE CONTAMINATION DOCWENT NO.
907875 ISSUE NO./LTR. N/C l
A review of the materials employed in the components exposed to primary f
coolant revealed that most were constructed from materials that are resistant to chloride-induced SCC.
The steam generators, for example, are fabricated from mill-annealed and solution-annealed Alloy 800 and 2-1/4 Cr-1 Mo steel --
materials highly resistant to chloride SCC / hydrogen embrittlement.
- Likewise, f
the control rod cladding and spine are solution-annealed Alloy 800.
The thermal barrier cover plates are Haste 11oy X, an alloy with a high resistance to chloride SCC, and carbon steel, a material which does not suffer chloride SCC / hydrogen embrittlement at moderate strength levels used in these compo-nents.
Similarly, items such as expansion bellows used in the steam generator j
assembly and the instrumentation systems are Inconel 600, a material which also
}
is resistant to chloride-induced SCC. Many other reactor components are fabri-cated from carbon steel and low alloy steel.
These materials are not suscep-4 tible to chloride SCC / hydrogen embrittlement unless they contain regions of.
high hardness due, for example, to certain types of welding.
Nevertheless, a preliminary review of FSV primary loop components showed that some components contained materials susceptible to chloride-induced
{
corrosion as indicated in Table 1-1.
i Both metallurgical investigations and an engineering failure consequence evaluation for the selected components are presented in Section 4.
Appendix A provides a background discussion of the possible compatibility issues raised by
,tl$epresenceofchloridesintheprimarycoolant.
4 EVALUATION OF CHLORIDE-EXPOSED COMPONENTS l.
The susceptibility of materials to chloride corrosion and the components selected for review / evaluation are discussed in the following subsections.
1 Page 8 n
. - - -.._ _ _...., - -.. - _. _.. _. _. _._.= _ _ _..-__.,_
s-GA TECONOLOGIES INC.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUMNT NO.
907875 issue No./LTR. N/C Table 4-1 stamarizes the overall results of the metallurgical and i
engineering evaluation and is included at the end of this section.
4.1 Cavity Liner 4.1.1 Metallurgical Evaluation As noted previously, the low alloy steels used in the cavity liner are not generally susceptible to stress corrosion cracking unless they contain regions of very high hardness. Evaluation of the liner fatrication records indicates I
that such high hardness regions do not exist except beneath the Nelson studs holding the thermal barrier (this is discussed further in Section 4.8).
Fran the standpoint of liner integrity, SCC was not considered to be a relevant failure mode.
General and pitting corrosion were viewed as possible concerns and were studied further.
I 1
The structures most likely to be adversely affected by enhanced general or pitting corrosion rates are the carbon steel liner and the core side attach-ments. This is based on the assumption that, subsequent to water ing*ess, much j
of the water will condense and collect on the liner, particularly in the bottom j
head of the PCRY.
The pH value of the condensed fluid cannot easily be detined. However, if the solution is initially acidic, it is very likely that reaction with the surrounding structures will rapidly dissolve sufficient corrosion products to move the solution back toward neutrality. The condensed solution probably represents a r1ther unique environment which has little industrial precedence (most industrial environments contain significant levels of oxygen, whereas the dominant gas in FSV is helium with very low levels of gaseous impurities).
It is therefore difficult to find a practical experience basis to quantitatively define the likely corrosion rates.
However, one assumption that could be made is that the condensed fluid will act like warm I
Page 9 j
CA TECNDOLOGIg3 INC.
J l
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED 1
TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUMNT NO.
907875 ISSUE NO./LTR. N/C seawater (the PCRV liner is normally at 90*F - 150*F).
Corrosion rates of carbon steel in warm seawater environments typically lie in the range of 5 to 10 mils (0.005 to.0.010 in.) per year.
In some cases shallow pitting is observed, and worst case pitting rates are in the range 15 to 20 mils per year.
These rates, which are applicable to warm, flowing, oxygenated seawater, are probably very conservative relative to the low oxygen, static conditions existing within the FSV reactor. However, for the purposes of a conservative l*
analysis, a corrosion rate of 30 mila a year was assumed and is t. sed in the engineering review.
I 4.1.2 Engineerina Evaluation 4.1.2.1 Assessment Summary i
i The PCRV liner is constructed from carbon steel ASTM A537 Orade B and as such is subject only to general corrosion.
It is estimated that the maximum loss of effective liner material is 0.15 in. over a period of 30 years. Since the liner is fabricated from a 3/4-in. thick plate, it is concluded that breaching of the liner by general corrosion is very unlikely.
- However, requirements have been established and employed to monitor the liner thickness in the bottom head, sidewall, and top head regions by ultrasonic testing (UT) per Technical Specification SR 5.2.14 The UT readings as of 1981 have shown no decrease in liner thickness.
In the unlikely event of a liner breach occurring, radioactive release from the primary coolant will be detected by radiation monitors as described in the FSAR and corrective actions will be taken. Hence, the public health and safety would not be adversely impacted.
4.1.2.2 Failure Consequence Moisture in the primary coolant can be assumed to condense on and collect f
on the liner, especially at the bottom head of the PCRV.
If, by a series of i
l Page 10
~...
1 OA TECi.NOL00IES INC.
1 TITLE:
EVALUATION OF FORT ST. VRAIN lETALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION D0Giram No. 907875 ISSIE NO./LTR. N/C such condensation and subsequent drying, the condensed fluids become relatively l
hign. in chloride concentrations, corrosion and pitting of the steel liner f
ASTM-A537 can occur.
Assaing that the wet condition exists for 2 months per year, it would take approximately 150 years to fully corrode the 0.75-in thick j
liner with a corrosion rate of 30 mils per year. Since the plant has a 30 year 1
life, the most corrosion that could occur during that time using these j
conservative assumptions is 0.15 in., leaving 0.60 in. of effective material 1
(local areas of the liner were ground to a thickness of 0.69 in., leaving a
~
4 minimum effective thickness of 0.54 in. if the most severe corrosion rate is i
assumed in the local area). Therefore, breaching the liner by general chloride corrosion is extremely unlikely. The PCRV liner seismic loads are based on the maximus ground acceleration of 0.15 g for OBE and 0.30 g for SSE.
Even at the estimated maximum corrosion rate, the effects of seismic events on the liner of reduced thickness would not be of any significant concern when considering safe operation of the reactor or public safety.
Should the liner be creached, i
release of the gas containing fission products would be detected by the 1
radiation monitors.
I 4.1.2 3 Safety Consequence It is very unlikely that the liner could corrode to the extent of causing l
failure within the reactor lifetime. Such corrodion, should it occur, would be j
gradual and released radioactivity through the liner breach is detectable by f
radiation monitors as described in Section 11.2 3 of the FSAR.
1 L
4.2 Plenus Elements j
4.2.1 Metallurgical Evaluation 1
{
Certain of the core inlet plenus elements are type 347 SS castings. During j
normal operating modes, the tea.perature of the plenum elements is too high i
l Page 11
OA TECNNOLOGIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUMNT NO.
907875 ISSUE NO./LTR. N/C 1
i (650*F) for moisture condensation to occur unless very high partial pressures of water exist in the reactor.
Even during reactor shutdown, the temperature 1
of. the elements is high enough to avoid moisture condensation. However, it is conceivable that condensation could occia on these elements during very extended reactor shutdowns if high primary coolant moisture levels exist.
Thus, some stress corrosion possibility exists in the plenum elements.
4.2.2 Er.gineering Evaluation 4.2.2.1 Assessment Summary i
Failure of type 347 SS plena elements by SCC is unlikely, but if it occurred, it would not adversely impact public health and safety.
Only the control column plena element material is type 347 SS.
All other plena I
elements are 2-1/4 Cr-1 Mo material which is not susceptible to this type of
)
corrosion.
4.2.2.2 Failure Consecuence A typical fuel region plenum element / orifice valve assembly is shown in l
Fig. 4.2-1.
Failure of the 347 SS control column plenum elements, which might be caused by SCC, would occur only in locations where the chloride concentration is significant, water is present, and a high stress level exists.
1 t
The plenum elements are relatively free from moisture condensation because their normal operating temperatures are above the dew point of the water vapor j
which can be present in the primary coolant, except during prolonged shutdown.
l The infrequent occurrence of water condensation on the elements limit,s the i
opportunities for stress corrosion.
Page 12 i
907875 N/C Pago 13 e
+
i 4
e 7
'~
t t
4y CONNECTING RCD
- 2 >-
c-Q V
CONTROL ROD GUIDE TUSE Y.
4 n
s RESERVE SHUTDOWN y
SYSTEM GUIDE TURE i
I s
v i
W'b SLIP RING ORIMCIINt.IT QPENING i
7 P
g:
4 e<
.sf1 Egg g.>
m NET,uNiNG yoxa-N g%
i t.cWEn auloE Tust L- -
1M 1
W M. - - -
,1,1,r:
s 1
a, KEyEn j
PLENUM
(
. J s,L. *,,.
I l'k _l r 2 ls.
o*.
=
ELEMENT 5
- w m _.
- .D I
(
i STEEL)
N f
,'i! [ *-.-- "
i #
s I-l g,,<
,7 KEYED CCNTMCL j,
RCD El.5 MENT (TYPE 317 SS)
(
g'
/
GulOE PIN x
s
\\,#
TCP REPLECTCR ELEMENT 5 i
~--
j
, Fig. 4.2-1 Top Planum and Orifice Valve Arrangement l
I
GA TECHNOLOGIES INC.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUMNT NO.
907875 ISSUE no./LTR. N/C The long-term stress levels in the plena elements are generally low. The L1ong-term primary load on the element is the differential pressure caused by the coolant flow through the plenum element and the weight of the orifice assembly.
Some of the plenum elements are subject to relatively higher loads for short durations, caused by the dynsmics of column movements in the core. These loads are transmitted by the region constraint devices attached to the periph-j eral plenum elements around the control coltan.
The movements are associated j
with pressure fluctuations, thermal expansions and contractions of the core coltains,
and the occurrence of seismic events.
However, assessment of these loads showed that they are not large and they do not exist long enough to cause any concern.
i' During a seismic event, the moor loads are transmitted by the core i
restraint devices which attach only to the nonaustenitic peripheral planta 4
elements.
The control colan plenum elements do not share the core restraint j
loads. SCC does not cause any degradation in the ability of the control column plenum elements to resist seismic events.
The plenum elements are also subject to secondary stresses caused by thermal gradients. These gradients are small because the elements are exposed to the uniform temperature of the ccre inlet coolant flow.
The effects of l
neutron irradiation on the embrittlement of the metal elements is within acceptable limits. The boron graphite granules in the' planta elements do not generate internal structural loads on the metal element by neutron irradiation swelling because of the void spaces inherent to the granule packing.
No SCC is anticipated in the plenum element because of the absence of moisture condensation in the upper plenum of the core and the absence of long-term, high-stress levels.
i l
1 Page 14 t-
__.,__..._..____.m mm- - _.. _.
i-GA TECENOLOGIES INC.
i:
1 TITLE:-
EVALUATION OF FORT ST. VRAIN PETALLIC COMPONENTS EXPOSED TO PRIMARY. COOLANT OfLORIDE CONTAMINATION DOGurum No. 907875 ISSUE NO./LTR. N/C 4
Even if a plenum element failure is postulated, it would require a severe fracture and distortion to produce conditions which would plug some coolant holes and prevent proper cooling of the affected fuel elements.
Although unlikely, such a failure could locally increase fuel temperature and increase fission product release into the coolant system.
However, the coolant system
{
is continuously monitored for fission product buildup; hence, such a plenum element failure would be detectable. The increased fission product inventory would be retained in the PCRV and processed by the helium purification system.
I A planta element which would produce the above scenario can be removed and replaced during normal refueling operations.
l 4.2.2 3 Safety Consecuence Accelerated fission product buildup could occur from a fuel temperature i
increase if the coolant flow to the fuel elements was restricted by an unusually severe plenum element failure. - However, there is no adverse impact (Ref. Technical Specification LCO 4.2.8, " Primary Coolant Activity") on public health and safety by a severely damaged plenum element because fission product buildup. is continuously monitored and would not increase sufficiently rapidly that corrective operator. action could not be taken (Ref. Technical Specifica-tion SL 3 1, " Reactor Core Safety Limit").,
4.3 ~ Core Restraint Devices 431 Metallurzical Evaluation The carbon steel core restraint devices contain hardened austenitic stainless steel (A286) bolts with 304 SS retainers which hold the Inconel 718 l
pins.
The operating. environment surrounding the core restraint devices is j
similar to that of-the planta elements.
Since it is conceivable for 1
l Page 15 r
j
--y--.we wi.
w.
f w wwv w
,va.vc,wo-y yvv-y
.c w 9y---p.'w--wwyp.g-y-w-T--
eywvey tmw-
--,-we,gwy
%-g-7y y3, gww,www-w w1mmyw y---
y--.----,-y-y-,9.-,-wew-wpw9-goi.-wc---=g-yw,--,-pw-p tw g -
GA TECHNOLOGIES I N C.
TITLE:
EVALUATION OF' FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY-COOLANT CHLORIDE CONTAMINATION DOCUMNT NO.
907875 ISSUE NO./LTR. N/C condensation to occur in the regions, an assessment of potential stress corrosion of the core restraint device was performed.
4.3.2 Engineering Evaluation 4.3 2.1 Assessment Summary Chloride corrosion induced failure of the stainless steel bolts and retainers of the core restraint devices is not expected. If a large number f restraint devices were postulated to fail, it could result in the initiation of core temperature and nuclear flux channel measurement fluctuations.
Such fluctuations will be readily detectable as described in Section 3.6.6. I of the FSAR.
It is assessed that the public health and safety would not be adversely impacted by core restraint device failures.
4 3.2.2 Failure Consequence The core restraint devices (Figs. 4.3-1 and 4.3-2), like the plenum elements, normally operatie at temperatures too high for moisture condensation to occur. In addition, owing to the configuration of the stainless steel bolts and retainers in question (Fig. 4.3-3), exposure to moisture is considered an unlikely occurrence.
Further, redundant bolt / retainer sets are employed and these are stressed to only approximately 60% of yield strength. Assuming that condensation could occur on these items with resultant chloride-induced stress corrosion, the torqued bolts and their retainers would have to be rendered completely ineffective for the core restraint devices to fail to function by allowing separation or unlocking of. adjacent plenum elements.
-e Failure of one or all 84 core restraint devices will not impact.he safe-operation of the reactor.
The reactor operated without them until 1979 with occasional temperature and nuclear flux channel measurement fluctuations.
i i
Page 16
g o
p' ~...
l
{
~)
907875 N/C
- j T
i xa uo if I.8"2--
suasT_ ~ 3 Pago 17 50 si
~
=
!3
%-j 3
U ! 8 *s $,,
s.= W.I
-a ST L l !g
- n I
l*
i
}%{
w $ o i=*
f
}';'p 3
h lf!
l l
5 "dW Y
,l 9.?..
\\
/
,/
\\
se I
i
/
!!jl s
11
']
.,,8 g
N -J
/
k, i iii 8'
/
\\
si s
s/
- ; o.- i
?
r
- )
- l e
I !
(,1.,
s
\\.
i
\\ '-
/
O s
a
,/3
\\
I j
.xg; U,-
I I
a o
j
/
.. (g i
=
=
e j
p
,3,
/,;
/*
,I g,
~
i g
/
i
- Qj 6-e
!Q 3l
(
a
.\\g
.,i
/7
\\.
g
=
'-Q /
..\\
/L M.
,y
/ 7 A.
, e
=
sx;.py,l$r
.'s.
hp S
t
907875 N/C Page 18 GRAPPUNG HOLE
)
l ANDUNG INDEX 0T.WL 1
\\
i k
I
[!
e ENGAGERAENT PIN
)
I
/
g
@ (
x 4
-l s
s (
. I:
I
% e =e. -
n
.g
> c C 8 P
'l if A
3
~
- -]
N I
s
- i ed og /
l.a i i
c o
i 28e
's-6 c l
l O O O
.J l
/
O cC s
cC g
o C.0,
'YSL* $ c.
a' \\/M l/
\\
+
s x
u N
'.)
?
Fig. 4.3-2 Core Restrainc Device Installation e
-r,
,..... ~
n m
907875/N/C Peg 2 19 5
1 A
CARBON STEEL
~
1-304 RETAndidq Pld 8
A 288 g
l SCREWS t
E l
M df C
i.
( k ',
[/,G I
l
\\
l s
\\
l n
y
,,3,,/
CARBON STEEL plNC 718 b
Fig. 4.3-3 Cross Se kon'ofCoreRestraintDe ce O
-,-w----r
- - - - - +
m
=*
9 GA TECHNOL0GIES INC.
TITLE:
- EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUMENT NO.
907875 ISSUE NO./LTR. N/C Failure of one or several of the devices would not likely be detectable during plant operation owing to the redundancy of the configuration, i.e., each core region contains six restraint devices.
Failure of a core restraint device would be detected during refueling of a region with the failed device.
A failed restraint device would be removed and replaced.
The failure of many core restraint devices could result in core thermal fluctuations. These would be observed in the primary coolant circuit at individual core region outlets, at the steam generator module helium inlets, and in the main and reheat steam outlets, as described in Section 3 6.6.1 of the FSAR.
Individual nuclear flux channel fluctuations wo'uld also be -observed.
In assessing the effects of seismic events following the failure of core restraint devices owing to stress corrosion, it was concluded that the results would not be of any more concern than those previously discussed; i.e.,
failure of all the restraint devices will not impant the safe operation of the reactor.
If the bolts fail to retain the Inconel 718 pins to the body of the restraint device through the effects of chloride stress corrosion, the pins would remain engaged in the plenum element handling holes upon removal of the device.
The pins would then be accessible for removal by the in-core service handling equipment which are provided as accessories to the fuel handling i
machine.
4.3.2 3 Safety consecuence I
Failure of the core restraint devices could result in temperature and I
nuclear flux channel fluctuations. However, as discussed in Section 3.6.6.1 of the FSAR, a safety evaluation concluded that fluctuations would result in no undue risk to the health and safety of the public.
l l
l Page 20
GA TECHNOLOGIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION f
DOCUIENT NO.
907875 ISSUE NO./LTR. N/C 4.4 Helium Purification System and Hydrogen Getter 4.4.1 Metallurgical Evaluation The helium purification system contains some austenitic stainless steel components.
The high-temperature filter-adsorber, for example, contains 300 series SS filter elements. - The hydrogen getter is 300 series SS except for the titanium sponge.
These components are susceptible to SCC by exposure to primary coolant chlorides.
4.4.2 Engineering Evaluation 4.4.2.1 Filter Elements 4.4.2.1.1 Assessment Sumar y Postulated failure of these filter elements will not impact activity releases, will not introduce debris into the PCRV larger than about 100u in diameter,. will not impact the operation of any other system, and will not prevent continued operation of the helium purification system.
Furthermore, such failures will not adversely impact the public health and safety.
4.4.2.1.2 Failure Consequence Filter elements are installed in the high-temperature f11ter-adsorbers (HTFAs, A-2301 and A-2302), purified helium filters (F-2301 and F-2302), and hydrogen removal filter (F-2304).
These filter elements consist of a blanket of type 347 SS fibers which is wrapped around the outside of a perforated support sleeve fabricated from 1-1/2 i
Page 21
GA TEC5NOL0GIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION i
DOCUNDIT NO.
907875 ISSUE NO./LTR. N/C b
in. Sch. 40 pipe.
The support sleeve and associated fittings are fabricated l
fra 300 series stainless steel. The fibers are compacted and sintered to form a rigid filter mat.
Each filter element is nominally 2 in, dia x 20 in, to 40 in, long, and the perforations in the support sleeves are 1/4 in. dia. Each of the listed filtering vessels contains seven to ten such filter elements. All other parts of these vessels are fabricated from carbon steel and low alloy steel, and are not subject to stress corrosion cracking.
Each of these filtering vessels is characterized by a top helium outlet chamber,'a dead-leg catch basin below the filter elements, a hellun inlet line near the top of the catch basin, and a very low flow velocity of about 1 ft/
sec.
In the event of a filter element failure, the low helium flow velocity would not be capable of transporting any particle greater than'100u in dia.
Larger particles would only settle harmlessly to the botta of the catch basin.
The HTFA filters are the only ones which should be exposed to the chemical impurities necessary for stress corrosion cracking.
The HTFAs are also the
{
only filtering vessels which have the potential for containing radioactive particulates. Any such particulates transported through failed filter elements would be expected to be deposited in downstream components such as the knockout drum, dryer, low-temperature charcoal adsorbers and back-up filters.
. TheiHTFAs are located in the top head of the PCRV and are the best candidates for " dropping" filter element debris into the PCRV.
However, any such debris would have to contort into the high-elevation filter chamber inlet port ~ and then pass counterflow through two carbon steel screens (0.034 in.
opening), a charcoal bed, and a cooling coil (0.125 in, gap) before it could enter the PCRV.
4 Therefore, failure of the filter elements in the helium purification
- system will not impact activity release, will not introduce debris into the PCRV larger than 100u in diameter, will not block helium flow through the i
Page 22
-o
GA TICHNOLOGIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUMNT No.. 907875 ISSUE No./LTR. N/C helium purification systems, and will not impact the operation of other system.
Likewise, the consequences of an earthquake will not be altered if these filter elements fail prior to or during the earthquake for the same reason.
Failure of these filter elements will not impact plant depressurization following a loss of ~ forced circulation accident (DBA-I, Design Basis Accident-I), since such failures will not interfere with the continued opera-tion of the helium purification system.
4.4.2.1.3 Safety Consecuence Failure of filter elements in the helium purification system will not adversely impact the public health and safety.
Such filter element failuras will not impact activity release, nor introduce debris into the PCRV, nor interfere with the continued operation of this or any other system.
4.4.2.2 Hydrogen Getter 4.4.2.2.1 Assessment Summary Stress corrosion cracking failures are not expected to occur, since the impurity conditions necessary for the corrosion process (chloride and water) do not exist within the hydrogen getter section.
Nevertheless, even if such a failure were postulated to occur, it would not impact activity releases; would not introduce debris into the PCRV any
-larger than 100u in diameter; would not impact safety-related plant depressur-ization capabilities, since the hydrogen getter section is bypassed under such i -
con!1tions: and would not adversely impact the public health and safety.
l l
l l
Page 23
1.
GA TECHNOLOGIES INC.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUIENT NO.
907875 ISSUE NO./L7R. N/C 4.4.2.2.2 Failure Consecuence The hydrogen getter s'oction is fabricated from 300 series stainless steel and consists of titanium sponge filled hydrogen getter units (A-2309 and A-2309S), plus piping up to and including the isolation valves. The hydrogen i
getter section does not receive the impurity conditions necessary for stress corrosion cracking. The hydrogen getter section is installed at the outlet end of the helium purification system and processes only helium which has had all impurities removed with the single exception of molecular hydrogen. This is a process and design requirement, since even ppa levels of other impurities will
" poison" the surface of the titanium sponge gottering material and render it ineffective for hydrogen removal. Whenever such titanium " poisoning" occurs, the titanium sponge material must be either removed and replaced or must be reactivated at highly elevated temperatures.
For this reason, the process gas to the hydrogen getters is first purified by the helium purification dryers (molecular sieve) and the low-temperature adsorbers (charccal at -320*F) to remove all impurities except molecular hydrogen. Since chloride-induced stress corrosion cracking requires the simultaneous presence of chloride and water, the necessary conditions for stress corrosion cracking do not exist in the hydrogen getter section.
Nevertheless, even though conditions do not permit stress corrosion cracking, the implications of such a failure have been evaluated. In the event of an internal screen failure, the low bed flow velocity of 1.5 ft/see would not be capable of transporting any titanium or steel particle larger than 1009 diameter.
Furthermore, even if ~all included 300 series stainless steel compo-nents were to simply collapse, the impact would be as follows:
The hydrogen
. getter section would be isolated and bypassed; plant depressurization capabil-ities of the system would not be impaired; and, the health and safety of the public would not be endangered.
Page 24
.m- -, -..
GA TECHNOLOGIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCmENT NO.
907875 ISSUE NO./LTR. N/C i
Failure of the hydrogen getter section would not impact plant depressur-ization following a loss of forced circulation accident (DBA-I, Design Basia Accident-I), since the getter section is bypassed under such conditions. Like-wise, the consequences of an earthquake will not be altered if the hydrogen getter section fails prior to or during the earthquake for the aute reason, 4
Full helium purification flow during normal operation could be regained by installing one non-nuclear isolation valve on the cold-inlet side of the hydrogen removal economizer.
4.4.2.2 3 Safety Consequence i
Stress corrosion cracking failures should not occur, since the impurity conditions necessary for this corrosion process do not exist within the i
hydrogen getter section.
Nevertheless, even if such a f ailure were to occur,-
I it would not adversely impact the public health and safety.
4.5 Instrument and Sensor Lines 4.5.1 Metallurgical Evaluation A review of the Moisture Monitor System and the primary coolant pressure and analytical instruments of the circulator auxiliaries indicate that some of the instruments and sensor lines are made from 300 series stainless steel and therefore may be susceptible to stress corrosion cracking.
)
l I
Page 25
GA TECBDOLOGIES IDC.
4 TITLE:
. EVALUATION OF FORT ST. VRAIN lETALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUIENT NO.
907875 ISSUE NO./LTR. N/C 4.5.2 Engineering Evaluation 4.5.2.1 Assessment-Summary An analysis of the Moisture Monitor System indicates that the failure of an instrument, sample line, or other part containing primary. coolant would result in an alarm in the control room, and that there would to no loss of safety function.
The sampling lines are small; therefore, a ?.ine break cannot cause a significant primary coolant leak. The public health and safety are not adversely affected relative to the primary coolant leakage scenarios evaluated in Section 14.7 of the FSAR.
1 Failure of,an instrument line that provides a primary coolant pressure signal to the PPS may result in the trip of a single PPS channel as ' evaluated in Section 7 of the FSAR.
However, public health and safety are not adversely affected because there is no loss of safety function and because the size of the sampling line is small.
The analytical instrtment sample lines are not safety-related.
Primary
{
coolant released due to the failure of an instrument line is contained inside scue of the penetrations.
Because of the small size of the line, a release of primary coolant is not s'gnificant. The health and safety of the public are i
not adversely affected.
4 f-Breaking of any instruments or sensing lines in the circulator auxiliaries I
cannot result in a significant primary coolant leak because of the small size I
of the sensing lines. Stress corrosion cracking is not credible for any of the sensors at the upper and of the circulator because the instruments normally see only dry, moisture-free helitat.
They are the instruments that monitor and l
control the buffer helium supply to the circulator.
i e
Page 26
-e
,--n
- - ~ - - -,.,
,_,--,,,-,-a
-,--,-.,n
-.,..,---n
,,-.-----e e.--,
GA TECHNOLOGIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUIENT NO.
907875 ISSUE No./LTR. N/C
)
t Instruments and sensing lines below the midbuffer line of the circulator cartridge normally or occasionally do see wet helium and, though' the chloride content of the mixture should be low, these components may be subject to utress corrosion cracking.
However, the he'alth and safety of the public is not adversely affected due to the PPS action and control room alarms.
I 4.5.2.2 Failure Consecuence t-The following failure consequence was identified for the moisture monitor system.
Primary coolant gas is. circulated from the sample rakes at the circulator diffusers, through the process and moisture penetrations, and back to the discharge point at the circulator inlet plenum.
Most of the parts in contact with the primary coolant are stainless steel.
Valves, fittings, and other machined metal parts are welded in the sample system inside the I
penetration interspaces.
Pipe breaks or cracks are automatically detected and alarmed by the primary coolant sample bypass control system as discussed in Section 7 3 2 of the FSAR.
The affected channel is then manually tripped by the operator.
Furthermore, even a sudden guillotine rupture (e.g., that caused by a seismic event) of all instrument lines penetrating the secondary interspace of a l
penetration cannot result in a significant primary coolant leak due to the l
small size of the sampling lines.
Therefore, there can be no degradation of safety functions.
The following failure consequence was identified for the primary coolant j
pressure sensing system and analytical instri,mients.- The primary coolant and analytical instrument sample system contains series 300 stainless steel. If a crack occurs in a line close to the sample point, purge helium may leak back into the reactor, in which case an increase -in the helium purge flow would be alarmed.
Or, primary coolant may leak into a penetration interspace.
If the Page 27
4, G A-TECCNOLOGIES INC.
TITLIs.
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLAlff CHLORIDE CONTAMINATION DOCUMNT NO.
907875 ISSUE NO./LTR. N/C leak into the interspace is small, primary coolant would be contained within tne interspace.
If large, the PPS channel would trip and the primary coolant would still be contained.
This discussion appears in Section 7 3 3 2 of the FSAR.
Furthermore, because of t!'.e size of the lines, there can be no significant s
l primary coolant leakage. Therefore, there would be no release of radioactivity or loss of safety function, and the health and safety of the public would not
- be adversely affected.
l The investigation of failure consequence for the circulator auxiliaries i
showed the following.
Cracks or breaks in instruments or sensing lines that monitor or control the helium circulator auxiliaries do not cause a loss of i
safety function.
Malfunction in these instruments above the midbuffer level result in rapid acticn by the PPS to isolate the circulator and alert the operator (this is discussed in Section 7.1 of the FSAR). This limits exposure
~
~
of these instruments to water. Therefore, they are not susceptible to credible stress corrosion cracking.
Other instrtaients and sensing lines that may be exposed to chlorides are located lower on the circulator and monitor the various water drains.
These instruments are exposed to helium and water most of the time.
They monitor conditions of circulator operation by monitoring differential pressures across L
the helius-water, main drain, and steam-water drain cavities.
'If stress l
corrosion cracking did occur in one or more instrument / sensing lines, it would be detected and that channel would be isolated or the affected circulator would be shutdown as discussed in Section 7.1 of the FSAR.
Furthermore, due to the small size of the lines involved, there can be no significant primary coolant leakage nor loss of safety functions due to seismically caused breaks or due to breaks from other causes.
i Page 28 i
l
._=
5 GA TECHNOLOGIES INC.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUIENT NO.
907875 ISSUE NO./LTR. N/C 4.5.2 3 Safety Consecuence Becsuse the PPS acts to isolate equipment and trip a failed channel, there are no loss of safety functions.
Therefore, there are no conceivable safety concerns, including impact on public health and safety, for the instrument and sense lines of the Moisture Monitor System, the primary coolant pressure and analytical instruments, and the circulator auxiliaries.
4.6 Control Rod and Control Rod Drive Mechanisms 4.6.1 Metallurgical Evaluation The control rods and control rod drive mechanisms utilize 347 austenitic stainless steel cables and numerous precipitation-hardened austenitic stainless steel A286 bolting. A high-strength austenitic stainless steel (A286) bolt and a captured nut attach the cable assembly to the control rod. The top portion of the control Pod, the clevis, as well as a jam nut and tab washer, which prevent the. control rod spine from unscrewing in the clevis, are all made of austenitic ' stainless steel (see Fig. 4.6-1) and therefore susceptible to stress corrosion cracking.
4.6.2 Engineering Evaluation 4.6.2.1 Assessment Summary The cable and bolt end fittings have been changed to Inconel 625. Certain other elements of the cable assembly that may be susceptible to stress corrosion cracking are now changed to materials resistant to chloride-induced stress corrosion.
4 Page 29
..m
GA TECBNOL0GIES INC.
TITLE:-
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUMIrf No. 907875 ISSUE NO./LTR. N/C Stress corrosion failure of the bolt, nut, and clevis is unlikely.
The tab washer and jam nut, however, may have significant stress levels.
If failure of any or all of these parts is postulated (even if failure occurred during a seismic event), the control rod would drop into the core.
Public health or safety would not be adversely affected due to the failure of one or
- more of these parts.
4
~
The stainless steel (A286) nut, which attaches the control rod shock absorber to the control rod spine (see Fig. 4.6-1) supports only the weight of the shock absorber, and therefore the occurrence of stress corrosion failure in this nut is not expected.
Should the nut fail, an existing expansion of the spine below the nut will' prevent the shock absorber fra detaching from the rod.
1 The possibility of moisture condensing at the top plenum is low, but stress levels in the A286 nuts and bolts connecting an actuating rod to the outer - cylinder of the orifice valve (see Fig. 4.6-2) may. be significant, depending on torque values. Stress corrosion, although unlikely, is credible.
Failure of these parts while the plant is operating would cause the outer cylinder to fail, which would fully open the orifice and provide the maximum amount of cooling flow.
As discussed in the FSAR Section 3.6.7, this would degrade plant performance but would not affect public health and safety, even if failure occurred during a seismic event.
No adverse -impact on public health or safety would result fra the failure of any or 'all of the above parts.
Page 30 c-r.
,,.. ~, _...,
_,_..m
907875 N/C Pago 31 k ;
'l
{g:
p.
l g
e I
l Ms
\\
fI l
I TAM C_%25 q}
I 9 hc/
I e.P. ' "'>
u t
l
. hg s:
yp-
[
4, p
w.,
Th
'. /
$Nk. )y)
"S a
._I I
- 3 g
,T
,,,,(/
t ll1 i
's r
4
- .1 k
w-l4p
~
a c
pp s,
t e
m c
Lv7 m'
- .:-& i n.
%as 4m ata.
I hm/
9 1
ic a
t w jV
+
p + En K
R.
p t
i
- '=
3 s
D D k' g k
m h /g
'ta.
13 ?
la s
y
.-- S R.
I t
Q kj giils d.
g x
e
]..
Q
_.4 4
i L.
1 s<
i.
4
-g p
e((k.p j b DMD'
.]
,cu--
ls@E.$
[
v /A SM i
i
-3p. ngi,, n,;,
1 e.
i s
e
+
u 8
/
~ (l I
N' 4 V i
i
{
.~.
l
~
c f
ni i.
-l ty 7
3 1
1 y
I6 C
,.=J l
i
- it i
=
c g
W L
Q.
p.; pA 4
I.= ~
N l
P ge i
5f
- g DL 4
s, d T.
^ ;- G.
/
i c
t,
{
Il ga !<-i,ili e._
sp i
i i
s il l
- 1. I';j.l!:, l' 1
'I d, -b l y...
U s
t
$t
.D I p-i; l -.s Fig. 4.6-1 Control Rod Assembly 1
907875 N/C i.
l l
Page 32
"=
S 5
g I
1 il
/
t
~
E I
g l
i
. mum l
rlNI N
l g
i
.,..,... g 4
I 4
P
[ lL i nl 1
t I'
i'" k, I
3 I
l 1
a i
g" Fig. 4.6-2 Control and Orificing 3
Assembly Installation
~
a
e, 1
GA TECHNOLOGIES INC.
l -
1 i
TITLE:
EVALUATION OF. FORT ST. VRAIN METALLIC COMPONENTS EXFOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION Docuterr NO.
907875 ISSUE NO./LTR, N/C 4.6.2.2 Failure Consequence e
l When the rod is in the fully raised position, any condcnsed water from the upper control rod drive cavity could run down the control rod drive cable and wet the bolt, nut and clevis, as well as the jam nut and the washer.
If the water contained dissolved chlorides there would be potential for stress corrosion.
The nut and bolt can be assembled into the clevis without torquing the bolt. In this condition the only load on the bolt and clevis would be the dead weight of the control rod, which would induce a shear stress in the bolt of
<560 pai, which is much less than the level normally required to initiate 1
stress corrosion.
Stress levels in the clevis would be even lower and the nut would have no stiress.
Stress corrosion in these parts is therefore very unlikely (provided residual stresses in the parts are low).
The tab washer may have large residual stresses in the taba and may therefore experience _ stress corrosion.
The jam nut is torqued against the i '
washer and clevis with 20 ft-lb, resulting in a stress of 26 kai in the nut threads.
Failure or loosening of these parts would reduce the friction resistance that inhibits the rod from rotating and unscrewing from the clevis.
It would then be possible (but not necessarily probable) that the rod would unscrew and fall into the core.
i
.If a rod detaches from the cable assembly, the slack cable switch would indicate that the control rod had fallen-(see Section 3.8.1.1.1 of the FSAR).
In addition, the fallen control rod would cause a local reduction of core gas outlet temperature and degrade plant performance, which should be detectable.
4 If the control rod fell, the shock absorber located at the bottom of the rod would absorb the impact and prevent damage to the rod or the bottom l
Page 33
^
OA T E C E N'O L 0 G I E S I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN ETALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUMirr NO.
907875 ISSUE NO./LTR. N/C reflector blocks.
The plant would be shut down, the rod would be retrieved, and a new shock a"bsorber installed (see Section 3.8.1.1.1. of the FSAR).
The nut attaching the shock absorber to the control rod spine is located at the bottom of the control rod where water is very unlikely to condense. The only load on this nut is the weight of the shock absorber; stresses are low and stress corrosion is therefore unlikely. The nut is the same captured type used I
~
at the clevis it would have to fail in more than one location to separate from the control rod threaded spine.
The threaded spine is expanded below the nut to keep the nut from loosening.
This same expansion also prevents the shock absorber from falling off the spine if the nut dropped off.
Failure of the nut in a manner which would permit the nut to separate from the spine is probably not credible; and even if this event occurs the shock absorber will remain attached to the control rod.
d 4
Water is not likely to condense in the orifice valve area, and stresses in the bolts and nuts that connect the orifice valve actuating rod to the orifice valve outer cylinder are low. Stress corrosion of these parts is not probable.
Failure of these bolts and nuts would fully open the valte and result in f
increased gas coolant flow through the valve, which would cause a local reduc-tion of core outlet gas temperature. The failure scenario is discussed in the
[
FSAR Section 3 6.7.
This change in gas temperature might not be detectable.
l Failure of the parts would be detectable in the hot cell during routine l
l maintenance.
As failure results in a local increase of cooling gas flow, l
there are no adverse safety consequences.
Page 34
GA TECHNOLOGIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUMNT NO.
907875 ISSUE No./LTR. N/C 4.6.2.3 Safety Consequence
' The drooping of a control rod into the core and/or accidental opening of the orifice velves are events which were evaluated in the FSAR and shown to have no adverse impact on public health or safety.
4 4.7 Helium Circulator 4.7.1 Metallurgical Evaluation s
High-strength steels are utilized in fasteners (e.g., bolts) in a number of components of the helium circulator. The primary closure bolts that failed in circulator C2102 were of AMS 6487 type H11 steel which has a minista yield strength of 215 kai.
As indicated in Appendix A, these bolts fall into the chloride-induced failure prone group.
The circulator also contains A286 bolts, which are susceptible to SCC.
i The helita circulator compressor blades are of 422 stainless steel l
hardened to R 40 maximum and the compressor disk is of D6AC alloy hardened to n
the range R,40-44 (see Fig. 4.7-1).
Potentially, these components are suscep-tible to stress corrosion.
However, the stresses on the blades and compressor
. are low except during reactor operation. During operation, water cannot reside
{
on the compressor due to the speed of rotation.
During reactor shutdown stresses will be low.
i J
Page 35 i
-. ~.
[
SLADES th0T $NeuAll ARE A22 TEtIPERES 3
TO R 4888AX I../ TO C
4 OlA 27.000 M
8800V7181 CORPORATION.
OSAC F044sIIS TEaIPE9ES ll26 HERIO50CES CAUSEWAY, C
f RIGEFIELD.leEWJERSEY y
2 AFTER BLADES HAVE BEEN 5
ASSEGASLED AND LOCKED Q
"hE'"'
te.sese.se2 dea REF g
y 3 AFTER SPINNING. MACHINE g
OUTSIDE DIAteETER AS A
SHOWN e
A
_L_
I ei lo 1",.
y.
- a da SEtil001A-A SCALE 1/5 ELECTRO ETCH BLAGE NoteeER lli THit
\\
AREATYPS2 PLACES 8.030IAATERIAL REF reevioto FoA Fles SH MACHHNIIS j
2
- l$
's a 1
!w-a 4
=
Fig. 4.7-1 IIelium Compressor 3
1 1
G A.
TECHNOLOGIES I-N C.
t TITLE:
EVALUATION OF FORT ST. VRAIN ETALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUM Nf No. 907875 ISSUE NO./LTR. N/C i
4.7.2 Engineering Evaluation
)
l 4.7.2.1 Assessment Summary 1
Thirty-one different types of fasteners utilized in the FSV circulator machine assembly were examined for susceptibility to chloride-induced stress corrosion, four of them were found to be susceptible to the conditions that may
/
result in stress corrosion cracking.
Failure of-three types of bolts (90C2101-310-4, -340-9, -380-10) may affect only the functional capability of the circulator and will not cause a breach of the primary coolant boundary. However, failure of the inner diameter i
primary closure bolts (90C2101-300-40'), combined with fallure of turbine housing torus (90C2101-431) could release primary coolant into the reheat steam piping. The isolation valves provided in the reheat steam piping will inhibit release of primary coolant to the environment, if the primary closure bolts and the turbine housing fail.
j-1 The compressor disk (90C2101-362) and blades (SOC 2101-363) have been 1
l identified to be potentially susceptible to chloride stress corrosion.
It was also established that during compressor operation any liquid is thrown off.
This precludes - the simultaneous occurrence of stress and aqueous chlorides I
required for stress corrosion damage. In any case, the failure of the circula-tor compresMor disk or blades would be contained within the envelope of the l-.*
circulator inlet and would only affect the function of the particular l
circulator, with no other adverse consequences.
There would not be any adverse impact on public health and safety as.a result of the failure of the components identified.
The consequences of the circulator primary closure bolt failure causing primary coolant to enter the Page 37
GA TECHNOL0GIES INC.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUISNT NO.
907875 ISSUE NO./LTR. N/C steam piping system would result in a failure scenario as discussed in the the FSAR Section 10.2 3.4.
4.7.2.2 Failure Consequence Figure 4.7-2 shows the circulator machine assembly.
The pertinent parts i
discussed in this section are identified.
s Postulated failure of the primary closure bolts (90C2101-300-40) under certain plant operating conditions (i.e., full helium pressure combined with very low or no steam pressure) would cause the circulator machine assembly to i
f move downward due to the helita to steam differential pressure force, causing the stresses in the toroidal housing (Part No. 90C2101-431) to exceed the yield j
and possibly the tensile strength values. The structural failure of this part would allow for the primary coolant helita to enter the reheat steam piping (see Fig. 4.7-2).
4 The failure of the primary closure bolts could cause primary coolant helium to enter the secondary closure and the reheat steam piping. There are l
two steam isolation valves downstream of the circulator.
The first valve, typically HV-2249, downstream of the circulator ' (i.e., in the cold reheat) would trip automatically on the high shaf t wobble and other. circulator trip i
signals.
The second valve typically HV-2253 (i.e.,
in the hot reheat) j downstream of the reheaters would trip on the radiation monitoring PPS signal (caused by the primary coolant in the steam) if the first valve at the
. circulator turbine discharge fails to close.
The closing of the second valve is estimated to be within seconds of signal initiation.
Closing of either valve will isolate and contain any primary coolant leakage within the steam piping system and cause shutdown of the affected loop. The consequences of the circulator primary closure bolt and turbine housing fa1. lures, causing primary coolant to enter the reheat steam piping system, would result in a failure Page 38
907875 N/C
]
Page 39 l
i l
90C2101-362 DISK 1
C2101-310-4 90C2101-380-10 BOLT BOLT j-i e
90C2101-363 a3 i
BLADE N
90C2101-340-9 f* M 4 m,, -.,..
EN i
BOLT 4
. :d' C h',-
f i
- ,1, w w y D:'
a mLT sw n' 'Miss.
a
gi#.!!N
. dk BEARING
!@st' tS
\\'
PRIMARY l
,: d4,p CLOSURE
_ f.,
51 _ HOUSING l,/
, A.,j,,sN.
~
v t s
s
/I l
V,
+-
i v
i <
y f$
l k
~
l x ~Y' s
\\
. 3.@
hy r
u
' ?),E$
1T
.s.
'L N
l ?C;
}
' ', iw x
I
'?l.,
'l
- h
4-P.
.si z r.
4
\\
m i
h:Ib
[/
. \\
01t be h,k,.
gl
ft Fig. 4.7-2 FSV Circulator Machine Assembly.
l
G A-TECHNOLOGIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUIENT NO.
907875 ISSUE No./LTR. N/C i
I scenario as discussed in Section 10.2 3.4 of the FSAR.
As discussed in the FSAR, this failure scenario has no adverse impact on public health and safety.
j The initial investigation indicates that failure of the primary closure bolts may or may not be readily detectable. This would depend on the pressure
~
differential between the primary coolant and the secondary coolant pressure, i.e.,
whether the flange is being compressed or separated.
If the flange l
separates, the drop in secondary closure pressure will 'cause actuation of PPS 1
alare.
i i
Postulated failure of three other circulator bolts (90C2101-310-4, 340-9, and 380-10) may all result in a loss of circulator function.
Any of these failures could interfere with compressor rotation and the the worst case failure would be loss of compressor blading, i.e., loss of circulator function.
The overall circulator design incorporates features to contain any disk or blade failure within the envelope of the circulator inlet.
Thus any bolt, blade, or disk failure in the compressor region could cause a loss of circulator function; but damage would be restricted to the circulator parts.
All consequential failures would be detectable with the shaf t wobble instrumentation.
The worst case failure (disk burst) would result in a repairable failure that would cause a plant shutdown.
The effect of seismic loads on a degraded component is negligible considering that the seismic loads would constitute a small fraction of the-load due to bolting forces. Seismic or circulator generated dynamic loads are well within the safety margins of the bolts.
l l
Page 40 i
l l
GA TECHNOLOGIES INC.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUMNT No. 907875 ISSUE NO./LTR. N/C 4.7.2 3 Safety consequence No postulated circulator failures induced by chloride stress corrosion have been identified that would result-either in fuel failure or breach of the t
primary coolant boundary in a way to adversely affect public health and safety.
Obviously, should a failure of all the primary closure bolts occur the primary coolant boundary would be breached, but the primary coolant would be contained within the reheat steam system as a result of isolation valve closure.
4.8 Thermal Barrier Attachments 4.8.1 Metallurgical Evaluation The class A and class B thermal barrier attachment studs are attached to the carbon steel liner by Nelson stud welding.
This process produces the i
approximate equivalent of a resistance flash butt weld between the carbon steel 1
stud and the carbon steel liner. The configuration of these studs is shown in Fig. 4.8-1.
Since the liner is relatively cold at the time of stud welding and the assembly is not post weld heat treated, there is a significant probability 1
of ferming a hard, martensitic, heat affected zone (HAZ) in the liner immedi-l ately underneath the studs.
Examination of available records indicates that the liner steel has a carbon equivalent of about 0.6% which would provide the potential for significant hardening. In recognition of this in process welding l
j controls were employed during construction so that the liner HAZ stayed below a l
hardness of approximately Rc 35.
In the final assembly of the thermal. barrier to the studs, a sleeve is threaded on the stud, butted against the liner, and tightened so as to produce a tensile stress of approximately 37 kai in the stud.
The hard zone beneath the stud is therefore continuously subjected to tensile stress.
Thermal barrier studs are ASTM A108 Grade 1018 carbon steel.
and the PCRV liner is ASTM A537 Grade B.
l Page 41 l
1 e
-,,<w4 w..
e.
,9 9,m-7
,,,e_,,3 g
,.,.y
907875 N/C Pago 42 POST GUTER NUT SPHERICAL BUSHING OUTER q a
/
WASHER COVER WASHERS
^
N p%.
PLATE uninsmann ej s
w//n--a s,, '
- d. N.:,+.: :@
, - ' [
- i.
'3 y,,,,
g--
l SHEET c-~~~~'.-
,)
, :^,t,sL un s,
8' ! OUTER
' ^':
.[ OUTER e
RETAINING:s '
O '
^
6'/ l KA0 WOOL ^' '
s
,'}+~+-SPACERM
>l, I-
'; RINGc -,. '~ ~,';r,s_' w.,
"s
.:M j BLANKET '
s
,', n n.
- n. _.
,s s
WASHER',s,',.
~s s
4;. ',
~,',,.'s s-s s >
s tr,
- . ',,,, v///n/ns/, nu kw
//nn un g
s h.-
INNER y,..s: :.' g&3 y
y 0
TE <+.' ' d 'U',' l'0~"*-"~Y
^' '
1
\\SUSHIN"G.
SEAL
,,s i
,/
~'v s
lNNER NUT -
t'
.,> ^:. ;,s '
'., ;;ax.a,:.,,,.,
,s SHEET
': / mmy,
e~
.3
- y. -
'J NNER s
%,, s,J' - a,/
1 RETAINING ' '
^
a - a'
, ',' jlNNER l ;., /
- j i.:;:
$ RING
? $
T. '
~~^7~~.
r -,
f
- ,f 3 KA0 WOOL:', ;
l[.,A 'J
- 1:
'"' 'i sa,_8 LAN,K_ET,', s
..:.; :-)
.-(
-<.,,,..,s.>.
~m_.,
s
..1 s
s L-
',S/8 IN. WELD,. STUD, ;
',..,, g s
s<
s
<s,
,o'
' ' ^
-.f 9.
- -~,= 2,, x:xws.,
,-,. ',,./;,
+
- ~.
' u,.LET WE LD,s FIL w<, ~.
3 s
's
,^>
g
/s###
A A
- :.....*.7 r.:'.:...... 1.: ~........-...?
- 7..::. :.
"" ~ "
"w
- ..:l:::.:::: STUD WELOING.:;
y::. -
.: :: :{
}/..':l:f!;:;T
.: - {l::) CREATED A HAZ::.
'.*.::l-l.
. ll.2..lY.5:i HAR0 NESS IN ;i
.:::..'.;.. ;.1 ll
- .C*..'l..:(('*
- :../ -l.
l.".;.i. ?J..:::::'.,:.~illll::::..:...:Y.:&:::.:.. kl. ; LINER STEEL
.......,.....;jj:.;;;;
wm.
- 0F Rc35
.:.::...: :::::::::.::... :.,:.}.: :t :i.::' :.. ::::.:.; :: ?:.:.:..:::::::::::.:.l. ::..;. :..::..:, ;;,.:......
Fig. 4.8-1 Typical Thermal Barrier Attachment Stud Fixture
GA TECHNOLOGIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSEL TO PRIMARY COOLANT CHLORIDE CONTAMINATION
~
DOCUMENT NO.
907875 ISSUE NO./LTR. N/C i
L Since the studs are attached to the liner, the integaity of the stud to liner joints is of potential concern for all thermal barrier regions.
The regions of most concern are at the bottom head of the PCRV where chloride bearing water can collect and pool. The top head and upper sidewalls will not be subjected to standing water and, therefore, chicride corrosion is much less likely.
It is evident (see Appendix A) that the stud HAZ hardness is above industry-accepted values for resistance to stress corrosion.
In addition, in
[
view of the large ntaber of studs and the sensitivity of stu'd welds to weld gun
~
operation, it is possible that some HAZs may be harder than Rc35.
4.8.2 Engineering Evaluation 4.8.2.1 Assessment Summar.y I,f the stud weld is postulated to fail, a fillet weld having a larger area than the stud weld will hold the thermal barrier attachment in position.
In the event that both sets of welds were to fail, local voids between the insula-
)
tion and liner are possible.
This could result in liner hot spots over a sizable area which can be detected by the liner cooling water system as deacribed in the FSAR (Section 9.7.3.5.6) and Technical Specifications (SR 5.4.4 and 5.4.5).
w It is concluded that the loss of thermal barrier attachments associated with the bottom head and lower liner barrel section will not adversely affect public health and safety.
4.8.2.2 Failure Consequence The thermal barrier cover plate attachments or posts (Fig. 4.8-1) are threaded onto studs that have been welded to the liner in accordance with GA l
Specification 11-R-38.
The post is torqued to generate a prestress in the weld l
l Page 43 1.
v, GA TECHNOLOGIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUMNT NO.
907875 ISSUE NO./LTR. N/C stud of 37.5 i 2.5 kai.
Yield stress for the stud material is 50 kai. Three 1-in. fillet welds at tihe liner / post interface complete this installation. At the time of installation (and until failure), the stud weld-is in tension, i
while the fillet weld is lightly loaded, he area of the fillet weld is greater than that of the stud weld. Should the stud weld fail, the load on the fillet weld becomes tensile, but will not exceed about 30% of the yield strength.
The presence of the redundant fillet welds substantially decreases the probability of total stud failure.
However, if stud failures do occur,
.there will be no immediate impact even at the bottom head due to the failure of the attachments.
Attachment failures will not be detectable until the thermal barrier-has degraded significantly or has permitted excessive primary coolant flow due to voids between the liner and the insulation. An indication of the possible stud failure can be provided by monitoring the liner cooling tube tesperature as described in the FSAR and Technical Specifications.
The temperature scanner provides for continuous monitoring of the outlet water temperature of each liner cooling tube and is to be checked on a monthly basis 3
during power operation.
Postulating complete loss of an insulation assembly
)
during power operation, a-local liner hot spot of up to 400*F could result.
Such a condition is expected to be tolerable without jeopardizing the primary i
coolant boundary for a period of time not exceeding the interval between cool-ing tube thermocouple checks according to the requirements of the Technical Specification SR 5.4.4.
Failure of the lower liner barrel section thermal barrier attachments
~.'
could result in -an initial slight -pull away of the insulation from the -liner.
Tetal failure of the cover plates is unlikely because of the integrated ~ and
'I staggered nature of ' the insulation,. real sheets, and cover plates. Since each thermal barrier assembly employs eight or. nine fasteners in this region, some individual-failures will be inconsequential. A substantial ntalber of fasteners within individual assemblies would have to fail before ai failure consequence need be addressed.
As with the bottan head, failure of the sidewall thermal l
Iage NN m.
,.m
- ~
GA ' TECHNOLOGIES INC.
t TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED l
TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCO Wrf NO.
907875 ISSUE NO./LTR. N/C barrier can be detected by the required monitoring of the cooling water tubes as apecified in the FSAR and Technical Specifications.
The thermal barrier is designed for seismic loads with the maximum horizontal ground accelerations being 0.15 g for OBE and 0 30 g for SSE.
Due to the relatively low mass of the thermal barrier assembly, a seismic event is not expected to result in component dislodgement. However, in the event of the failure of the studs due to stress corrosion, it was concluded that the added results of a seismic event would not be of any more concern than those previously discussed.
That is, the failure of the studs in the areas of concern would not impact the safe operation of the reactor nor adversely affect public health and safety.
4.8.2 3 Safety Consequence Individual liner cooling tube thermocouple monitoring -is used to detect increases in liner temperature which could be indicative of potential failure of thermal barrier assemblies due to stud failures.
Thus, the failure of the studs on the bottom head and/or the lower liner barrel sidewall will not adversely affect public health and safety.
9 l
l
- rage u5 l
.b a
TABLE 4-1 SUtttARY OF POSSIB1'E CONCERNS AND ENGINEERING FAILURE EVALUATION Adverse Impact on Other Failure Public Health Component Material Issue Considerations Consequence and Safety 1.
Cavity liner-ASTM-A537 Gr0 Perforation-Conservative No failure is expected
.None.
due to Ct, corrosion rates during the life of the enhanced based on seawater
' ratiotor. However, if water experience suggests a failure should occur, corrosion perforation very release of fission rates unlikely.
product in the primary coolant would trigger radiation monitors (Channel 7312--see FSAR Fig. 11.2-1).
2.
Plenum Cast 347 SS Susceptible Condensation No major SCC expected.
- None, elements (control to SCCu probably coeurs column only) only rarelys so Damage to the plente opportunities for elements must be severe stress corrosion to prevent proper flow are limited.
to the fuel elements.
Overheated fuel can cause fission product buildup in the primary coolant system, but fission product buildup is continuously monitored, mm
- SCC - Stress corrosion oracking gS
! e$
s~ ta g
i
s TABLE 4-1 (Continued)
Adverse Impact on Other Failure Public Health Component Material Issue Considerations Consequence and Safety 2.
Plenum The reactor would be elements shut down for replace-(continued) ment of damaged plenum elements.
No off-site radiological release will occur.
3.
Core restraint A286 SS bolts; Susceptible Condensation occurs May impede removal
rarely, so oppor-of restraint' device tunities for SCC and/or plenum elements are limited, and core elements.
Failures could result in core' thermal 1
fluctuations which are detectable. Failed restraint devices can be replaced.
4.1 HPS filters Filter elements Susceptible Filter failure will not
- None, are sintered -
to SCC.
Introduce debris larger 300 series SS.
than about 100p into the PCRV or into the HPS; Will not prevent con-tinued operation of the HPS; will not increase activity releases; and will not impact the operation of any other system.
78
,55
,>u
.J' i
s TABLE 4-1 (Continued)
Adverse Impact'on other Failure Public Health '
Component Material Issue Considerations Consequence and Safety L
i 4.2 Hydrogen Piping and Susceptible All impurities Postulated failure None.
getter valves are 300 to SCC.
including chlorides should not occur. due series SS.
are removed from to absence of reactant pr1 mary coolant impurities. Neverthe --
i before reaching less, such a failure this unit.
would not introduce debris larger than about 100p into the PCRV, and would not impact plant depressurization capa-bilities, since the hydrogen getter unit is bypassed under such conditions.
5.
Instrument 300 series Susceptible Tubing probably Motature monitor system:
None.
1 and sensor SS tubing to SCC.
contains residual lines stresses due to
- 1) Any primary coolant cold formed bends.
leak will be contained in the reactor or
{
within process and moisture penetrations j
B1-B6.
High helium i
purge flow alarm indicates interspace depressurization.
Note: All lines containing primary coolant are 50.5 in.
- AG 1.d.
~
t*
1 k
i n
G TABLE 4-1 (Continued) e, Adverse Impact on.
Other Failure Public Health Component Material Issue Considerations Consequence and Safety 5.
Instrument
- 2) Component failure and sensor causing loss of. flow I
lines will be alarmed in (continued)
-control room and channel trip may be initiated.
Primary coolant pressure and analytical instruments:
- 1) Same as above,
- 2) If break causes 50 pai t
drop below primary coolant pressure, the pressure channel will j
trip.
Circulator auxiliaries:
- 1) Instruments above i
buffer midline see only dry helium.
Those below see helium and water. But if a break occurs, PPS or operator action shuts otroulator down, i
a i
- r4 f
25
i
.s TABLg 4-1 (Continued) s Adverse Impact on Other Failure Public Health Component Material Issue Considerations Consequence
.and Safety 6.
Control rod 321 SS Susceptible Applied loads are The control rod falls
'None.
olevis,, bolt, to SCC.
Very low.
back into core ar.d a washer, nuts slack cable signal is activated.
I Shock absorber 321 SS Susceptible Condensation prob-Shook absorber remains None.
nut to SCC.
ably occurs very attached to rod after l'
rarely, so oppor-complete failure of nut.
tunity for SCC is limited.
I orifice valve 321 SS Susceptible Condensation prob-Oriftoe opens fully None.
connecting bolts to SCC, ably only occurs after failure of bolted l
and nuts very rarely, so connections coolant opportunity for flow through the region SCC is limited.
Increases.
7.
Helium Compressor:
Susceptible Water thrown off in Loss of otroulator None.
nirculator D6AC R 40-44 to SCC.
operations stresses function.
e Blades: 422 SS low when stationary:
R,40
.therefore, stress corrosion risk low.
I High-strength Susceptible Stresses high in Loss of circulator bolts
.(H-11, to SCC.
-some cases one function.
A286)
H-11 bolt failed, another oraoked.
Primary coolant Crack indications ingress into the on some A286 bolts.
steam piping for bolta 90C2101-
..o e,
300-40 only.
ASi i
Io i,
E, n
1
.)
TABLE 4-1 (Continued) l i
Adverse.
Impact on Other Failure Public Health i
Component Matierial Issue Considerations Consequence and Safety 8.
Thermal barrier Liner ASTM-Nelson stud HAZ hardness will Stud weld failure will None.
attachment A537: studs HAZ8 may be vary: oondensation transfer load to post-A108-GR1018 SCC will occur'in these to-liner. fillet welds susceptible.
regions; potential of greater area than for SCC exists, post.
If fillet welds also fall, loads trans-ferred to other studs.
If all studs fail, thermal barrier may t
detach and cause hot spots over a sizable area which can be detected by liner cooling system instru-mentation as described in FSAR and Technical Specirloations.
3 aHAZ = HJat affected zone.
i I
e o
a!
! w" U i
x!
l bI
.~
OA TECHNOL0GIIS INC.
TIT 12:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT CHLORIDE CONTAMINATION DOCUlsIrf No. 907875 ISSUE NO./LTR. N/C 5.
CONCLUSIONS 1
The following are the conclusions from the evaluation study performed on the selected FSV components which are exposed to primary coolant chloride S
contamination:
1.
The component failure modes and failure consequences of credible P
chloride-induced failures were found to be either similar or less severe than those failure cases which have already been considered in the FSAR.
2.
The consequences of the chloride-induced failures identified would not adversely affect public health and safety, which is consistent with similar results in the FSAR.
6.
ACKNOWLEDGMENT Acknowledgment of their contribution in the preparation of this report is given to the following lead personnel and their associates:
H. Jones C. McDonaid C. Perry D. Roberts C. Rodriguez i
L. Swanson J. Wistros i
Pase 52 i
i.'
.Q A TECENOLOGIES INC.
3 TITLE:
EVALUATION OF FORT ST. VRAIN METALLIC COMPONENTS EXPOSED TO PRIMARY C00LAlfr CHLORIDE CONTAMINATION
!~
D0cirram NO.
907875 ISSUE NO./LTR. N/C i
APPENDIX A EFFECTS OF CHLORIDES IN FSV PRIMARY COOLANT i
The presence of significant levels of chloride in the primary coolant of 1
the reactor could have two types 'of potentially adverse effects.
One is the i
possibility of accelerated corrosion of the materials of construction at
)
elevated temperatures found-during normal reactor operation. The second could
~
~
occur in the event of water ingress to the reactor which would create chloride j
bearing aqueous solutions with the potential to cause enhanced general corro-sion, pitting corrosion, stress corrosion or hydrogen embrittlement.
~
A.1 High-Temperature Cverosion i
l
(
To-determine in-reactor corrosion rates due to elevated temperature t
interaction with all primary coolant impurities (including chlorides), data were collected from a plate-out probe, which was reoved from the reactor in f
November 1981.
The probe had been placed in the reactor at start-up.
This l
probe had corrosion coupons on its outside surfaces exposed to the hot primary coolant helium.
The coupons included samples of Alloy 800, Inconel 600, and I
type 304 SS.
These coupons were exposed to temperatures close to the primary coolant outlet gas temperature (up to -1400*F).
Following removal of the probe, the coupons were examined for corrosion and the results showed no 1
i evidence of either damage or the formation of surface chlorides of any kind.
[
Accordingly, it was concluded that the chloride levels present in the reactor primary coolant stream, at least during the period preceding the renoval of the I
plate-out probe, did not cause enhanced elevated taperature corrosion.
Page 53
QA TECBDOLOCIES INC.
TITLES-EVALUATION OF FORT ST. VRAIN ETALLIC COMPONENTS EXPOSED TO PRIMARY COOLANT QiLORIDE CONTAMINATION DocumuT N0. 907875 ISSUE No./LTR. N/C
~
J A.2 Chloride Interactions with Water Ingress.
h possible interaction of chlorides in the primary coolant with water which accide ntally ingresses into the reactor does pose a concern.
The c,
chlorides may be initially distributed fairly uniformly around the primary coolant circuit.
If the water vapor pressure rises to the point where condeensation occurs on chloride contaminated surfaces, the condensing water will dissolve the plated-out chlorides and produce pools of chloride containing water. In addition, as the reactor is dried out following a water ingress, the chloride levels in the remining pools of water will concentrate until finally all the water is removed to leave chloride enriched debris. During subsequent i
water ingress, dissolution of this debris may form relatively chloride-rich ustar.
i The process' of repeatedly forming chloride enriched aqueous regions is most likely to occur in areas either where water ingress occurs directly (i.e.,
accend the helium circulators) or in regions subject to condensation when high prir.ary coolant water levels occur (i.e., the PCRV liner).
However, the process can also occur on additional components during some modes of reactor operation.
l The nature of the chloride bearing aqueous solutions that is formed during the process described above is not entirely clear.
If, as it is currently postulated, the chloride is being introduced as hcl gas from the core, then the initial condensate will be acidic.
However, since acid chloride is very i
)
reactive, it is quite likely that the condensate will be rapidly neutralized by interaction with ocerosion products from the structural materials. However, for the purpose of this raview, it has been assumed that, at least on some occasions, the condensate is acidic.
Pass 518
,__r
,----,~,y,--,.,-m--_,-.
-.---m.--.-.------.m.
,f, GA TECENOLOGIES I E C.
TITLE:
EVALUATION OF FORT ST. VRAIN ETALLIC COMPONENTS EXPOSED TO. PRIMARY COOLANT CHLORIDE CONTAMINATION i -
vvw m m NO. 907875 ISSUE NO./LTR. N/C E
i A.3 Meta 11urrical Impact of Chloride Solutions Chloride-bearing solutions, particularly if acidic, can have a ntaber of adverse impacts on the types of structural materials used in FSV. There are at i
least three potential effects:
- w.
j 1.
Enhanced general corrosion and development of pitting attack, jl 2.
Stress corrosion oracking (SCC), and i
3.
Hydrogen embrittlement.
I Enhanced general corrosion is most likely to be of concern for carbon steel and low alloy steel components. Pitting is a possible concern in carbon j
steels and low alloy steels, stainless steels, and aluminus alloys.
Stress
}
corrosion oracking, in the presence of chlorides, poses a significant concern to components of 300 series stainless steel even at moderate stress ' levels..
4 1
j Stress corrosion cracking is also significant, along with hydrogen embrittlement, to components made of high-strength steels which are significantly stressed.
With respect to the latter phenomenon, stress corrosion / hydrogen 4
embrittlement cracking of high-strength steels occurs in several indratries.
)
One is the petroleus industry where the principal stress corrodents are i
typically warm, moist H S environments which are acid and may also contain chlorides.
In this industry, it is a standard practice to use steels with a l
hardless of less than Ra22 and yield strengths below about 90-100 kai to minimize the possibility of stress corrosion.
In the milder aqueous chloride
)
environments that may occur in FSV, it seems likely that steels below these strength and hardness levels will not suffer stress corrosion.
j Evidence from many industrial uses and laboratory experiments indicates j
that steels with a haedness above about Rc35 or a yield strength above about Page,55 i
. ?_*,
CA TECHNOLOGIES I N C.
TITLE:
EVALUATION OF FORT ST. VRAIN ETALLIC COMPONENTS EXPOSED TO PRIMARY C00LAlff CHLORIDE CONTAMINATION DOCUMNT No.- 907875 ISSUE No./LTR. N/C 150 kai are, if subject to high stress, prone to stress corrosion / hydrogen embrittlement failure.
These general trends are illustrated, for carbon steels, in Fig. A-1. Hardness and yield strength are not the only criteria 3
governing susceptibility. The exact microstructure and chemical composition of I
the steel also has a strong influence on resistance to stress corrosion.
o Steels containing martensite, which have been only lightly tempered, for example, may be much more susceptible to cracking, at the same strength levels, l'
than steels with other carbide strengthened microstructures. Thus, each' steel must be assessed specifically for stress corrosion susceptibility.
Neverthe-4 t
i less, the trends shown in Fig. A-1 provide a useful engineering guide for initial screening to assess where problems may occur.
i In all the above corrosion processes, the presence of dissolved oxygen in the aqueous solutions plays an important role. It is suspected that there may have been circusstances in which water and oxygen were jointly present in the FSV reactor. Moreover, if the condensed solutions are acid, base metal attack i
can proceed even without oxygen.
For the purpose of this review, it has therefore been assumed that oxygen levels are high enough, or pH levels are low enough to permit the corrosion processes to proceed.
This is a conservative assumption.
l 1
l l
i 1
l l
l Page 56 f
n.-...,,...
,,,.n,,
--.n n.,
.._._.,n r
-,,. - - - - - -. - - - - ~,... - - - - - -,, ~.. - - -
l en 907875 N/C
,s.
Pass 57 300
/
APPROXTENSILS STRENGTH
=
200 240 APPROX YlELD STRENGTll T
220 gm E
l g is0 E
I C
l IN 8
140 VERY SUSCEPTI8LE l
l TO SCC
' SUSCEPTIBLE l TOSCC 120 I
l POSSIBLE l
l# "-
SCC 1
RESISTS SUSCEPTlBILITY l
- l l
l a
l l
80 -9 I
l l
I !I l
l l
I l
Il l
I I
20 24 28 32 38 40 44 48 52 58 60 HARDNESS (RC)
Fig. A-1 Appr5ximate Relationships Between Hardness, Strength and Stress i
Corrosion Susceptibility of Carbon and Low Alloy Steels
+-,,
.,,. -