ML20042C771
| ML20042C771 | |
| Person / Time | |
|---|---|
| Issue date: | 06/27/1983 |
| From: | Dircks W NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| To: | |
| Shared Package | |
| ML20037D485 | List: |
| References | |
| FOIA-92-436, TASK-PINC, TASK-SE SECY-83-254, NUDOCS 8307200140 | |
| Download: ML20042C771 (46) | |
Text
IffMMMMSMNNNSSSNSS p#*"*%
ry2 N/
June 27, 1983 SECY-83-254 POLICY ISSUE l
The Commissione(NEGATIVE CONSENT)
For:
rs From:
William J. Dircks Executive Director for Operations
Subject:
THERMAL ANNEALING
Purpose:
Response to Comissioners' request (M830421) dated April 28,1983, for a SECY paper addressing thermal annealing and the need for additional revision to Appendix G, 10 CFR Part 50.
Issue:
Should Appendix G be amended to remove or revise the requirements that concern annealing of the reactor vessel?
Background:
There are two references to annealing in Appendix G.
Paragraph IV.B. states:
1 Reactor vessels for which the predicted value of upper-shelf energy at end of life is below 50 ft-lb or for which the predicted value of adjusted reference temperature at end of life exceeds 200*F (93*C) must be designed to permit a thermal annealing i
treatment at a sufficiently high temperature to recover material toughness prope, ties of ferritic materials of the reactor vessel beltline.
Paragraph V.D. states:
If the procedures of Section V.C. of this appendix do not indicate the existence of an equivalent safety margin, the reactor vessel beltline may, subject to the approval of the Director, Office of Nuclear Reactor Regulation, be given a themal annealing treatment to recover the fracture toughness of the material. The degree of recovery is to be measured by testing additional specimens that have been withdrawn from the surveillance program capsules and that have been annealed under the
Contact:
P. N. Randall 443-5903 8 3 8.
?j ghoogo)
Gl A
m
- u The Comissioners-2-
1 c
same time-at-temperature conditions'.as those given the beltline-material. - The results stogether with; the results of other pertinent annealing-effects a
studies,'are to provide the basis for. establishing:
the: adjusted reference temperature.and upper-shelf energy after annealing. The reactor vessel may1
{
continue to be operated for that service, period j
within which the predicted fracture toughness of l
the beltline region materials; satisfies-the' requirements.
of Section IV.A.'of this appendix.usingcthe j
values of adjusted. reference temperature'andL J
upper-shelf energy'that include the effects ~ ofL j
annealing 'and subsequent irradiation.;
1 These paragraphsido.not' require annealing of reactor. -
vessels. Paragraph V.D. offers annealing as. a fallback j
1 position, which may be proposed.by the license'e.when:
continued. operation can no L1onger becjustified because-1 the reactor vessel has become embrittled to-the extent:
that it cannot be shownlto meet the requirements of'-
i Appendix G. ; Paragraph IV.B. provides' a; warning to licensees.that annealing may need to be considered at:
the design stage,~and it provides them the basis for
.i knowing if the warning applies _to them.-
.]
Three options will be' discussed:
_1.
-Retain the paragraphs on annealing unchanged, 2.
Remove paragraph IV.B or V.D. Lor both, and 3.
Revise the 20D*F-criterion for design to --
permit annealing. given in oaraaraphTIV.B.
Option 1 -- Retain the paragraphs on. annealing ' unchanged.
At present, annealing is.the only known way.to remove j
. neutron embrittlement from a reactor vessel... Other actions now being taken ' slow the rate.of embrittlement j
and refine our: calculations of the margin of safety:
against' brittle' fracture.1but none of: these actions reverse the embrittlement effect or improve the'present day margins for existing plants. Based on~ the: PTS i
report, in 1984 there will be four or five reactors that have RTNDT values above 250*F and about twenty reactors wi1T have RTNDT values.above-200*F:in either ~
an axial or;a circumfarential: weld. ~ Flux reduction i
.....a.
2 h q.
p k
- The Commissioners t q
procedures are being' implemented to= slow-the rate'of embrittlement ~ in al number of reactors. 2More material property measurements-are being made, Land. batter.
trend curves' are being developed to remove some ofs i
J
-the conservatism in the estimates of' embrittlement
~
'as a function of neutron fluence and chemicalicontent of the steel.. Fracture analysis: methods are:being refined'to remove' unnecessary conservatism inithe-3 determination of how much embrittlement is: acceptable.
- But,' annealing is the.only known way_ to remove embrittlement
'and 'actually: regain safety _ margins,_ short of replacing -
- j the' reactor vessel.
y A future need to remove embrittlement from operating -
Lt reactor vessels certainly may arise. LThe following 4
three~ examples ~ outline situations in which ' annealing-might-be the only~ alternative; to~ shutdown or granting i
an_ exemption to the' regulations: to. permit _ operation-at reduced safety' margins. "Although their individual:
i,
- probability of occurrence is low, when -taken.as a group (and_ knowing that this may be anLincomplete.
list), these situations provide-a' basis;for predicting
.l that annealing may someday be'needed.
~
\\
j 1.
Severe pressurized ^ thermal. shock events could j
a occur with sufficient frequency.and severity to -
l cause an upward revision of the' estimates'of H
j probability of occurrence.: The consequent.
j reduction in::the screening criterion could i
impact several: plants..
l 2.
- Advanced analysis and experimentation.could show that radiation damage'in the form of low-
. upper; shelf energy of the beltline material-
- poses a threat;to:some reactor. vessels by-a -
mechanism known as'" low-energy ductile" tearing."'
(Radiation damage takesttwo forms:
it shifts i
the transition temperature upward and it lowers the maximum fracture toughness that can be developed at' temperatures above the transition.
_In Charpy test-terminology,~ it lowers the'
upper.- s hel f." ) Appendix _G requires a minimum
.'of.50 ft-lb of Charpy upper' shelf ener?
r throughout' plant ;11.fe.
If radiation 1ca ::es the upper shelf to drop below 50 ;ft-lb,;a chr:n of g.
.)
m
~
r 4
_w,w_
,, e m
m c
e,.,
W s 3
- 1 j.o i
- Tha Comissioners
' 4
. activities : including extra inservice : inspection,-
tests'of: archival.materialiand advanced analysis-4
' is set:in motion. Guidelines.for the analysis are e
' now in place. in-NUREG-0744,. as. the resolution of-
.the Unresolved Safety-Issue A-11..1 Several plants-
-are to undergo the analysis in'the'near. future.'
l s
L Meanwhile, this' subject'is the focus of fracture:
R i --
mechanics. research,:the results Lof which"couldi j
i:
cause revision;of our procedures.; zIt could.be:
1 concluded for certain plants that fallingibelow 50' 4
ft-lb. upper shelf energy 1 reduces the safety margin..
j l
below:that required'by Appendix.G.-
3.
Power shortages could occur?in the futurei even as:
certain' older plants approach _a critical, state. of' embri ttl ement. Thi s L i s : pa rti cul a rly l 11 kely 1 for.
1 plants that are. operated beyond;the, original...
J
-expiration date ofitheir license. tThere could-be
. requests to grantiexemptionsito permit operation,-
j-despite reduced safety margins,Ljustified by:the
'argumentLthat the power was needed.for public r
l health andLsafety.1 and in wartime; for the' common.
4 defense and. security.
ij' The-foregoing-examples =are meant to show that'public; welfare.and safety will be better-served 11nTthe future -
by having an alternative to.'the-options of plant shutdown?
or operation atisafety margins;1ess:than we normally..
required. LAnnealingfis such an' alternative, provided it really'does remove' embrittlement and provided it.
4
'does' not induceLsome hidden hazard that detracts from plant safety.
1 1
)
Appendix M of.the PTS. report, SECY 82-465,/ describes-the state of the art of annealingk There.is an experimental
. basis for choosing the annealing temperature and timei F
(850'F for 1 week). -: The benefits. to be-gained are l
clear - about 80 percent recovery of. fracture toughness 1
properties. The rate of reembrittlementiafter annealing-'
can be expected to be much-less than the initial 1 rate -
of '. embrittlement, possibly as low as the rate just:
preceding annealing. More testing is-needed to confirm.
l these~ estimates,;but there is ;no questionL that annealing.
u removes ~ neutron embrittlement.
I
- ~
'I U
L a;
i,5-
.~.,
u4 l
The Comissioners 5
J The' subject of engineeringofeasibility and hidden hazards deserves special-attention,? because Commissioner Ahearne (in his coments on SECY 83-80); questioned why we should keep the annealing requirements-in the regulation.
when.certain factors are so poorly understood that the NRC is funding researchito provide the answers. - He was:
~ referring to a memorandum (Enclosure 1) which recomended -
1 i
against-a demonstration annealiat an existing plant but
- did recomend a smaller research program on separate'-
pieces of the problem..
'Coments of this ' kind apply to most of our research projects. The arguments used to ' justify NRC-sponsored -
l research can be turned against the NRC regulator.1 saying, "How can you. regulate before you. understand the
. problem?" e A. specific response in the-case of annealing
-is that:the evidence at hand indicates annealing is-. ~
feasible and free of results detrimental to safety, but-we recognize annealing to be a difficult; engineering' job,'one that a utility should not be permitted to undertake without thorough understanding of a number,of.
factors. Moreover, in order to review their' submittal properly, we must have understanding ourselves.
The prime' engineering difficulty concerns:the risk-of-distortion rf the reactor vessel. This is a> function of the _degrae of control of the-temperature distribution-in the vessel during annealing.1The initial studies-funded by EPRI* concluded:. Dry in-situ thermalJannealing of an embrittled reactor. vessel: using an annealing -
temperature of.850*FL(454*C):for a-period'of 168 hours0.00194 days <br />0.0467 hours <br />2.777778e-4 weeks <br />6.3924e-5 months <br /> is feasible and can be perfomed at most of the' existing.
nuclear power plants without jeopardizing-plant integrity."'
A discussion of the experience posa ssed by-one vessel J
fabricator,-Combustion Engineeri.. Inc.,-relative to:
local. heat-treatment of a vessels <.ontained in Enclosure' 2, is reassuring evidence that vessel fabricators have considerable experience'with localized heat treatment -
of vessels. - Quot*ng from part of that report: " Reactor vessels have several critical dimensions that relate-to
-l
~
O " Feasibility of and Methodology for Thermal Annealing of an 'Embrittled.
Reactor Vessel," prepared by Westinghouse Electric Corporation for
' EPRI, NP-2712, Volume'2, November 1982, T. R. Mager, Principal Investigator.
j r
b; w
+
.,i y
..,ev
. - +.
ey a 4,
-ep-*
-o
,w m,,-
..my.
w
<,.,.ae,w. +- ee -, e, w ef.,,
g w.v * :, yvy y.,p -ny-g o w-e,, --
e
ae J
The Commissioners
-6 t
core alignment, th'e most critical one being the sealing surface between the core barrel and the hot leg nozzles. This-dimension is machined to + 0.010 inches. Other locations include the seismic ~ alignment.
keys and core stop lugs at the lower end, and the alignment between the control rods and the core.'.
Standard CE fabrication procedure' calls for these surfaces to be finish machined while the vessel is
- still in two sections and. prior' to the final closure weld. The two sections are individually post weld heat treated (PWHT) at 1150*F in a furnace before finish machining. A field PWHT is made on.the final closure weld with the completed vessel in.the vertical-
.4 position sitting on the support pads. This heat treatment is-done from the_outside with' gas heaters that extend for approximately 3-1/2 feet on either side'of the weld and with the inside surface covered eith 4 inches of mineral insulation.
The vessel weld area is heated at 100*F/hr to approximately i
1150*F..Despite this heating, the' nozzle region-remains at approximately 80*F even though no specific cooling is applied. CE has-observed.no dimensional distortion problems due to this heat treatment."
3 In.sumary, alternative 1 presents the case' for retention of Paragraphs IV.B. and V.D. in Appendix G.
Annealing effectively removes embrittlement when performed-at recommended temperatures and times.
There are a number of circumstances that could lead to a need for annealing as an alternative to plant shutdown or operation at reduced margin. - Experience indicates that annealing is feasible, but.the engineering difficulties need to be carefully evaluated. This requires a research effort partly funded by the NRC, but annealing remains an acceptable solution to embrittlement problems.
Option 2 -- Remove paragraph IV.B. or. V.D. or both, i
The first argument for removal of the paragraphs on annealing.is that we might encounter some hidden difficulty or saf ety problem in the ongoing research that would invalidate these paragraphs. As discussed under Alternat've 1. we do not-believe that this is a likely result of the research work.
i aw-E e m e
w
.-----r e
w
~
1 j
The~ Comissioners
'7 t
j-l There has been criticism that the requirement to-
" design to permit annealing": receives small consideration-in the FSARs and-a. cursory; review byLthe staff.
To illustrate the meaning of the requirement, the~following r
list provides examples of the considerations which the-Appendix _G requirement was intended to bring into the.
review process.
j 1.
Configuration-of the area around the1 reactor vessel:
a) clearances for heating < fixtures',' b)l working: space'above the, reactor vessel, and c) storage space for; fuel and internals removed from -
the reactor.
2.
Accomodation.of, vessel expansion:on its. supports.
with' consideration for the large-diameter piping attached to the vessel.-
1 3..
~ Power source.for. heating. -
i 4.
Radiological hazards-to workmen.-
5.
' Adequate surveillance ' program to monitor the -
effects'of annealing and reirradiation'.
i An example of an FSAR response 'to this requirement is.
J the'following -
"There'are.no special design _ features which would' prohibit the in-situ annealing of'the vessel.
If the unlikely need for an annealing operation was required to restore the properties of the: vessel' material opposite'the reactor core because of:
neutron irradiation damage, a metal' temperature-greater than 650*F for a period of.168: hours maximum would be. applied. -' Various modes of. heating ~
may be used depending on'the~ temperature.--
1 The reactor vessel materials surveillance program is adequate to accommodate the annealing of the; i
reactor vessel..- ' Sufficient' specimens are available-
)
to evaluate the-effects of the annealing treatment."
i i
The Commissioners 8
It appears that there have been few,'if any, actual design features added to accomodate annealing. Other maintenance, inspection and repair operations place similar requirements on clearances and radiological hazards. Vessel expansion at the nozzles is not expected to exceed that experienced in operation at the design temperature (650*F). The extra features of the surveillance program are referred to briefly in ASTM E 185, which is referenced in Appendix H.
Thus, it can be concluded that this paragraph has little impact on design of a plant. On the other hand, it does not cause unnecessary cost or the incorporation of features that might never be used. As stated earlier, the purpose of Paragraph IV.B. is to provide a warning to licensees that annealing may need to be considered at the design stage, and it provides them the basis for knowing if the warning applies to them.
One argument for removal of a11' references to annealing from Appendix G is based on the statement that annealing is an economic issue, not a safety issue.
Our response is that a feasible annealing procedure can keen embrittlement from becoming a safety issue. Moreover, if the Comission expects that annealing will be regarded as an acceptable alternative, plant owners and designers should be forewarned of conditions where annealing might be required and should be offered the annealing option in the regulation.
Finally, to remove Paragraphs IV.B. and V.D. now would rignal a change in policy -- implying that annealing is no longer an acceptable solution to embrittlement problems -- which is not the case.
Option 3 -- Revise the 200*F criterion for design to permit annealing.
Aside f rom the general question of whether references to annealing belong in Appendix G, there have been specific wstions about the numbers in paragraph I
IV.B.; i.e., is an end-of-life RTNDT of 200*F the proper limit beyond which the reactor vessel "must be designed to permit a thermal annealing treatment?"
~-.
- 3
- ' i l
i l'l'
- The Comissioners-9-
f i
The figure, 200*F, was chosen originally on the basis NDT of 200*F_.was L
of. pressure-temperature limits. An RT i
judged to leave ~ a sufficient' temperature interval between the-P-T -limit curve.and the. saturation curvel to permit the. operator to start up;and cool down the:p1_ ant without violating either the fracture prevention: requirements.-
on-the low-temperatureLside of the interval. or core
(
protection' requirements:on the high-temperature side'. -
of -200*F is well_ below the temperature Actually, an RT that would restNt operator action significantly.'
This was done deliberately because 200*F is a projected.
there has to [e allowance-for error. in the prediction.ND, made at;the 1
. value of RT The proposed rule concerning pressurized thermal shock.
(PTS) sets:forth as a: screening criterion for susceptibililty-to PTS an RTNDT'at the' surface of 270*F.for_ plates..
forgings and axial welds and.300*F for circumferential, 1
' welds.
For: plants' that will exceed the screening;
- criteria before end of life, annealing:is one optionL t
among several'that may.be considered. : This raises the'
' question:- Is the 200*F limit;given in paragraph IV.B.
of Appendix.G still the proper value'.in light of the
~
l-PTS screening criteria? - We believe the answer is yes..
L The 200*F limit applies at the 1/4.T ~ position and--
l corresponds to'about 235'F at the surface.:whichiis 1
sufficiently below the~ PTS ' screening criteria..to allow for error in the prediction.
Summary:-
The references to annealing in Appendix G have;been.
discussed with regard to their relationship to current research on annealing.
The Commission has asked:- Is- ~
thisEa consistent relationship?L Our answer is yes, to -
the best'of our knowledge..and we recommend that no changes be'made in Appendix G.at this time. Nevertheless,'
if the research reveals some reason why annealing.
i p
1
- u --.
.. a
-....~.a. a--
a
The Commissioners 10 is not practical or some effect that. could. reduce _ plant safety, we will take the necessary steps to propose an amendment of Appendix G.
s William J.' Dircks Executive' Director for Operations -
Enclosures:
1.
Memorandum: iMinogue to Denton,
" Proposed Experimental. Program-to Support Pressurized Thermal Shock. Issue " February 25, 1983 2.
Memorandum: Taboada to Serpan,
" Trip Report to Combustion Engineering," June 1,.1983 SECY NOTE:
In the absence of instructions' to-theecontrary,.
SECY will notify the; staff on Tuesday, July 12, 1983 that the Commission, by negative. consent,. assents-to the action proposed.in'this paper.
DISTRIBUTION:
Commissioners OGC OPE OCA' OIA OPA REGIONAL OFFICES I
l-
n..
....n.
~
.n.~.- ~ ~ -,
.-...-n.+ - - - - -
.~-ng-~,.
-.----.---~<--.~n--,-~.-....~--
. v.
e e
3 F
b a
b ENCLOSURE.1-I t
I E
I e
4
"~T"W+*fMt"we3 p, q:. 3, g,,
NN#W'WNeWee e pm.mp
hC/o.sure A~
o
,I i
February 25, 1983 MEMORANDUli FOR: Harold R. Denton, Director Office of Nuclear Reactor Regulation FR0!!:
Robert B. Minogue, Director Office of nuclear Regulatory Research
SUBJECT:
PROPOSED EXPERIMENTAL PROGRAM TO SUPPORT THE PRESSURIZED THERMAL SHOCK ISSUE
, RR-82-03)' JUNE 15,1982
(
Your nemorandum of June 15, 1982, requested that RES make a study of the feasibility of conducting a demonstration of "in situ thermal annealing" on an irradiation-embrittled reactor pressure vessel to restore fracture toughness properties. We' have completed such a study based on the published information available on the subject and a survey nade for us by INEL of reactor vessel manufacturers, architect-engineering companies, major field heat treating companies, EPRI, and others. He have reviewed this natter with the NRR staff on several occasions' and have reflected the outcome of those discussions in this document. As your letter suggested, the study reviewed possible candidate vessels, gave careful consideration to the value and general applicability of information to be gained and evaluated the cost effectiveness of such a program.
Basic Concerns The basic concerns of a demonstration anneal fall into one of two ca tegories. One, primarily related to safety and economics, deals with engineering and systems problems.
These include the practical problems of heating and instrumenting the vessel and the possible detricental effect of such heating on the rest of the system.
Concerns include, for exanple, (1) rotention of dicensional accuracy to avoid unacceptable i
distortions that might interfere with replacanent of reactor head and internals or with control rod operation, (2) avoidance of excessive thercal stresses and strains due to tenperature differentials that night lead to localized damage of associated systeas or crack growth in overstressed areas; (3) protection against thercal degradation of such comronents as the concrete containment, supports or instrumentation; and (4) controls against radioactive contamination and excessive personnel exposure. As you night surmise, these effects are to a great cxtent plant specific and difficult to extrapolate.
The other area of concern, primarily.
related to safety, deals with material property changes due to irradiation effects and includes the need for a demonstration or other neans of assurance that natcrial toughness properties have been restored and that the vessel surveillance progran would continue to be valid for conitoring irradiation enbrittlement during subsequent reactor operation.
<8 7
- C n >
' " ~ '
- sasn >
!!arold R. Denton 2
FEB 2 51983 Candidate Vessels Five vessels were identified as possible candidates for an anncaling demonstration, including those at Indian Point-1, Shippingport, KRB-A, Humboldt Day, and BR-3.
All have been irradiation embrittled to some degree. Also, since all but the BR-3 have been shut down pcmanently, significant r.1odifications of the plant and destructive examination of the vessel would be possible. However, these plants have significant differences from the large PURs of concern in those features that would inpact an annealing demonstration including the vessel caterials, irradiation tmperature, nozzle and piping configurations, and ex-vessel cavity and support configurations..
Feasibility Based on the study requested, we have concluded that a general demonstration of in situ thermal annealing is feasible.
The processing, heating equipment and instrumentation that would be applied to anncaling a reactor vessel bcitline region are well developed and considered state of the art by coacercial heat treaters.. However, such an operation requires a significant engineering effort to establish logistics and to detemine the appropriate tecperature profiles, heat-up and cooldown rates, and holding conditions needed to control diniensions and avoid distortion and overstressing during the heat treatment operation.
The determination of these temperature conditions should be based on the limiting stresses and strains permitted in the vessel (and its attachments) during the annealing operation, and would probably require a sophisticated and costly elastic-plastic analysis which considers piping reactions, peak stress locations and levels, dissimilar metal effects, reactions at safe ends, control rod drives, and bolting.
Our conclusion that it is feasible to perfonn an annealing denonstration without causing excessive distortions or localized damage due to'overstressing, provided the appropriate heating parameters are established, is consistent with the Uestinghcuse report for EPRI on the subject referenced in the enciesure as well es by infomation developed by IMEl. in its survey.
INEl. reported several cases of successful localized post weld heat treatnent of courercial nuclear reactor vessels in the t.eltline region at tcaperatures considerably higher than those proposed for a denonstration anneal.
In these cases, elastic-plastic analyses accurately predicted the vessel reactions.
- " M Daf t k
FEB 2 51983 Harcld R. Denton 3
for the purposes of detennining feasibility, we have concluded that the materials aspects of an anneal are sufficiently well understood.
The effects of time and temperature on the metallurgical properties of irradiated pressure vessel steels have been demonstrated ir. enough detail to predict that annealing a vessel by one of the acthods proposed will restore a major portion of the fracture toughness properties.
However, the exact arount of recovery.and the specific characteristics of rcembrittlement are raaterial and plant specific and cannot be predicted l
to the required level of precision at this time without additional expericental testing.
It is reemphasized that a denonstration anneal would produce specific infomation for only one steel composition and set of annealing conditions,,
P.esults Expected from a Demonstration Experiment
\\
A successful in situ thermal anneal experiment using one of the candi-date pr'ototype reactor vessels available would demonstrate proof of the general kbility to perfom such an anneal.
It would also verify engineer-ing design features such as the successful operation of the equipment and instrupentation and the general ability to predict and control temperature and loading conditions and thereby the contr.ol of vessel growth and dtst.ortions, piping, reactions, themal gradients and dirensional stability. Further, experience would be gained in limiting personr.el /
exposure, centrolling contanination, decontamination of heating equipment as well as the nore routine planning, scheduling and perfoming the individual steps of the anneal.
Additional data on annealing effects on irradiation embrittled material would be generated if destructive exenination of the vessel is permitted.
Inspection operations to be applied subsequent to the annealing would be demonstrated.
Because the candidate reactor vessels are not prototypic, those detailed engineering procedures, stress and themal analysis and irradiation effects data developed would not be directly applicable to the PWRs of concern, and in that sense, the experinental anneal could not be considered a ca1plete denons tration. Geometric features that control themal stresses and flcu patterns are significantly different between the two groups of vessels so that the potential distortions or excessive loadings that right lead to subsequent assenbly problens would not be simulated.
This is especially true of the nozzle configurations where the najor stress concentrations occur. The ability to control the temperature of concrete supports to a safe level to avoid degradaticn of the concrete would not te demonstrated because the available vessels are designed without concrete supports in the heated region. The size and shape of the evailable vessels are sufficiently different that the heat treatrent equipent design would not apply directly, liniting the ability to extrapolate operational experience, control of contaraination and other engineering details.
Also, although the material properties would be applicable to the cata bank of properties en irradiation effects, it would be a single point of a Lody of data affected by naterial cc position and fluence levels which vary with each reactor vessel.
.../'"
.., o
FEB 2 5 553 Hhrold R. Penton 4
Cost An anncaling demonstration using one of the available reactors is expected to cost from $30M to $50!! including the cost of developing procedures, perfoming analyses, procuring equipoent and instrumentation, perfoming the annealing operation, and any subsequent cicanup and examinations required. Planning of this operation is expected to take 1 year and perfoming all of the engineering and operations 1 to 2 years. This cost does not include the existing materials irradiation program which is required to detemine parametric effects of chenistry and heat treatment on the recovery of toughness properties.
Related NRC Research
- w, RES has completed or has currently in procrass research programs of testing and analysis which will contribute to the data base on the subject and can fem the basis for establishing HRC positions on the safety-related aspects of in situ themal annealing for specific plants.
These programs are developing the following infomation in detail:
fracture rechanics crack initiation / arrest methodology, irradiation embrittlenent and recovery trends, neutron dosinetry and fluence prediction methodology, code and regulatory criteria that would have to be satisfied for conduct of annealing and requalification for service, a listing of systers effects considerations with the results of analyses developed by RES and others, evaluation of material properties and annealing characterisitics of retired-frca-service reactor vessels, and material recovery and engineering and 1cgistics data available from such nuclear plants as have actually been annealed inservice.
Conclusion Finally, although we have concluded that a deconstration'of in situ themal annealing is feasible, we believe it is not advisable to undertake such an operaticn at this time since it would necessitate using a nonprototypical reactor which would have limited applicability as well as very' high costs. As we noted, none of the reactors presently available for a deconstration anneal have the same vessel naterial/ irradiation tmperature, nozzle-piping-stcan generator configuration, nor ex-vessel cavity configurations as the large PURs which are potential annealing candidates. Thus, a demonstration anneal on one of the available plants would produce only a limited body of infomation that could be directly transferred to the engineering analysis for ruch larger, core nodern plants. Of course, a wealth of data could always be gained frca the annealing of any plant; however, the questions to be resolved are:
Is the infomation gained of direct applicability to the next annealing undertaken; i.e., is it generically applicable or can it be directly extrapolated to a large PUR that is being considered for annealing? Is such infomation from a demonstration necessary or just helpful? Does the high cost justify the value of the results?
... ~..........
... ~. ~ ~.
. ~ ~ ~ ~.
~. - ~..
. ~. ~.
............ ~
~ ~.. ~.....
~.... ~ ~.
. ~.....
\\
~~
0 N
gg 2 b Harold R. Denton 5
j Rather than evaluate the effectiveness of annealing through an integral demonstration experiment, we believe that many of-the safety related
~
questions can be resolved nore efficiently through' studies,~ experiments:
and tests specifically designed-for this purpose, and as such, c'an be more applicable to a concercial vessel anneal.
Enclosure.l' presents a '..
list of rescarch activities that:are in; progress or-should be ieplccented to establish feasibility and safety requirenentsifor an in situ anneal.
Instead of a deir.onstration, we, therefore,- propose to continuo separate studies on-the subject to detemin'e the needed-infomation in a nore efficient canner. Ue propose to continue our program of experimental work mainly on neta11urg,1 cal aspects, to_ include expert consultations as appropriate, and-to define a program of analytical / experimental studies that would emphasize aspects of. systems engineering and logistic problems.
Ve would consider participation in a-devonstration annealing conducted by others at a future date if we detemined that specific data and results of particular value to liRC could be obtained by our participation, and that such infomation was-not to be obtained in the ' original-plan.of' the deronstration; the exact level of effort and_ financial contribution from HRC would have to be determined at that future time. The NRC progran would be designed to establish-the proper background and benchaarks to pemit accurate extrapolations. He believe that this approach is the nost ' effective way to.obtain the. appropriate infomation necessary to assure that the fracture toughness properties ~ of embrittled-reactor vessels have been restored without unduly danaging the reactor.
A core detailed discussion of the. background. infomatfon leading to these conclusions is given in enclosures 2 and 3.-
.D ert. flinogue, Director Office of Huclear Regulatory Research inclosures:
Distribution 1.
program Plan for P.cscarch RES Reading Chron Activities 2.
Evaluation of a Ceactor Subj ect Branch rf l
Vessel Annealing Denonstration, ATaboada rf l-l'EBR 83-1, February 1983 3.
EGt.C Coport, " Evaluation of CSerpan GArlotto j
a Ceactor Pressure Vessel t.Shao j
I.nnealing Demonstration,a October 1902 Dross RMinogue See attached sheet for concurrences.
MEB R :DET
- DET:MEBR:C*
DET:DD*
DET:DIR*
RES:DD
......................................m-~~~~~~
~ ~ ~ ~ ~ ~ ~ -
~ ~ ~ - ~ ~ " -
"4..AJ,a boa g a /,mp,,,,g,Z,{,ey,p a,n,,,,,,,,y,S ha,o_ _ _ _ ;, G,g_gp_ _ _h s.s _ _"-_ "W og e 4
~~/24/83 b..
e e n ). 1/17/83
.... /.10/.83.--.1/. ~ / 83 1/14/83 1
2
~~"~~~
.... ~
~
~~~~~
. - - ~ ~ ~ ~ ~
TO ESTABLISH FEASIBILITY AND. SAFETY: REQUIREMENTS TOR IN: SITU. ANNEALING.
t
'Research: Activities NRC Programs EPRI/ Industry Programs-Coments :
.7: METALLURGICAL EFFECTS
. i
- 1'.c JEstablish. toughness requirements for irrad-HSST Program In Progress
[
rated vessel-j 2.
Establish ' effects of.. annealing on irradiation ENSA/ MEA Program Westinghouse (EPRI)
.In Progress--
embrittlement andfsensitivity to reembrittle--
ment.
Coordinat' ion progra.
- 3. = Establish-procedures for. prediction of neutron HEDL NBS
- ASTM /EURATON
~ progress-
- fluence OP.NLL
EPRI - Small1Compi. - Program
'4'.
Validate laboratory results using operating or KRB-A-In' negotiation l retired power reactors
--- B R-3 JIn. negotiation-SM-1A" Completed...
~ i Shippingport; IP-1l ho act %n at this t.
I/ ENGINEERING FEASIBILITY-
'i 4
ll'.11dentify engineering and logistic problems.
EG&G EWestinghouse (EPRI)I
-NRC program-in pror j
s*
EPRI' program compig
.2.
Detennine details of heat treatment operations EG&G-Westinghouse (EPRI)~
~
'NRC programjin pror
^
'EPRI programLcompiE.
03.. ' Establish controls to avoid distortion 'and.
EG&G'
?In: progress overstressingiof vesseliduring annealing
- cperation :-l detailed. analysis ~
+
r F f4.3 ' Possible heat ireatment' demonstration of?
Noactionat'thislt(
i
! surplus / scrap vesse1LwithLrestraints{to simu-
~
Llate system loads
~
~
=
1
~
15'.
Cooperate 11n~BR-3! Annealing decision to anneal'y 1
1EG&G e
Being ~ developed 7perL iII; JCRITERIA AND: STANDARD REQUIREMENTS
~
.l.^ Criteria for new toughness detenninations?
- NRR-RES
.To=be deve1oped-1 L
.for annealed material ~
In progress l
- 12. Criteria forisurveillance program. forx--
EG&G'-
. ASTM-j reirradiated vessel:
- 13. : Identify lrequalification requirements EG&G:
ASME, ASTM:
.In:progressi U
'-4.
Identify augmented inservice inspection-EG&Gc ASME'-
- In progress 1 requirements
.e
. -~ ~.
- -,...... - -... - _. - - ~.. ~. -,..
w L n 13-1 february 1983 EVALUATION OF A REACTOR YESSEL ANNEALING DEMONSTRATION q
C. Z. Serpan, Jr.; A. Taboada; W. L. Server * -
i
~
I.
Definition and Criteria for an Annealing Demonstration l
A demonstration of annealing.is.taken to'mean that an irradiated reactor L
vessel has been subjected-to a given level of heating for a specific l
time, that the mechanical properties of the vessel steel have been measured before and af ter the annealing and correlated with results of measurements made on surveillance capsule specimens (that have~been irradiated either in the reactor itself or' in test reactors and subseq)uently-subjected to the same time at temperature of' annealing heat treatment,
a that a suitably.high: degree of ' recovery of material properties has been l
achieved'.. that following the heat treatment the reactor has been completely l
reassembled and is capable of resuming power-producing service and that the reactor is no less safe following the annealing than.it was before.
l the annealing as a result of the activities involved with the annealing operation (Ref.1).
The criteria developed for evaluation of a successful annealing demonstration fall into the two areas of !!aterials Evaluation and'of Engineering and Logistics Evaluation. They are discussed in greater ' detail below.-
~
Materials Evaluations Verification of Degree of Recovery-of Itaterial Properties.
In order to prove that the fracture toughness properties.of the reactor vessel:were actually recovered by annealing, and that the vessel steel was in an acceptable toughness condition.
verification would be required.
This can only be done by making fracture toughness measurements of steel specimens machined from samples removed out of the vessel wall both before and after annealing.
This is a very severe requirement and is one that could only be done on a reactor that was not going back into service.' To be established would be the RT NDT and upper shelf enenjy of the irradiated vessel steel as measured by Charpy-V (or compact tension fracture mechanics) specimens both before and after the annealing to demonstrate the degree of recovery and establish the absolute value of the RT and upper shelf energy upon restart of the plant.
NDT Correlation of Surveillance da ta to Vessel Steel Condition.
Indirect means of verification, such as af forded through surveillance capsules (Ref. 2), should be validated in connection with the requirement noted above on testing steel removed from the reactor vessel. The main objective here would be to.
detennine how closely the surveillance measurement of annealing recovery matched.the measurement made from the vessel itself.-
This would be of high value to any future licensing evaluation j
because a utility contemplating annealing could only infer the embrittlement condition of the reactor vessel through use of i
the surveillance specimen results and the existing data base.
j i
INEL
o 2
Engineering and 1.ogistics - This area describes factors that relate to-J the piysical accomplishment of an annealing operation;~ that is, heating t
a reactor vessel and bringing-the plant safely back on line (Ref. 3).
Application of Heat to a Vessel. 'This is the most' critical.
aspect of the engineering and logistical; area because. heating must be directly applied to the vesselito gain-the benefit of j
recovered properties.c To accomplish this aspect, a heating system or device must be introduced into the reactor. containment, the heat must be applied to the vessel, unifonnly, for the correct length of time and at-the correct heat-up and cool '.
down rates, and.the heating system or device must be removed.-
For purposes of a demonstration, verification would be. required that the specified heating was~ actually sustained!by the--
vessel through its entire thickness for the specified length; of time Retention of Dimensional' Accuracy.
This aspect of annealing is almost as critical as the. application' of heat, because ~1f.
the plant loses dimensional accuracy ~as a-result of. annealing and cannot be put back together for safe, continuing operation, then the annealing operation is pointless and unnecessary.
This aspect refers to such items as warpage off the head-flange,
~
or distortion of the piping from the vessel to the= steam -
generators, continued free motion of control. rods through vessel penetration and guide tubes, and movement or distortion ~
~
of the nozzle support pads or vessel support columns because of the differential expansion resulting from the heating.
If unacceptable levels of-warpage:or distortion occurred during 3
annealing, they would show up. as the vessel. internals were..
being ^ installed, the' head was being replaced and control rod motion being checked prior' to restart. -
Stresses and " Hidden" Damag'e.
The stresses referred to here arise because heat is applied.only locally to the vessel beltline. Thus, a temperature differential of several hundred degrees F can develop between the beltline and the lower head 1
as well as nozzle and upper flange areas.
It is'not thought that these th'emal stresses 'will be too severe, but this aspect must carefully be reviewed to-predict the stresses, especially in areas such as nozzle corners; for a demonstration, the stresses must be verified through strain gage measurements.
The higher-than-nomal -level of stresses. which are not uniformly distributed, have the potential for causing crack growth, again at locations such as nozzle corners where the stresses J
- ^
even in nonnal service are higher than1 average... Thus, carefu1~
and detailed thermal and stress analyses must!be made to i
establish stress levels and. determine regions of high stresses as input to further evaluations 'of -the possibility of localized crack growth.
l
i o
i Protection from Thermal Degradation.
Care will be necessary to avoid heating to an excessive, unsafe level of items such as the concrete containment and support cavity.-or instrumentation that would not be qualified for heat levels above 650*F.
While the instrumentation might be removable, the concrete-surely is not; thus, care rust be. taken to insulate or cool the concrete to assure that it does not lose its strength because of an annealing operation (Ref. 4)..
Personnel Exposure. Because thh entire annealing operation would have to be done in and around an irradiated and contaminated reactor and is both extensive and complex, the possibility exists for a high level of radiation exposure, both individual and collective, to the operating personnel who would be required to emplace the heating elements, provide for insulation or protection of thE concrete, etc. Any preannealing logistical study would have.to estimate the number and type of crafts personnel _ needed for an annealing operation, and would have to provide assurance that an adequate number would be available, based on estimates of radiatinn levels for different tasks and workers, and the rate at which they could be expected to reach their pennissible levels.
Additional Considerations (Ref. 5)
The following itens are noted because they are important and should be.
very strongly considered for an annealing although they are not critical for demonstrating feasibility.
Air-transported Contamination. This refers to the radioactive contamination that could come off the reactor vessel inside wall and be blown around during annealing.
It is especially.
likely under the-current reference conditions which call for annealing the vessel dry rather than under water. '
teak-Proof Vessel Cofferdam Seal. The present refererice conditions for annealing suggest that a cofferdam will be built to separate the dry inside of the vessel during heating frun the upper head of water in the refueling tank. - The water head is clearly necessary for shielding of personnel and for continuing storage of. fuel, thennal shield, and other internals.
Thus, the cofferdam and seal must be leak and failure proof to prevent the in-leakage of water which would then-flash to stsam and potentially blow out the cofferdam resulting in sudden flooding 'of the vessel with' cold water and potential warpage or cracking of the vessel from the thennal shock.
i l
l i
l
~
Code and HRC Requalification Criteria (Ref. 6).
Pres su re vessels built to ASME Code Section 111 are presently rated only to 650*F. Assuming that an annealing were to be conducted at 850*F, an extensive code case, or a modified section would have to be added that would spell out: the limits of stresses,-
for example, and any 'other-items thought necessary' for requalification.
of the vessel. The NRC would surely follow development of the.
code requalification criteria to be assured that all necessities -
were covered. This item is reviewed in mJch greater detail in.
a later section.
Inservice Inspections and Hydrotest (Ref. 7).
In order to assure that the annealing did not introduce any harmful defects and to assure that the annealing did not cause.an otherwise small and benign defect to grow to anLunacceptable level, it would be highly recommended to perform a 100% pre-and postanneal inspection of' the vessel and nozzle areas potentially affected by the heating'and the differential thermal stresses.
Serious thought should also be given to performance of a pressure test -
such as required by Section XI of the ASME Code following the annealing as further evidence that the annealed vessel is "as-safe"'as it was prior to annealing.-
4 4
4 4
1 4
4 4
a l.
5 1
II. Candida te Vessels for Annealing Demons tration Five reactor vessels were reviewed as candidates for annealing demonstrations.
These included the vessels of Indian Point-1. Shippingport XRB-A, Humboldt Bay, and BR-3.
The first four of these plants have been shutdown permanently and would potentially be available for an annealing demonstration experiment which might involve significant, modification of the plant and destructive examination of the vessel (for metallurgical and mechanical testing). The BR-3 is still operating and may well be annealed during September 1983, but not under conditions of especially high interest to NRC.
An annealing demonstration..wi.th any of the above reactor vessels would give plant specific infom'a~ tion on the recovery of properties of irradiation embrittled material, engineering logistics on the process and procedure applied, including infomation on system response (support movements, vessel distortions, concrete temperature control, etc.).
However, none of these plants is sufficiently similar to the commercial PWRs in question to pemit the reproduction of all the major parameters important to a successful commercial vessel anneal simulation. The following is a discussion of the candidate vessels and the limitations associated with their use in an annealing demonstration.
Indian Point-1 (Ref. 8) 1.
The Indian Point reactor is made of 7-inch thick A212-B steel, and had an operating temperature of about 475"F.
It is expected that the kinetics of embrittlement would be different because the vessel steel is A212-B (rather than A533 typical of new plants), and the vessel operating temperature was very low compared to current practice.
Very little infomation exists on the embrittlement characteristics of the IP-1 vessel steel, and no surveillance specimens or archive material are available.
2.
Several features of the design would be expected to result in significantly different stress patterns due to temperature gradients during annealing which would make system response and distortions nontypical.
These include (a) bottom inlet nozzles, (b) skirt support, (c) smaller vessel, and (d) stitch weld cladding.
3.
Rather than concrete shielding around the vessel. IP-1 has a layer of insulation, a 3 to 4-inch air gap and a 12-inch water-filled steel shield tank.
Thus, annealing IP-1 would not generate infomation on protecting concrete from themal degradation.
4.
The head flange configuration is pinched in a manner that restricts the direct entry of heating systems sized to fit up against the vessel I.D. wall.
e
l i'
6 L
1 l*
5.
The four steam generators are of horizontal t!-tube design rather j
l than the current upright design.
Furthemore, the inlet and outlet l
nozzles are at the bottom and top of the vessel, respectively (a situation. which is completely atypical of current plants), and the nozzles are placed on a single 90* segment of the vessel circumference rather than being placed evenly around the entire 360' of the vessel circunference.
6.
The IP-1 reactor is currently shutdown, with no plans at all on the part of the utility to restart the plant.
ShippingportRPV(Ref.'9) 1.
The Shippingport reactor vessel is 8-3/8 inches thick A302-B steel, with an operating temperature of about 490*F.
The material and wall thickness are fairly typical of current commercial PWRs, but the operating tenperature of 490*F compares poorly to the current typical operating temperature of 550*F.
No surveillance specimens or large base of irradiation effects data exist for the vessel.
2.
Design features exist which are similar to those in IP-1 and are expected to result in different stress patterns due to temperature gradients during annealing. These include bottom inlet nozzles, upper outlet nozzles, different support design, and a smaller vessel.
The four steam generators are located in pairs dir
- ,rically opposite each other across the vessel, but they are horizontal rather than vertical.
3.
Shippingport has a water filled, steel-lined thennal shield tank which also supports the vessel.
This arrangement is not amenable to generating infonnation on protecting concrete from thermal degradation.
4.
The roll-bonded cladding was approximately 30 unbonded. This condition would seriously affect the heat transfer characteristics and the associated temperature and stress gradients in the vessel during an anneal.
5.
The vessel was designed and constructed to Section I of the ASME i
Code and thus has very low design stress levels.
Later plants were constructed to Section III of the ASME Code.
6.
The Shippingport plant will be turned over to DOE during the summer of 1984 for decommissioning and disposal, leading to burial of the vessel at Hanford, Washington, late in 1986.
n-r.
L o
l l_
7 e
Hunboldt Bay-(Ref.10) 1.
The Humboldt Bay reactor is made of 5-inch thick A302-B steel. The vessel operating tenperature was about 550*F..Although the plant operated. for 12 yeans, the vessel doe,s not have-a large irradiation-induced NDT shift because _of the large water gap between fuel and vessel. Nevertheless, a fair amount of-irradiation effects data is available from' the surveillance program of the plant.
2.
Humboldt Bay -is 'a BWR and thus has.a. design very atypical of PWRs; especially in that,it..h,as no pumps and only has small feedwater inlet nozzles about m'id-way up the vessel, and one large steam exit line out of the top head.
3.
The vessel resides in a steel drywell surrounded by concrete. 'An atypically large gap exists between the vessel _ and drywell. The vessel is suspended from the top. by steel suspension bars (another' significant atypicality).
4.
The plant is currently shutdown, primarily for reasons _ of noncompliance with seismic restraint requirements and is not expected to ever be l.
res tarted.
KRB-AGundremmingen-(Ref.11) 1.
The V.RB-A reactor is a small BWR similar to Dresden-1 and is located in Gunzberg, about 50km west of Augsburg FRG. The reactor vessel is 4 7/8 inch thick A336 steel (German designation 20 HiCr26),' and-had an operating tenperature of about 545'F. A small: amount of irradiation effects data is available from the surveillance program, i
but RES currently is planning to remove material from,the vessel for testing to determine the as-irradiated and laboratory ' annealed H
(850*F,168 hours0.00194 days <br />0.0467 hours <br />2.777778e-4 weeks <br />6.3924e-5 months <br />) properties.
2.
This plant is a BWR with design features atypical of PWRs such as bottom nozzles, skirt support, and lower-head control rod drives.
3.
This plant is unavailable for an annealing experiment because it is scheduled for complete decommissioning and sealing shut by the Fall of 1983.
e a
i i
e
8 I
BR-3 (Ref.12) '
l.
The BR-3 f a small"PWR being operated as a power and test reactor by the Belgian Government at the CEN/SCK Hol Laboratory.
The reactor was originally built for the;1958 World's Fair held in B russels.
The reactor vessel is 4.5-inch thick A302-B steel, and has an operating temperature of about 490*F. A very limited amount of infonnation is available on the exact material of the vessel and -
the single, vertical seam weld, but a great deal of periferal information has been and is being developed for the. weld metal to support studies of.the safety evaluation of the plant. The studies 3
are underway because~the Belgians believe that the vessel weld is sufficiently embrittled such that annealing may be the only way to safely keep the plant in operation.
2.
The BR-3 is quite similar to a current PWR in that it only has nozzles in the upper part of the vessel and is supported by those nozzles.
The three nozzles extend from one 180' segment around the vessel-circunference rather than being evenly spaced. around the vessel. The vessel is quite thick for its size, inasmuch as it was designed to ASME-Section I rules and thus has very low design stresses.
l 3.
Surrounding the vessel is an air gap containing the vessel insulation.
)
a 4-inch thick, steel-lined water shield tank and finally a concrete support cavity. This arrangement is a very poor nodel of the current PWR cavity arrangement.
4.
It appears very likely that the BR-3 will be annealed during the summer of 1983. The Belgians would be most eager to have the NRC participate in the annealing in any way.
Items that mitigate against our gaining much information from the anneal, however, center on the specification of 650*F annealing under " wet" conditions, no opportunity to remove and test material from the vessel wall either before or after the annealing, and the fact that the low temperature is within code allowances so that stresses would be well within acceptable limits.
e 0
e O
g i
1 III. Properties Recovery Throuch Annealing __
l-i The principle _of annealing recovery of material toughness properties in
~
steels degraded by neutron irradiation has been under study since the.
i 1950s. One of the first detailed studies of_ annealing effects and parameters was reported by Carpenter, Knopf and Byron in 1964 (Ref.13).
i Soon thereaf ter, the Amy SM-1A reactor wa) annealed in-place to. recover toughness properties and extend its lifetime-(Ref.14).. The details of.
l that operation are described later in this section.
Laboratory testing in support of the SH-1A anneal (Ref.15) and additional studies by the.
AEC (Refs.16 and 17), the NRC (Refs.17 through 20), by EPRI (Ref. 21) and by others have developed the major parametric effects of annealing on toughness recovery of. irradiated pressure vesse1~ steels; furthermore, as early as 1968, there was an analysis of possible changes in plant-i operations :to limit or mitigate abrittlement so as. to preclude _ or minimize annealing (Ref. 22). These _ studies utilized various grades of pressure vessel steels, including those used in commercial LWRs. Furthemore, e
the ability to completely restore certain-. toughness -properties (transition temperature shif t and upper shelf drop) has been demonstrated under -
a certain limited conditions (Refs.17, 20,.and 21).
.The body of data
~
accinulated to date is relatively large.but has significant variability, and only some of it is directly applicable to annealing of the large-commercial PWRs under current consideration.
Annealing Variables The major variable controlling annealing recovery is.the time-temperature rela tions hip.
Other major variables include the irradiation temperature, i
neutron fluence and spectrum,- the composition of im?urity elements..and the type and grade of steel.
Some observations with respect to annealing include:
l.'
Annealing at progressively higher temperatures results in commensurately
]
higher recovery.
2.
Much recovery occurs in the first few hours of annealing; additional recovery is inversely proportional to the annealing time (typically about 1 week).
3.
Westinghouse,.in their EPRI study (Ref. 21'), and'NRL (Ref. 20) both showed recoveries of 80 to 100% of the transition temperature shift, under specific conditions of high temperature annealing, and j
Hestinghouse showed over 100%. recovery of the upper shelf drop for i
annealing at'850*F for 168' hours. However, since the reasons.for the variable recovery response are not known, full-recovery _for these condi tions cannot be guaranteed.
4.
The temperature at which a stee1Lhas been irradiated effects.the ability of the steel to recover due to annealing. ' Thus, steels irradiated at higher. temperatures respond less readily to annealing.
i f
treatment; i.e., although. steels irradiated at PWR tmperatures (550*F) require annealing at 800 to 900*F for complete recovery, steels irradiated at 500*F can recover completely at 740*F.
i
l 10 5.
Impurity element composition has an affect on the propensity of a steel to recover, which is secondary only to time-temperature (of annealing) effects.
For example, increasing copper content results l
in lesser recovery of Charpy upper shelt energy (Refs. 20 and 21).
1 6.
Neutron fluence does not appear to affect the recovery kinetics of the transition temperature shift but'does appear to have an effect on recovery of upper shelf for lower temperature anneals.
Reirradiation Effects Infomation on the rate of reembrittlement of steels that have been irradiated, annealed and reirradiated was developed under different sets of conditions and, therefore, is somewhat conflicting and shows considerable scatter in the ita.. For annealing at 750*F, reirradiation at 550*F tends to result in the same trend of embrittlement caused by the initial irradiation (Ref.17)
However, other tends have been observed at higher annealing temperatures (Ref. 21). Westinghouse data indicate that reembrittlement as measured by transition temperature shift continues at i
a rate expected if no anneal had been perfomed; since this rate is significantly less than the rate at the start of the first irradiation, the time to reach the same embrittlement level would be longer.
ENSA/ MEA currently has experimental work nearly completed to elecidate the actual rate of reembrittlement following annealing, at least for several weld o
netals. The reenbrittlement rate as measured by upper shelf drop does not follow the same trend as the transition temperature increase but rather is between the initial rate and the rate expected if no annealing had occurred.
Because reembrittlement is highly specific to the material and the proposed annealing conditions, it is necessary. for a utility planning to anneal an irradiated vessel to develop an appropriate, modified surveillance program that would measure the effects of the annealing operation and subsequent reirradiation.
Such a program should include surveillance specimens that are representative of the vessel material in the irradiated and annealed conditions.
Amy SM-1A Annealing Operation i
Because of its compct design, low operating temperature (430*F) and sensitive vessel siael (A 350-1.F1, modified), the SM-1A reached a point where annealing re:overy of the reactor vessel was required after only a j
few years of operation (Ref.14).
The 46-inch diameter, 3 3/8-inch i
thick vessel was annealed by raising the reactor system temperature to 1
572*F for 143 hours0.00166 days <br />0.0397 hours <br />2.364418e-4 weeks <br />5.44115e-5 months <br /> using nuclear heat. The tmperature was established by ASME code limitations, and the time by experimental testing.
Fran a study that complemented the annealing operation and from capsules in the j
reactor, recovery was estimated to be 73% (Ref.15).
No significant j
radiation exposure to plant personnel resulted frun the operation. The total time required to install the necessary additional equipment and to perfom the anneal was 5 weeks.
The maximum pressure vessel wall transition temperature was changed from 190*F before anneal' to 0*F af ter the anneal.
Extended life as the result of the anneal was estimated to be 124.7 iM-years of reactor operation, based on evaluation of surveillance capsule data.
IV. Annealing Heat Treatment Process The pr blem of in situ annealing heat treatment of the beltline region of commercial PITii reactor vessels was discussed (Ref. 5) with a number of architect-engineering fims and commercial heat treating fims who specialize in field erection of large structures. These included Bechtel Power Corporation; Chicago Bridge and Ironi pyromet Industries;~ Applied Energy Systems, Inc.; and Cooperheat.
Also, a Westinghouse report for EPRI on the subject discusses procedures for such an operation developed by World Stress Corp., Inc. (Ref. 3).
It was generally concluded from these discussions (Ref. 5) that such an annealing operation would' fall within the state. of the art existing today. Much larger vessels have been annealed at higher temperatures.
However, the operations would require a significant amount of engineering-in two categories: the'themal design problem which includes heating element design,-location and heat transfer; and the mechanical design which involves allowance for entry of the device.through a reduced section and the adjustment of components outwardly into the proper locations once the device is in position.
Two state-of-the-art methods of heating have been proposed by heat treating companies. One would include commercially available electric heaters placed along the vessel wall with appropriate ducting to allow s-for forced air circulation. This approach would-requireg a completely dry vessel to avoid arcing. The alternate method would circulate hot gas using externally filtered gas fired equipment as the source of heat.
Such a procedure would require the development of procedures to control airborne contamination.
In either case, an intricate temperature measuring and control system would be required. Such a system would probably utilize sheathed themocouples mechanically pressed against the vessel inside walls.
Such a. system has been used in the past for other applications (Ref. 5) but may need additional testing at the temperature of interest.
Temperature control of the concrete structure outside of the vessel would be effected by the existing' gas cooling system with any appropriate supplementary flow rate equipment and baffling deemed necessary.
Temperature profile and cooling and heating rates would need to be established by appropriate themal and stress analysis.. It would be necessary that these analyses establish the limiting conditions to avoid any significant distortion of the vessel that might affect the reinsta11ation of the reactor head and vessel internals.
l 4
12 l
l l
j Y.
Critical Factors Affecting Annealing l
l Inherent in perfoming a themal anneal of a reactor. pressure vessel are l
critical factors which must be resolved on generic and plant, specific bases.
)
Guidelines for evaluating these factors wi11 evolve out of ASME and A5174 activities and NRC safety guidelines.
The following list of factors may not be exhaustive, but does represent the major engineering and logistical concerns _ needing resolution prior to an actual anneal.
Critical _ _"sctors for Engineering and L.ogistical Concerns (Refs.1, 3, and 5) l
.., ~ '
Juction and removal of a heating apparatus into the vessel and 1.
.;essful application of heat per specifications. Type of heating 2system to be employed and proper filtering of' air to avoid radioactive j
contamination if convection type system is used.
2.
Proper themal and stress analyses to insure that themal stress gradients and eventual residual stresses are minimized and to avoid l
vessel distortion. Design of the heating system and heatup/cooldown rates will be dependent upon these analyses.
3.
Thermal expansion or growth that could affect dimensional stability.
4.
Adequate measurement of temperatures during the annealing cycle to allow precise control and monitoring.
5.
Cavity seal integrity if coffer dam approach is used to store internal and coolant, and to separate vessel inside from fuel handling pool.
4 6.
Personnel exposure beyond that required for an inservice inspection.
7.
Control of airborne contamination.
I 8.
Adequate review, control, and measurement of concrete temperatures-to avoid concrete degradation (including knowledge of actual cavity dimensions).
9.
De+ ailed preplanning to guard against loss of reactor. internal instrumentation, loss of heaters during anneal, etc.
- 10. Heating of the vessel lower head (where 'some vessels have a lot of instrumentation).
- 11. Requalification and assuring compliance with the appropriate codes and standards, and establishing inspection requirements for after annealing (Regulatory Guide 1.150).
e e
!1' 13 o
- 12. Application of current experimental data to real. reac_ tor welds, neutron environment, and actual fracture toughness.-
- 13. Measurement of annealing response to the critical welds or materials in questf on.
- 14. Subsequent rembrittlement rate as. measured through an appropriate i
1 j
surveillance program.
~
Critical Factors for Codes and Standards (Refs. 5, 6, and 7)'
i The nomandatory fracture mechanics procedures specified in Appendix G of Section III of the ASME Code for construction of nuclear components are required for compliance with Appendix G of 10 CFR Part 50 to ensure aderitrate safety margins for. nuclear reactor-pressure vessels.
If these-safety margins cannot be satisfied during the lifetime of the reactor. -
1 Appendix G of 10 CFR Part 50 identifies themal annealing of the beltline l.
region as a means to recover material-toughness' properties (and restore l
adequate margins of safety). Once a reactor vessel is put into service.
l the ASME Code,Section XI, for inservice inspection of nuclear components-is followed for maintenance operations in case of flaw' or defect detection.
l Only by inference from Section XI isSection III invoked again with regard to a procedure such as themal annealing, and this includes the general requirement that-the operation not be " harmful." 'An important
~
goal of any new code or NRC guideline would be a definition of "hamful."
These l
Other critical factors exist which need ASME Code guidance.
factors are:
postanneal inspection requirements, themal and stress l
analyses requirements to establish permissible heating / cooling rates and gradients, assurance against damage to other plant components (such as concrete) and other requalification items (such as a cold rehydro test).
It is anticipated that an equivalent 'of a Section XI,10-year 10.07.-
volumetric inspection of all welds will have to be perfomed.. possibly j
following Regulatory Guide 1.150. An adequate elastic-plastic stress l
analysis will probably need to be perfomed, and themal analysis studies l
for eva'luating concrete temperatures will be required. : Some activity l
within the ASME Section XI Subgroup' on-Repairs and Replacements has begun, but action within this subgroup appears to be slow.. A new subgroup i
is tentatively being fomed on requalification in general, which would include the issue of themal annealing; ASME action may be faster through this new subgroup.
l Most of the' information needed to evaluate the safety and outcome.of L.
performing a themal anneal is plant specific.
The first concern. relates to the actual material properties prior to and af ter the anneal cycle.
Many of the older commercial reactors do not have the actual, critical 3
weld metals in their surveillance program.
Therefore, an assessment of L,
14
/
?
material property recovery mt;y be contingent upon comparison to like materials in the data base for test reactor conditions.
Definitive relationships between time, temperature, chemistry, weld metal and flux type, and product fom very likely could not'be established due to the limited basis of information. Further expansion of the current data base of annealing recovery is needed to assess real vessel-weldment toughness properties for different chemistries (especially copper and.
nickel variations) and under low flux irradiations typical of _.those for pressure vessel walls for commercial reactors- (Ref. '20). This infomation is being developed by Haterials Engineering-Associates Inc. (EllSA/ MEA).
- * * "g en e
l e
j.
i e
4 l
i
-.~
. -....f
1 i
15 VI. Research Basis for Regulatory Actions In order to satisfy oneself that an annealing can be successfully accomplished, it is necessary to either perfom an integral demonstration as discussed in Section I of this report, or to study and prove the individual parts in a series of separate effects tests and analyses.
and the work, which has been The latter choice has been followed in RES,is described in this section.
accomplished and is currently in progress, The parts which have been and are being developed in detail by NRC, which are felt to be sufficient for comprehensive NRC review, include the fracture mechanics crack initiation / arrest methodology, irradiation embrittlement and recovery trends, neutron dosimetry and fluence prediction methodo1ngy, code.and regulatory criteria that would have to be satisfied for conduct of annealin'g'and requalification for service, a listing of systems effects considerations with the results'of analyses developed by RES and others, evaluation of material properties and annealing characteristics of retired-from-service reactor vessels, and material recovery and engineering and logistics data available from such nuclear plants as l
have actually been annealed in-service.
Fracture Mechanics and Analysis Methodology (Ref. 23)
The accuracy and applicability of Linear-Plastic Fracture Mechanics as well as the extension into elastic-plastic fract.ure mechanics for irradiated,
reactor vessels have been developed and validated through research programs in RES, notably the HSST Program at OJ Ridge and also a prior program at the Naval Research Laboratory.
The critical approach in these programs is' to develop an analytical methodology for prediction of j
reactor vessel per4mance under loading conditions representing normal operation and severe accidents, with the size and placement _of flaws plus a specified :naterial fracture toughness condition as variables. A very extensive series of thick-walled pressure vessel tests has been carried out to prove this methodology.
The straightforward fracture analysis methodology has been extended to encompass the predictive analysis for themal shock and pressurized themal shock conditions; thus, the basis has been set for defining conditions under which embrittlement can no longer be tolerated and remedial measures such as annealing would be required to recover, fracture toughness properties.
The studies in this program provide deteministic assessments of reactor vessel integrity under nomal or accident conditions, and thu's the remaining margin prior to the necessity for remedial action, while another program provides a similar assessment but from a probabilistic viewpoint.
The probabilistic l
approach uses the VISA (Vessel Integrity Simulation Analysis) code developed by the RES staff to evaluate reactor pressure vessel integrity.
The VISA code utilizes Monte Carlo simulation techniques and has the ability to treat flaw depth, copper content, fluence, initial RTNDT'
!~
shift in RTNDT, and fracture toughness as random variables.
i e
i6 t
l The code estimates the conditional probabilities of crack initiation and through wall crack penetration given a specific pressurized themal i
l shock transient.
These probabilities coupled with estimates of transient-l frequencies fonn the basis for the pressurized themal shock risk analysis i
which has been performed by the HRC staff.,
Annealing and Reembrittlement Researcii
~
Very extensive and long-tem efforts have been underway under NRC (and formerly AEC) contract with the Naval Research Laboratory and }!aterials i
l Engineering Associates, Inc. for studies of the annealing and reembrittlement characteristics of pressure vessel steels. This has been discussed in Section III above. The current program is intended to characterize i
metallurgical factors affecting irradiation sensitivity, postirradiation i
properties recovery, and sensitivity to reirradiation. A major part of the work is directed at determining the specific reembrittlement path.of pressure vessel steel test specimens irradiated at commercial nuclear reactor vessel conditions, annealed at the conditions simulating an in situ anneal, followed by reirradiation at reactor vessel conditions.
l The questions to be ~ answered are: how rapid is the rate of reembrittlement l
and what factors affect the rate? The program will also study the synergistic interactions of selected combination of elements (especially Cu and Ni) on radiation sensitivity levels.
Finally, the program will l.,
test and analyze specimens from the dosimetry benchmark carried out by ORNL, as described below.
Neutron Dosimetry and Fluence Predictions The predictions of mbrittlement in reactor vessel walls is dependent upon a knowledge of the flux of neutrons which causes the property change. This is obtained from flux monitors or detectors located in every experimental or surveillance capsule; the flux data-is used in turn, to adjust and validate theoretical reactor physics calculations of the neutron flux and energy spectrum performed-for locations in the vessal wall of interest for fracture predictions.
It is the trends of embrittlement versus neutron flux (or time-integrated fluence) that are the basis for predictions of crack initiation, propagation, and arrest in irradiated vessels,.especially under accident conditions.
RES programs at Hanford Engineering Development Laboratory, Oak Ridge National Laboratory
~
and at the NBS, all cmbined in a Surveillance Dosimetry Improvement l'
Frogram, (Ref. 24) are working to define and reduce the uncertainties in i
the flux and fluence measurements and methodologies predictions and to establish a series of benchmarks for proving _ the accuracy of prediction methods for both the neutron flux and embrittlement in surveillance analyses and for reactor vessel walls.
The studies have been instrumental in improving the ac' curacy of the Regulatory Guide _l'.99 basis for prediction of vessel embrittlement, and setting forth displacements per atom (dpa) l as a new criterion for fluence determinations in place of the previous criterion of " flux > 1MeV."
l
~
]
1 2.
17-Annealing Criteria and Plant Systems Response The only existing criteria for annealing are given in ASTM Standard Recommended Guide E 509-74, "In-Service Annealing of Water-Cooled Nuclear Reactor Vessels." This provides a fair outline.of material requirements, but it is not adequate for systems response' considerations.
It is being revised and will benefit from3tudies being conducted under i
i RES contract at EG&G on the subject of annealing criteria and systems requirements.
This effort has objectives to establish criteria for requirements and standards for materials and systems response aspects of f
in situ annealing of commercial nuclear power reactor pressure vessels, and to identify technical areas which require additional 'research before l
such criteria can be established. Through this contract EG&G is reviewing I
all past and ongoing reses"rch dealing with the parametric effects of annealing on fracture toughness recovery of irradiated embrittled vessel steels, including programs sponsored, by.NRC, DOE /AEC, and 'EPRI, and will.
i identify the advantages and disadvantages of considered annealing procedures.
In addition, EG&G is reviewing all past and ongoing programs dealing.
i with the systen aspects of in situ annealing of pressure-vessels'and i
will identify, if any, those aspects of in situ annealing for which there are no or insufficient technically proved. bases.
Those system
)
aspects which are generic and those covered by the ASME Code or NRC j
provisions will be identified as will.those whir.h are clearly plant specific. Criteria will be developed for any in situ annealing: procedures to ensure the continued safe operation' of the reactor pressure vessel and the remainder of plant components.
Active participation in ASME cnd ASTM Committees dealing with these matters will be maintained by NRC staff and. contractors.'
Analysis of Commercial Reactor Vessel Material
}
1 KRB-A Gundremmingen - RES presently is exploring a program to remove material from the pressure vessel of this reactor (mentioned.in 'Section II above) and to have specimens machined and tested therefrom to establish fracture toughness and annealing characteristics of steel. irradiated in situ at temperatures, stress levels. and flux rates typical of service conditions.
Very important elements of verification are expected from this program:
Charpy-y and fracture toughness measurements of in situ irradiated steel; correlation of in situ. measured properties with those from surveillance irradiations and test reactor irradiations; annealing characteristics of in' situ irradiated steel for correlation with characterist'ics from surveillance and test reactor irradiations; extrapolation of fluence 4
and embrittlement from surveillance locations to in-vessel locations; correlation of dpa (displacements per atom) damage predictions in the
~
wall with Charpy-V and fracture toughness measurements; and crack arrest measurements of the vessel wall from inside to outside providing-an indication of the potential crack stopping power of the increasing toughness lying in that direction.
e e-
18
~
Shippingport - The option exists. for a program for removal and test of material from the Shippingport reactor pressure vessel (described in Section II above) beginning in the Summer of 1984, when the. reactor becanes available for such testing. The goals' of this work would be :
similar tu those outlined above.for the Gundremmingen reactor vessel s teel. The steel of the Shippingport reactor vessel is closer to that
~
of current.U.S. practice, but the. service irradiation tenperature, being lower, makes the potential value of this study of. considerably less value for verification of annealing procedures:and effects.
BR Details of this1rcactor, loc'ated.in Belgium, have been discusssd also in Section II above.-v RES has a-cooperative arrangement.with.the LEll/SCK Laboratory in Hol, Belgium, that will allow us to. obtain~ all infonnation developed in support of and resulting from any annealing-done ~ in.this reactor. We also have. defined an' outline of a program of l
our participation in any BR-3 annealing, which is aimed at measuring systen effects parameters, rather than material characteristics.as 'this latter item is the prime focus of the' Belgian efforts.
SH-1A - This small Army reactor was successfully annealed, so that appropriate material and ' system response data _ are available for NRC consideration.
A summary of that annealing operation was.provided in j.,
Section III above.
}
i l
e pa f
~
[.
I f
is i
VI. Conclusions on Feasibility and Demonstration of Annealing Feasibility of Annealing A survey of expert personnel from pressurized water reactor suppliers (Combustion Engineering, Inc., Babcock and Wilcox Co., and Westinghouse Electric Corporation), architect engineers (Bechtel, Stone and Webster),
heat treaters (Pyromet Industries and Cooperheat), and others (Chicago Bridge and Iron Co., Electric Power Research Institute, NUS Corporation.
Enenjy Incorporated, and General Electric, Knolls Atomic Power Labo'ratory) has indicated that thermal annealing of a reactor pressure vessel is l
feasible. There are several areas of concern which need to be addressed l
prior to an actual anneal, however. These areas of concern were presented L
earlier as critical factors which need to be addressed. The attached l
EG&G report describes 'in' detail the response of the industry experts.
The need for a demonstration had a mixed response from the experts.
Part of this difference was related to the type of demonstration. proposed.
Many thought that an engineering logistics demons tration would be beneficial,,
l while the heat treaters dismissed this test as insignificant.
Others agreed that a full demonstration on a decommissioned reactor pressure vessel would be invaluable. However, when the candidate vessels were analyzed in detail, the cost effectiveness of-performing a demonstration anneal on any of these vessels was questioned. Although there was no consensus agreement, the overall conclusion was drawn that a demonstration
(
should be restricted to a vessel with the appropriate geometry, materials, and fluence levels' of concern. None of the candidate vessels which might be considered meets this restriction.
In summary, then, both the EPRI study (Ref. 3) and the attached EG&G report indicate that the annealing of reactor vessel is feasible and that state-of-the-art heating equipment could be used for such an application.
However, a considerable amount of. engineering would be required prior to the actual conduct of such an anneal..It is estimated (Ref. 5)'that a 1 year engineering study would be required to identify the detailed requirements for such an operation and at lea'st another year of effort would be-required to design and construct equipment for annealing and to. make the necessary plant modifications to maintain the appropriate controls of' temperature profiles and personnel exposure levels.
l i
i t
I e'
20 Results Expected From A Demons tration Experiment A successful in situ themal anneal experiment using one of the candidate prototype reactor vessel available would demonstrate the general ability to perfom such an anneal.
It would also verify engineering design features such as the successful operation of tF equipment and instrumentation and the general ability to predict and coli,, trol temperatu.'e and loading conditions and thereby the. control of vessel growth and distortions, piping reactions, themal gradients and dimensional stability.
- Further, experience would be gained in limiting personnel exposure, controlling contamination, decontamination of heating equipment as well as the more ~
routine planning, scheduling'and perfonning the individual steps of the anneal.
. Additional dat'aon' annealing effects on irradiation embrittled material would be generated if destructive examination of the vessel is l
pemitted.
Inspection operations to be applied subsequent to the annealing would be demons trated.
Because the candidate reactor vessels are not prototypic, those detailed l
engineering procedures, stress and themal analysis and irradiation l
effects data developed would not be directly applicable to the PWRs of l
concern, and in that sense, the experimental anneal could not be considered l
a complete demonstration. Geometric features that control themal stresses and flow patterns are significantly different between the two groups of vessels so that the potential distortions or excessive loadings that might lead to subsequent assembly problems would not be simulated.
This is especially true of the nozzle configurations where the major stress concentrations occur. The ability to control the temperature of concrete supports to a safe level to avoid degradation of the concrete would not be demonstrated because the available vessels are designed without concrete supports in the heated region. The size and shape of the available vessels are sufficiently different that the heat treatment i
equipment design would not apply direct 1l,. limiting the ability.to extrapolate operational experience, control of contamination and other engineering i
details. Also, although the material properties would be applicable to the data bank of properties on irradiation effects, it would be a single point of a body of data affected by material composition and fluence levels which vary with each reactor vessel.
l
21 Summary A definition of an annealing demonstration is taken to mean that mechanical properties of the actual vessel steel can be measured before and after the annealing, and that following successful application of heat, that the plant can be reassembled and is capable of resuming power producing service in a condition no less safe than before the anneal. The demonstration involves a number of factors that fall int,,o the areas of Haterials considerations, and of Engineering and Logistical considerations, including especially a demonstration of the degree of properties recovery, successful application of heat to tha vessel and retention of the dimensional accuracy of the plant so dat it can be safely reassembled.
There currently are five atypical nuclear power plants available in the world that could be used for an annealing demonstration. All of them have serious deficiencies which would severely hinder acq0isition of data on the annealing that would meet the criteria outlined under Section I on Definition and Section V on Critical Factors.
The most serious problems are of low operating temperature (Indian Point-1, 475'F; Shippingport,,
490*F; and BR-3, 490*F), atypical nozzle placement (lower and upper nozzles in Indian Point-1 and Shippingport; lower nozzles only on KRB-A; small feedwater nozzles and one steam exit head-nozzle in Humboldt Bay; and three upper nozzles located around only 180* of the vessel circumference in BR-3), vessel support significantly different from current'PWRs in all except the BR-3, and reactor cavity differences from current PWRs (especially featuring steel-lined water-filled tanks and rather large air gaps instead of concrete-lined cavities).
Four plants could all yield material from the vessel wall for pre and post-anneal material property investigations, whereas, the BR-3 is still operating and would continue to do so af ter the anneal proposed by them for Summer of 1983.
The BR-3 anneal would be done at 650*F under " wet" conditions and so would be of limited interest and value to NRC.
The engineering, evaluation, s
annealing perfonnance, and material property recovery demonstration operations (which would necessarily include a number of experimental irradiations of specially fabricated duplicate weld or base materials) would be very costly and time consuming, and as noted above, would yield only partial data on a number of critical items.
It has been estimated that it could cost from $30' to $50H for such an annealing, and in consideration of that fact, the benefits and value resulting do not appear to warrant this level of expense. A very infomal j
estimate from the industry suggested that a vessel annealing might cost l
about $25M and require about 2 years to complete including initial planning. Actual annealing operations have been estimated to require 90 days.
A significant amount of infonnation already exists relating to annealing of reactor pressure vessel steels to remove the embrittling effects of neutron irradiation on the structural steel.
This infonnation has O
22 been generated primarily in laboratory tests of irradiated sacimens but i
was also demonstrated on the Amy SM-1A reactor vessel.
It 1as been established that the degraded toughness properties nomally measured by surveillance tests can, in some cases, be-completely recovered by annealing at 850*F for about 1 week.
However, the existing body of test data contains too much variability to predict specific toughness values in
]
adva nce. This infomation may only be obtained ~with an appropriate surveillance test program and an experimental irradiation and annealing program, most probably dp? 'ne using test reactor' facilities and duplicate or model materials.
Heat treatment of large pressure vessels and other similar structures is routinely done in situ on a commercial basis.
A number of fims exist today that employ such methods on larger structures and at tenperatures higher than would be required for the annealing of a reactor under the current reference conditions.
Careful engineering of the heating apparatus would be required as well as careful placement of themoccuples and control of heatup, annealing heat application and cooldown.
A large number of critical factors and criteria have already been discussed in preceding sections. One final aspect that nust be considered, and one that is foremost on the minds of the industry is that of the ASME Ccde and HRC requirements and requalification conditions that would have to be met for restarting any annealed plant.
These are not developed at this time, but are the subject of considerable activity by professional groups in ASME and ASTM who are working on various aspects of this problem. RES is also working to develop such requirements and requalificatibn conditions through contract research efforts; the contractor is a member of the various groups so that the maximum transfer of ideas on these -
subjects can be accanplished.
It is felt quite strongly that the separate parts of reactor and system f
perforinance related to annealing can be studied separately and that -
these can successfully be integrated into an overall basis for use by the regulatory staff in detemining a position on annealing of any plant. The parts which have been and are being developed in detail by NRC, which are felt to be sufficient for a comprehensive NRC review, include the fracture mechanics crack initiation / arrest methodology, irradiation anbrittlement and recovery trends, neutron dosimetry and fluence prediction methodology, code and regulatory criteria that would have to be satisfied for conduct of annealing and return to service, a listing of systems effects considerations with the results of analyses developed by RES and others, evaluation of material properties and' annealing characteristics. of retired-from-service reactor vessels, and material recovery and engineering and logistics data available from such nuclear plants as have actually been annealed inservice.
l l
a,
~
In summary, we have concluded.that it-is feasible to conduct an operation-of heating up of a reactor vessel to remove irradiation embrittlement and to return the mechanical properties close to those at the start of service, which is the essence of annealing. _ However, we believe that a.
" demonstration".of annealing implies a number of additional concerns that cannot be adequately met using any of. the currently available reactor vessels, and we, furthennore, recommend against the NRC undertaking a demonstration of annealing. The primary: reasons for this conclusion.
are reiterated here.
The metallurgical, logistical, and engineering aspects of annealing are highly plant sp'ecific, and there is -no reactor plant presently available for a demonstration annealing that is sufficiently similar to the large PWRs' currently in service to approach such plant specificity.
If a.large plant more typical of those currently under review for the pressurized thennal shock problem became i
more genera 1' applicability.
In.no case, however, siould " success" available, an annealing demonstration using such a 31 ant could have or " failure" in a demonstratio.n imply a guarantee of " success" or
" failure" in any other plant undergoing annealing.
There already exists a 'very large, credib1e body of data on recovery of fracture toughness ' properties through annealing.
There exists a large body of experience in successful opplication of heat treatment to reactor vessels in the field and at temperatures higher than those contemplated for annealing.
The data base and methodologies developed through RES and other programs, coupled with the results 'of ongoing research programs, should. provide a sufficient basis upon which to. form regulatory decisions concerning any proposed, in situ annealing procedures for presently operating commercial plants.
The cost for a demonstration, thought to be in the range of $30-50ft, may be higher than the actual cost for annealing of a commercial vessel, but in any case, the benefits derived from a marginally-l qualified plant would not Justify such a cost.
j 1
1 1
i
,l j
^
.~
~.
l.
REFERENCES 1.
ASTM E 509-74, " Standard Recommended Guide for In-Senice Annealing of Water-Cooled Nuclear Reactor Vessels."
1 2.
ASTM E 185-73, " Standard. Recommended Practice for Surveillance Tests for Nuclear Reactor Yessels'."
3.
T. R. Mager and R.' D. Riskel, " Develop ent of a Generic Procedure for Thermal _ Annealing an Embrittled. Reactor Vessel Using a Dry -
Annealing Method," EPRI. NP-2493, July:1982.
4.
ACI-349, " Code Requirements for Nuclear Safety Related Concrete Structures," American Concrete Institute, December 1975.
i 2
5.
W. L. Server and H. W. Spaletta. " Evaluation of a Reactor Pressure Vessel ' Anneal Demonstration," EGG-FH-6083, October 1982.
6.
Section III, " Nuclear' Power Plant Components." ASME Boiler and Pressure Vessel Code.
7.
Section XI, " Inservice Inspection I Nuclear Power Plant Components,"
ASME Boiler and Pressure Yessel Code.
8.
Directory of Nuclear Reactors, IAEA, Vienna, Vol. IV 1962, pp. 59-64.
9.
Ibid., pp. 21-29 10.
Ibid., pp. 109-114.
11.
Ibid., Volume II,1968, pp.117-123.
l t
l 12.
Ibid..' Volume IV,1962, pp. 53-58.
l
)
13.
G. F. Carpenter, N. R. Knopf and E. S. Byron, " Anomalous Embrittling Effects Obsened During Irradiation' Studies on-Pressure Vessel' Steels," Nuclear Science and Engineering:- 19, 18-38 (1964).
i l
14.
U. Potapovs G. W. Knighton and A. S. Denton, " Critique of In-Place Annealing of SM-1A Nuclear Reactor Vessel," Nuclear. Engineering and Design 8 (1968) pp. 57..
- 15.
U. Potapovs; J. R. Hawthorne and C. Z. Serpan', Jr., " Notch Ductility I
Properties of SM-1A Reactor Pressure Vessel Following-the In-Place Annealing Operation," NRL Report-6721, May 21,1968, s
I L
.-7,_,. _,,. _ _ _,. _ _ _ _ _ _, _ _ _ _ _. - _ _ _,. _ _ - _ _ _ _,,,. _. -,. _ _ _.,.,,. - _ _ _, _ _ - _ _ _. _ _ _. _ _ _ _ _ - -. _., -. -. _, _ _, _ - _ - _,.
4 80 e
1 3
1 9
ENCLOSURE 2-
.i e
i
G
[)fC /0.SU FC
.c.
g#
UNITED STATES NUCLEAR REGULATORY COMMISSION ys g
y-g WASHINGTON, D. C. 20555
.f June 1,1983 f
....+
MEMORANDUM FOR:
C. Z. Serpan, Jr., Chief, MEBR, DET FROM:
A. Taboada, MEBR, DET.
SUBJECT:
TRIP REPORT TO COMBUSTION ENGINEERING Enclosed are highlights of a trip taken with W. L. Server of EG&G related to the in situ annealing prog:am.
At CE, we discussed routine and.special post weld heat treatment cperations performed on' reactor vessels that are very similar to a proposed Y situ anneal operation.
Also, CE presently has surplus reactor vessels and the facilities in house capable of performing an annealing experiment that would simulate the engineering features of an in situ reactor vessel anneal.
i l ' fyj-@
A.,
aboada Materials Engineering Branch Division of Engineering Technology
Enclosure:
As stated I
i A
I h
4 a
' 1
} *i 3 -
j i
i p
April 19, 1983' r
i
}
In Attendance:
John P. Houstrup, Krish M. Rajan, W. L. Server, A. Taboada~
4 l
This meeting was a followup lto informal. discussions ~ between Server:and i
Houstrup on several. aspects of the in situ annealing problem. We-also
[
discussed CE construction and weld repair practices with respect to j
reactor vessels.
1 4
Reactor. vessels have several critical dimensions thatLrelate to alignment,.
d h
the'most critical one being-the sealing surface between. the core barrel j
i and the hot leg nozzles. This dimension is machined to t 0.010 inches.
Other locations include the seismic alignment keys and core stop lugs,at -
L the lower end, and the: alignment between the control rods and the core.
j Standard CE fabrication procedure calls for these surfaces 'to be finished
)
machined while the vessel is still in two sections and prior to the final closure weld.1 The two sections are individually pst weld heat.
1 treated (PWHT)-at 1150*F. in a furnace before finish machining. A-field
[
PWHT is made on the final closure weld with the completed -vessel in-the 1
vertical position sitting on; the support pads.
This heat treatment is-j done from the outside with gas heaters that extend for approximately-3:
)
1/2 feet on either side of the weld and with thelinside surface covered j
1 i
with 4-inches 'of mineral insulation. :The vessel weld area-is heated at' l
i 100*F/hr to approximately 1150*F.
Despite. this heating, the nozzle -
i region remains.at approximately 80*F even though nof specific: cooling'is -
1 applied.
CE has observed no dimensional distortion problems.due to this
[
heat reatment.
j CE has also perfonned several PWHTs on finished vessels following. weld j'
repairs.
In one case, they made a 10-inch repair in a: nozzle of. the TVA--
J Yellow Creek 2 vessel followed by a local post weld heat treatment of-i the nozzle shell course.
In:this. case, alternate supports were used at-l the bottom of the vessel to avoid slumping.- In another case,,a Westinghouse.
vessel was repaired in the belt line region followed by a-PHWT. : Again,
,i 1
L the vessel was in the vertical-position.during the PWHT.
CE generally L
calls in'a heat treat specialist such as Cooperheat to perfonn unusual
}
PWHTs such as these..
j.
In several cases, field repairs and-PWHTs were made on piping.at reactor t
plants (i.e., Palisades). ; In one case, this operation resulted in.the l_
vessel flange area going out of round. Although CE believes that the l
vessel could have been str'aightened by moving the steam generator along-
[
its-friction supports, which would require about 1000 lbs load, instead j
the owners elected to machine the core support. mating ~. ledge.on the vessel to establish the proper. fit. - CE stated that machining in place to reeastablish dimensions has been performed on vessels in several'
- operating plants.
i
!I h.
i.
~
i
~
e
~
%. m
-_,s,.a.,,
.....Ji._,-m.,
,._-.,,i.,.,6.~
,...,,,,,,m,,wm-.__.
1" i -
g 2.
y For each of the PWHT performed by.CE, they made an elastic-plastic.
i L
thermal stress analysis. These analyses take into consideration primary creep. which is significant at 1150'F but also exists to a lesser extent at'850*F. CE generated high temperature mechanical properties for this This data has'been purchased by EPRI;and reportedLin: EPRI NP-i.
purpose.
i 2763, "High Temperature Elastic Plastic and Creep Properties of SA 533 i
Grade B Class 1 and.SAtS08 Materials.": To~ take advantage of this background, Server is pursuing a subcontract with CE to.perfom a generic elastic-j j
plastic analysis for. the thermal and stress conditions in a vessel during an in situ. anneal at 850*F for 168 hours0.00194 days <br />0.0467 hours <br />2.777778e-4 weeks <br />6.3924e-5 months <br />.-
i 1
j CE will 'also submit a. proposal to EG&G1to.perfom a mockup of.an in situ annealing experiment.-(at CE) on a' surplus vessel with simulated pipe.
. i j
loadings. Six or.seven. surplus vessels exist at CE.1 The experiment.
l would use one of several completed reactorzvessels, stored at CE that; were abandoned because of reactor cancellations...-In every respect,..
r j
j these are acceptable.PWR. reactor vessels with typica1' materials; nozzle-j locations and critical dimensions. The~ heat treat experiment would be:
made in the CE hydrotest pit, a concrete pit with.all' of the appropriate i
4 4
i support fixtures for reactor vessels and adequate space-for any instrumentation required.
\\
1 Overhead cranes exist that routinely handle these vessels. TCE would.
-include the appropriate elastic-plastic analysis:to predict thermal.
treatment schedule, appropriate temperature. instrumenting and' strain j
gaging and dimensional measurements before'and after the-heat treatment i
at 850*F for 168 hours0.00194 days <br />0.0467 hours <br />2.777778e-4 weeks <br />6.3924e-5 months <br />. This experiment' could be considered a demonstration of the engineering aspects of an in' situ anneal-and would answer many of the questions concerning possible distortion and damage.- It should also l
result in a procedure that could be referenced by standards organization 1
j' and would be useful-in establishing the precautions to be taken with
[
such an operation.-
il' t
}:
i 4
i
)
i j
[.
4 c
n W
h-e w w
- v
.w w- +
-v.mI--.
e y
e-y y
p.
+eee,b.' #
,w o-
+~
4 --e
--+-su..
m - b-w w w ee=
we " ree e---
wme.
e v-ev < es m-*r=-*b-ce.-+ee=d- -wm -i = w-
-w e
y
<