ML18089A184
| ML18089A184 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 06/02/1983 |
| From: | Palladino N NRC COMMISSION (OCM) |
| To: | Biden J SENATE |
| Shared Package | |
| ML18089A185 | List: |
| References | |
| NUDOCS 8306150533 | |
| Download: ML18089A184 (49) | |
Text
.. WlllTED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D. C. 20555 The Honorable Joseph R. Biden, Jr.
United States Senate Washington, D.C.
20510
Dear Senator Bi den:
June 2, 1983 Your March 4, 1983 letter requested information and specific responses to questions concerning the events which occurred at the Salem Nuclear Gener-ating Station, Unit 1, on February 22 and 25, 1983.
The responses to your questions are enclosed.
My fellow Commissioners and I are also concerned that these events have occurred.
We have closely monitored the staff 1 s followup of the mal-functions at the Salem plant as well as the broader implications for the nuclear power industry.
The facts, data and circumstances associated with.these events have been collected and documented as NUREG-0977.. This
.information was used by the staff to determine the safety issues associated
.. *with the events.. These issues were grouped into three areas: (1) equipment
".issues; (2) operating procedures and operator training and response; and
. (3) management issues.
The staff evaluated each of the areas to determine the'licensee's actions necessary to resolve the issues.
The staff concluded, as reported in their safety evaluation NUREG-0995, that the underlying causes of the problems were identified and resolved and, as such, the Salem facility could be allowed to restart.
We concurred with these findings.
Concurrently, an N_RC task force with representatives from three NRC offices was established to review and evaluate the generic implications.
The events can be characteriied as failures of the safety system to automati-
. cally shut down the reactor. However, the operators did identify the need for plant shutdown and did manually shut down the r~actor on both occasions such that the events themselves posed no serious threat to public health and safety.
However, we view the failures as serious safety concerbs since the automatic systems did not function as expected and if other plant conditions had existed, such as full power, considerable overpressure of the reactor system would h~ve occurred without prompt operator action.
The licensee has attributed the cause of the failµres to a lack of adequate maintenance to a part of the safety system, specifically, the circuit breakers which de-energize the control rods to cause rod inserticins and reactor shutdown.
Additional means to trip the Salem reactors (albeit not as rapidly) are discussed in response to your Question 6.
Regarding calculation of probabilities that you mention in you letter, the industry over the past several years has provided the staff various estimates of the probability of failure to trip the reactor.
The staff has recognized the substantial uncertainties in these calculations and, I
I because of the probabilities calculated, the staff has.continued its efforts to resolve the Anticipated Transient Without Scram (ATWS) issue.
As indicated in the answer to your Question 11, the new proposed ATWS rule is currently being evaluated in light of the Salem events, and this reevaluation will be forthcoming.
In summary, prior to our decision to allow restart of the Salem facility, the Commission conducted a careful examination of the events and the circum-stances associated with them.
Based on this examination, we are satisfied that the safety implications of the short-and long-term actions have been resolved by specific commitments from the licensee.
In addition*, our review of the circumstances leading up to the events of February 22 and 25 led us to conclude that violations of the Salem operating license contributed to the failures that occurred.
As a result, we have proposed to impose a civil penalty of $850,000 on the licensee.
This is discussed. in more detail in the answer to your Question 3.
Commissioner Gilinsky adds:
I do not share my colleagues' confidence that the underlying causes of the problems at the Salem plants have been identified and re so 1 ved.
I must add that the documents referred to by the Commfsslori-were not a sound basis for decision on Salem restart in that they left out some of the most important safety violations reflecting management deficiencies -- as can be seen by comparing these documents with those accompanying the Commission's later enforcement action, which picked up the omitted items.
Si n ce rely, Original Signed By
- ~-
J oJ:i..n F. Ahearna /
Enclosure:
Responses to Questions
~
Nunzio J. Palladino Cleared with all Cmrs.' Offices by SECY C/R.
Ref.-CR-83-74 Cmr. Ahearne would have preferred the following:
- 1) In A to Q12 -- last sentence to have read~!
"This problem apparently occurred oe-cause there were no procedures requi ri ng*anyone to examine, evaluate or interpret the timing of events and there was inadequate or lack of, training in the use :::.ahd.><
- understanding of the SOE printout.
11
- 3) In A to Q14 -- second paragraph change 2) In A to Q14 -- ftrst ~aragraph to have word 11 requiremeilts 1
' to controls" additional sentence, "Additionally, main-tenance conducted on the reactor trip.*
breakers in January 1983.was*: conducted wi::th;1gui'dance. from a Westing-house (Apparatus Services* Division) representative who was also.:
Originating Office:
EDO/NRR
- unaware of the exi stance of the bulletfos NRC FORM 318110/80) NRCM 0240 OFFICIAL RECORD COPY
- , USGPO
- 1980-329*824
The Honorable Joseph United States Senate Washington, D. c.
Dear Senator Biden:
DISTRIBUTION CourtesyyCopy o.* Fischer w/incoming Central File w/inooming
-c.* !'a*rdsh NRC PDR w/incoming W. Dircks Local PDR w/incoming T. Rehm EDO Reading V. Stello OR'3#1 Rdg RP.g*: I.. Administrator OEL~
J. Roe OCA (3)
R. Minogue M. Bridgers (EDO#l2879)
R. OeVoung Program Support Staff, NRR G. Cunningham M. Jambor SECY D.
~Qttingham OPA D. Eisenhut H~Denton I am pleased to respond to you March 4, 1983 letter about the events that occurred at the Salem Nuclear Ge erating Station, Unit 1 on February 22nd and 25th of this year.
Your letter equested information and specific response to several questions concerning the wents; detailed responses are provided in the enclosure to this letter.
The ether Commissioners and I were a o concerned that these events occurred.
We closely monitored the staff 1 s folld~up of the malfunctions at the Salem plant as well as ~he broader implic~tio~s !or the nuclear power industry.
The facts, data and circumstances associated\\~ith the events at Salem have been collected and documented as NUREG-0977. \\rhis information was used by the staff to determine the. safety issues associated\\~"th the events.
These issues were grouped into three areas:
(1) equipment i ues, (2) operating procedures and operator tra~ning and respon~e, and (~) man ~ement.issues. The staff evaluated each of the issues to determine the licensee~ actions necessary to resolve the issues.
The staff concluded, as reported 'in fl eir safety evaluation NUREG-0995, that the underlying causes of the problem(s) w e identified and resolved and, as such, the Salem facility could be allowed to estart.
We concurred with these findings.
Concurrently, an NRC task force ith representatives from. three NRC offices was established to review and evaluate the generic implications.
The event can be characterized as a failure of the s.fety system to automatically shut the reactor down.
However, the operators did i ntify the need for plant shutdown and did manually shut the reactor down such
- at the event itself posed no serious threat to public health and safety.
Howeve we view the failure as a serious safety concern since the automatic system did ~o~ function as expected and, given other plant conditions such as full power, con iderable overpressure
- of the reactor system would have occurred without prompt o erator action.
The "licensee has attributed the cause of failure to be a lack f adequate mainte-nance to a part of the safety system, specifically, the cir it breakers which de-energize the control rods to cause rod insertions and rea tor shutdown.
Additional means to trip the Salem reactors (albeit not as idly) are given in response to YC!Ur Question 6.
Regarding calculations of probabilities that y6u mention in your letter, the Jndustry has provided the staff over the past several years various estimates OFFICE *************************
SURNAME> **********************,.
DATE. ************************
NRC FORM 318 (10-80) NRCM 0240 OFFICIAL RECORD COPY USGPO: 1981"'-335-960
.~ -
"'T 1
7 l?J.Honorable Joseph R. Biden, Jr.
2 of the probability of fail re the substantial uncertainti s probabilities calculated, th ATWS issue.
As indica~ed in ATWS rule is currently being e reevaluation is forthcoming.
to 'trip the reactor.
The staff has recognized*
in these calculations and, because of the staff has continued its efforts to resolve the e answer to your Question 11, the new proposed luated in light of the Salem events, and this In summary, we shared your concern~nd cons.equently, prior to our decision to allow restart of the Salem facility conducted a careful examination of the events and the circumstances associa ed with-them and assured ourselves that the safety implications of the short nd long term actions were resolved by specific commitments from the licensee Palladino
Enclosure:
Responses to Questions OFFICE *****.?!l~~~e.~....... ~~t.P.~:..........
~c~er
- l"lC:isenhut
..... ?.!.~!RR............ ~~ J........,.. **g*~-~...................9.~8..........................................
- ~****:............
SURNAME~........... l/:***:****.*,........................
[)ATE *
- ~21.f'.l.~.~.:.~..... Q~l.... l?.~...
..... HP.gnJ;Qn..... ~~-~-~E.~.~...........~.~?..~.11.ey9.ing........................
..... Q2!..... !.~~---.. ~!.2./.§1........... ~/..... /.§1.......... ~J..... /.8.3........
NRC FORM 318 (10-80) NRCM 0240 OFFICIAL R!!CORD COPY USGf'O: 1961-335-960
4-.
- nlbf Honorable Joseph R. Biden, Jr.
2
- e.
events occurred a d is an example of the defense-in-depth concept utilized in the design and ope ation of nuclear power plants.
In your letter, you ention that this incident demonstrates the extent to which reliance is pla don fallible 11 human 11 factors.
In my view, there must always be some relianc on the human factor since no design of this complexity can account for a 11 pos j bl e occurrences:. That is why nuclear p 1 ants are built using defense-in-depth c cepts and why we place so much emphasis on training, procedures, and managemen *involvement and oversight.
Regarding calculations of p babilities that you mention in your letter, the industry has provided the st f over the past several years various estimates of the probability of failure\\to trip the reactor.
The staff has recognized the substantial uncertainties ~p these calculations and, in spite of the low probabilities calculated, the su~!f has continued its efforts to resolve the ATWS issue.
As indicated in the,'f"nswer to your Question 11, the new proposed ATWS rule is currently being evaluated in light of the Salem events, and this reevaluation is expected to be com~leted by about mid-April.
In summary, we share your concerns ~Qd are conducting a careful examination of the events and the circumstances asso~*ated with them.
We are proceeding in a structured manner and intend to addre s your concerns, as well as others, prior to any restart decision.
\\oi ncere ly,
\\,
Nu~\\ J. Pa 11 adfoo, Chai rm~n U. S. N clear Regulatory Commission
\\
Hq:rno ~
~,m; re ks 0~ f /83 RI: R/\\
OFFICE. ************************..........,....
...... ~/.;~~............... ~!.~'.:~................ Pt~~............ ?t~l.!3?'................ !~.0.~f::QQ........
- ~~2~i.::: ::J~Y::z~L ::J~?::?.:~L :::3W:::i.ii.i.: :::~i.1.::::1.~3:::::: :::J:W=1.~:~::::
SURNAME. *... ~~.~ *.¥.~.~~.......
DATE *
.... fJ.~.f..... l~~....
.NRC.FORM 316 (10-60) NRCM 0240, OFFICIAL RECORD COPY USGPO: 1981-335-960
\\~
-o<-'- JCunni naharn
!'.?De Young MJambor D~*lotti ngham DEisenhut
_ -....-..11*uY1AN The Honorable Joseph United States Senate Washington, D. C.
Dear Senator Biden:
DISTRIBUTION Courtesy Copy Docket File w/incomino NRC PDR w/Jncoming
~
Local PnR w/incoming
- f'IRRJ1l 0d
., \\1.. 7(
.O. (3)
M. Bridgers (EDO-
)
Pro~ram Support staff, NRR JHe1 temes, J\\EOD DFi sch er w/i n*comi ng r.Parrish
!-1'1i rcks JRoe TRehm VStello RDeYoung JCunningham I am pleased to respond to your March
, 1983 letter about the events that occurred at the Salem Nuclear Generatin Station, Unit 1 on February 22nd and 25th of this year.
Your letter-requeste information and specific response to several questions concerning the events; etailed re~ponses are provided in the enclosure to this letter.
~
The other Commissioners and I are also cone ned that these events occurred._
We are closely monitoring the staff's follow*p of the malfunctions at the Salem plant as well as the broader implications>for\\~he nuclear power industry.. The facts, data and circumstances associated with lhe events at Salem have been collected and documented as NUREG-0977.
This ~formation is being used by the staff in evaluating the licensee's actions and ssessing when*a restart deci-sion for the Salem facility is warranted.
Concu rently, an NRC task force.with representatives from three NRC offices has been e tablished to review and evaluate the generic implications.
A plan of action (Salem Restart Status Report) has *een prepared which identifies the issues involved with the Salem events specifical y, along with short-and lorig term actions required of the utility to resolve, hose issues.
Before recommending restart, the staff intends to obtain spec"fic commitments from*
the licensee to complete the short term actions to the taff's satisfaction.
The Commissioners will make the decision on the restart o the Salem facility when we are satisfied that the underlying causes of the pro lem(s) have been identified and resolved.
~
The event *can be characterized as a failure.of the safety system to automatically shut the reactor down.
However, the operators did identify the ne d for plant shutdown and did manually shut the reactor down such that the event 'ltself posed no serious threat to public health and safety.
However, we view the failure as a serious' safety concern si nee the automatic system_ did not function as expected and, given other plant conditions such as full power, considerable overpressure of the reactor system would have occurred without prompt operator action.
The licensee has attributed the cause of failure to be a lack of adequate mainte-nance to a part of the safety system, specifically, the circuit breakers which __
-de-c;?nergi ze the -contra l rods- *to* cause rod insert i ans** arid reactor* siiutaowh~ -
Additiona] means to trip the Salem reactors (albeit not as rapidly) are given
- in response to your Question 6.
Regarding calculations of probabilities that you mention in your letter, the industry has provided the staff over the past several years various estimates of the proba~ility of f~il~re ~o trip the react~r. The staff has recognized OFFIC~~.................. :...... ************************ ********* 111 **************
~************* *****~** **
SURNAME~ *******************;.... ** ********************** ***** *******************. ****************** **.,,
- DA.TE~ ********* 9 ********-******
NRC FORM 318 (10-80) NRCM 0240 OFFICIAL RECORq COPY USGPO: 1961-33&-960
FOR:
FROM:
SUBJECT:
PURPOSE:
DISCUSSION:
RECOMMENDATION:
COO RD INA TI ON :
SCHEDULING:
{l Executiv.e* Director for Operations
/r
/
RESPONSES TO SEN TOR BIDEN'S QUESTIONS RELATED TO SALEM 1 EVENTS
./
f For Chairman's sig ~;t'~re.
This letter provi9fu a response to Senator Biden's letter dated March 4, l.983.
The Senator's letter contained fifteen questions which/are a swered in the enclosure to the letter to be signed by the Cl airman.
/
l I recommend/that the Cti irman sign the letter.
jl NRR, RES,/ Region I Prior t( Salem 1 restart.
I I
J l
/
/
I i
William J *. ircks Executive Di ector for Operations
Enclosure:
t Letter to Senator Biden
Contact:
D. Wi.ginton, Ext. 27354
/J/ofv-1
~mer 3fV/83 EDO WDircks 3/ /83
- =....
e*
~
~esponses To U. S. $enator Joseph R. Biden Jr. Questions (March 4, 1983 Letter
- ..o Chairman Palladino)
Question 1 PleasE provide me with significance.
Resoonse a complete description of the event and its safety Attachment l is a draft Abnormal Occurrence repor~ which provides details of t~e reactor trip breaker failure events at Salem.
NUREG-0977, a.
task force report on the facts associated with the circumstances of the* events h~s* been issued an9.is Attachment 2.
- 'r.'ith r~spect to the safety sign1ficance., the Salem AT'ivS events of February 22 and 25, 1983 posed no serious thre~t to public health and safety because the Salem reactor was _at l.ow power and the operators manually scrammed the reactor soon after the aut.omatic scram signal.
The event of February 22nd was a loss.
~f one operating,feedwater pump at low power. *Th~' event of the 25th was norm~l operation at 12 percent power with low level in one steam generator.
A dis-cussion of the safety significance_,of the events_ had they occurred at full power-is given= in the response _to Question 8.
p **
- .. ~*..
. r;.
1
- ** 1
. ~..
-~
.*Please provide me with copies of all memoranda and internal studies that analyze this event.
Response
There are two efforts underway with respect to analysis of the Salem events.
I am enclosing as Attachment 3, the Salem Restart Evaluation Report prepared by-the NRC staff. This report documents the bases for a restart decision.
Additionally, an NRC Task Force was established to reviev.' and e~aluate the generic implications of the Salem.events.
A report will be forthcoming in the near future. is a listing of additional internal documents which may relate to the events in question.
We will provide you with ~opies as you request.
2
. ~. :_
e*
1
- Question 3 Before restart, we would ask that you provide an assessment of whether NRC has reason to believe that either an operator during the in~ident or management action during or prior to the event acted inappropriately.
Response
An.NRC fact-finding task force was at the Salem site on*March 2-6,.1983, and
.they conducted a review of the circumstances surrounding the February 22 and 25 events.
The results of this review were published as NUREG-0977, dated March 1983.
This and other NRC and PSE&G efforts ~evealed significant deficiencies which contributed to the inoperability of the reactor trip breakers.
These deficiencies involved (1.) failure to adequately ihvestigate previous failures to identify and correct conditions adverse to* quality; * (2) failure to correctly*
inc1ude the breakers on the Master Equipment List (MEL); (3) failure to properly implement procurement procedures; (4) failure to properly.implement; control and distribute the MEL Which contributed to inadequate quality assurance review of procurement and maintenance; (5) failure *to identify and control safety
- related components; and (6) failure to implement surveillance testin~ requirements.
PSE&G efforts to correct these deficiences are addressed in the Salem Restart Safety Evaluation Report.
In.addition, the licensee failed to promptly report, a~ r~quired, certain events to the NRG.
On February 22, the reacto~ trip hreakers fa i1 ed to open automa ti ca 11y upon demand, apparently because of the deficiencies described in Item II of the:enclosed
- N6tice of Violation~ The licensee fa~led to recognize, prior to restart of the reactor on Februar.Y 23, that the reactor trip breakers had failed to open automatically on February 22.: As a resu1t, the reactor was operated for three
.additional days during which time the reactor.protection system could not be considered operable.
_=
The Commission has concluded that these_coritributors to.the events of February 22 an.d:.25 are,the result of insufficient manageTl!ent involvement in establishing a :safety perspective, in requiring 'attention to detail, and in ensurina procedural*.
adherence.... Furthermore, fhe Commission has determined that tnese contributors to the.events of February 22 and 25 are as significant as the events themselves. Accord-ingly, the NRC has pro:>osed imposition of civil oena.lties in the amount *of $850,000.
With respect to operator actions, the NRC staff review has *determined that the operators responses to both events were.satisfa~~~tY however, the post-trip review was*=;nadequate.
3
-~
t
. *. ~..
e*
Due st iJ.)n 4
.)
rn* this event, manual shut-down was achieved some.twenty.-four seconds after automatic cont~ols and ~ack-up failed; are there incidents of this kind where 30 seconds would not have provided for adequate public health and safety?
Resoonse Manual shutdown of the reactor in 30 seconds following any anticipated transient will *provide adequate protection for*public health and safety.
A more extensive discussion of the limiting anticipated transient with delayed r.eactor trip or failur~ Df reactor trip is given in response.to Question 8 below.
It should be noted that backup to automatic controls is a manual sh0tdown.
Since the plant was manually shut down, the:backup did not fail.
'.. ~ "
4
- --i
_J
Question 5 e*
I I.*am inrormed that.this type of event was calculated in the Reactor Safety Study (WASH 1400) as having an extermely low probability-.
- Wr.at does the NRC currentlv calculate the probability of this type of event?
How was it that the probability of occurrence was repeated in a three day time period at Salem I?
Resoonse In WASH-1400 a p~essurized water reactor designed by Westinghouie was analyzed.
The median probability of SCRAM failure was estimated at 3. 6 x 10- 5 per demand with an upper bound of 1 x 10- 4.
The frequency of ATWS is the product of the frequency ~f anticipated transients requiring scram a0d the probability of scram failure on demand.
With approx_i mate ly 10 scram demands per year the me di an estimated ATWS frequency would be 3.6 x 10- 4/year with an upper bound of approx-imately 1 x 10- 3.
However, only a fraction of ATWS events would result in reac-tor core damage.
For Westinghouse designed reactors, most are expected to be relatively mild.and controllable, as was the case at Salem, a plant of Westing-house design.
In addition it should be noted that the Salem event involved a failure to autom~tically ~cram, but manual scram worked as designed.
The NRC staff has been using an estimated sc*ram failure probability of 3 x 10- 5 per demand for value impact analyses being done as part of the ATWS rulemaking
. activities~ Consideration of the Salem event would increase this estimate by abciut a fa~tor of two.
While this approach makes it appear that all react~rs have the same likelihood of failure to scram upon demand, this is an over-simplification.
There are substantial uncertainties in these calculations,
- and experience indicates the potent i a 1 of a wider range of probability from plant to plant than might be inferred by the estimation of uncertainties in probability studies.
This could be due to variances in design and opera-tional factors (e.g., maintenance procedures and operations quality assurance which are important to the reliability of the reactor protection system).
The staff is aware of scram failure ~recursors which have occurred at a rate qf about. l x 10- 4 per demand.*,This is reasonably.consistent with the Salem event:. However, the Salem event raises the concern that the median scram failure probability may be higher than the value currently used in the generic ATWS rulemaking value impact analyses.
In light of this event, we are reassess-ing the ATWS rule proposals and technical bases.
The* incident on February 22 '" 1983 was very unusual in that the operator manually scrammed the reactor within a few seconds of the automatic trip signal.
A qufrk review of the incident on February 23, 1983 by the licensee led them to the erroneous conclusion that the reactor had scrammed automatically.
- Thus, the plant was restarted-on the premise that the reactor protection system was completely functional,* and no repairs were made.
After the February 25, 1983 incident, a close examination of the plant computer records from the February 22
- event showed that the automatic scram did not result in automatic opening of the reactor trip breakers.
Although the sequence 6f events on the computer printout shows that the automatic signal was received first, additional eval-uation was necessary to identify that the trip breakers* in fact responded to the manually initiated signal.
5
e*
Qi,iestiqns 6 What sequence of events would have followed the failure.of a manual SCRAM, both within an~ outside the plant gate?
-*Resoonse
- a.
Onsite Actions:.
- '~..
In accordance with Salem emergency operating procedure~. the actions to be taken by plant operators in the event that the plant fails to automatically trip (scram) on demand, and the manual scram also fails, follow below:
open the' reactor trip breakers manually by depressing either of the 11open 11 push buttons located in the control room, fo_r both reactor trip breakers..
tri~ *the tufbine by using the trip handl~ on the control room console.
(A turbine trip also provides a signal to the reactor protection system
.to,trip,the rea-ctor tr:ip _breakers.)*
manually initiate a safety injection from the control room.
open.the reactor trip br~ak~rs manually by depressi~g the p~sh button physically located on either reactor trip breaker.
manually trip both rod drive motor-generator (MG) sets at their local control panel; these can be tripped by opening either the power supply
~ase:electrical power to the.control rods is removed and-scram.occurs.
- Both the* MG set breakers.and the reactor trip breakers are _locat~d in a*
- switch gear room two floors below the control room.
It should take about 1 ~minute to go. from the control room to the switch gear room.. In addition
':to the procedural steps in 'Pl ace at the ti me of the event, there is the ability to deenergize the control rod drive MG set power supplies from the*
- *"**control'. rOom*.~ Revised =procedures since -the :event also include this addi-
, : tional*step.**
- b.
These actions are taken.in sequence and are progressive steps to accomplish
- the function of either *inserting the control rods into the core by gravity_,
reducing rector power or injecting boron in high concentration to shut down the reactor.
'Offsite' Acfions
- In the e~ent that a s~ram sianal existed and neither an automatic nor a manual scram o~curr~d (indic~ting a failur~ of the reactor* protection system) PSE&G 1 s Emergency Response Plan calls for declaration of an Alert Condition.
The purpose for declaring this condition is to ensure.that emergency personnel are readily available, in the event plant conditions degrade.
This condition requi.res noti,ication of Federal, State and local agencies, activation of site support centers and call-out of designated emergency response personnel.
6
_Qt!estiorr 7 Mhat other backup system would have been available if the manual SCRAM would have been ineffective and/or incomplete?
-Response The other backup system and/or operator action available in the event of a manual scram failure are delineated in response 11 a 11 to Question 6.
~"
.:. ~
'.. _:*/:
I.
~.
7
~.*
- 1 *,
i
Question 8 At the time of the event; the re~ctor was reported to be operating at 12% of its rated power.
Of what consequence and severity would the failure of the automatic system have been had the plant been operating at near or full power? Specifically, what other problems besides core endangerment might have occurred? What differences would have been relevant in operator reaction time?
Resoonse Befote answering your specific questions it is important to note that in themselves, the Salem An~s events of February 22 and 25, 1983 were not serious threats to public health and safety.
The primary reasons were that the reactor was at low power in both events and the operators alertly scrammed it manually without* depending on the automatic scram.
However, the events are important in that they are indicators or early warnings of more serious ATIJS accidents that could occur.
The Salem events would have been more severe if (1) they had*occurred when the reactor was at full power; (2) the initiating event had been*more severe*but within the range of anticipated events for this station; and (3) there were either human errors or additional equipment failures.
For the four-loop Westinghouse PWR design, of which Salem is typical, there have been many previous analyses of severe ATIJS accidents in connection with the An.JS rulemaking (Unreso:lyed Safety Issue A-9). *We will rely on these analyses performed *for the composite four-1 oop Westinghouse P~JR design in
-answering your specific que~stions.
To begin with, assume that the reactor were operating at full power soon after startup following a refueling (in the case of Salem 1 in February, the reactor was asceriding to full po~er after a refueling outage).
If the reactot wer~ to then experience a complete loss of main feedwater (this would be comparable to the event on Feb.ruary 22, which involved the loss of the one feedwater train that was operating) and if the automatic reactor scram fails entirely (this was the case at Salem), then the automatic turbine trip would also fail (as it did at Salem).
Loss of feedwater under these.conditions results in a large mismatch between the heat generation and the heat removal rates for the reactor
. coolant system because the secondary system can no longer.remove all the heat generated in the reactor core.
If the operator manually scrams the reactor within 30 seconds (as was the case at Salem), there would be no appreciable heatup or pressurization of the primary system and no serious threat to public safety.
If the manual scram were delayed, then there would be an increase in reactor coolant temperature and a decrease in coolant density, producing a surge of coolant to t~e pressurizer which increases the pressurizer level and the system pressure.
The pressurizer relief and safety valves would open to limit pressure bui 1 dup. __ -Steam genera tor inventory would decrease, as the result of boi.loff :With no replenishment by feedwater flow~ and further reduce heat
- transfer from the primary to the secondary system.
Operator actions to cause reactor scram within about l to 1-1/4 minutes after loss of feedwater would keep the system pressure near the normal operating pressure and there would be no damage to the reactor. If a manual scr(\\.m were delayed slightly more, but accomplished within about 1 and 1/2 minutes, Service Level C (about 3200 psi) would not be exceeded and it would be expected that emergency 8
e*
co)~e _c_o~ling could be established.
After that time, scram would do little to reduce the peak pressure, but it would assist reactor recovery.
For the typical four-loqp Westinghouse plant, a peak pressure of* 3650 psi is estimated if there is no*manual scram.
For conditions specific to the Salem plant, Westinghouse recently calculated that the peak pressure is only 3200 psi.
P~essures abpve Service Level C increase the likelihood of permanent deformation of valves needed to actuate emergency core cooling neecfed for recovery of the plant.
Some key assumptions in this severe accident wer~ as follows:
(1) a moderator temperature coefficient is minus 8 x 10- 5 ~K/K°F, a value that is not exceeded 95% of the time, (2) no credit for turbine trip, and (3) credit for the normal capacity of the pressurizer power ope.rated relief valves (i.e., they are assumed to be unblocked).
If the turbine were to be tripped by the operators (in ac-cordance with procedures at Salem) at 30 seconds into the event, the maximum system pressure would decrease about 950 psi even if there were no manual scram.
The result would be a peak pressure of about 2700 psi, well within the capability to establish emergency boration and core cooling.
There should be little or no fuel damage in this case. __ This is the rrost ]ikely event if there were to be a complete loss of scram capability (both manual and automatic).
If both power operated relief valves' (PORV) were blocked, there would be an increase of about 300 psi in the maximum* pressure of 3650 psi for the severe loss of feedwater *ATWS with no turbine trip. If the turbine is tripped, the blockage of both PORVs would increa~e the peak pressure from.2700 psi to 2950 psi.
Blockage of only one PORV (as was the case at Salem on February 25)
~ould increase. the pressure about half as much as blockage of both PORVs.
If only one of the two normal feedwater trains were lostJ a complete failure to scram.. from full power with no operator action should tesult in a mild pres-sure transient and no fuel damage (Westinghouse calculates_ a peak p~essure ~f 2330 psia for Salem).
If auxi 1 i ary f eedwater is initiated by the o_perator earlier than the* 60 seconds assumed in the severe ATWS analysis described above, there would be little decreas~ in the peak pressure (AFW from two of two tfains is equivalent to about_8% of full power).
Verification of initiation is called for in the I
Salem procedures.
. Design changes are being considered in connection with an NRC rulemaking that would reduce the likelihood of ATWS events in the Westinghouse desig*n by the decreasing the reliance on the manual scram addressed by your questions.
They involve the diversification of the present breaker design for interrupting power to the contro 1 roes.
Such a change, in the case of the Sa 1 em event, would have eliminated the need fo~ manual scram.
The rulemaking also considers practical changes that should reduce the consequences of ATWS events in the Westinghouse design.
These are the provision of diverse, automatic initiation of both turbine trip and auxiliary feedwater.
It can be seen from the discussion above that the automatic turbine trip femoves the need for' the operator to manually trip in the first two minutes of an extreme ATWS event to avoid exceeding Servic\\ Level C (about 3200 psi).
9
~Yhe Wa~tinghouse analysis for Salem is attached (Attachment 5).
Except for sm~ll differences in the pressure values due to the design details for that p)ant, the Westinghouse_analysis generally agtees with the staff description provided above (which d~rived from earlier generic analyses by Westinghouse).
~'.. '..
- 1::
. ' ~
10
Ooesti~n 9 Are there other initiating events (i.e., besides low steam generator water le~e1) in which 'operators would have had less time to respond?
Resoonse The most limiting an~icipated transient combined with delayed reactor trip or failure of reactor trip is the loss of feedwater transient which leads to low steam generator water level.. This transient is discussed in the answer to.
Question 8.
~...
11
The Preliminar;.Notification of Event or Occurrence notes that arr alert was
belatedly declared.
11
\\{hat was the cause and effect of this delay?
Given the relative urban proximity of the plant, of what consequence would this delay
-have been?
Resoonse Fellowing the manual reactor trip on February 25, 1983, the operators' *atten-tion was* first devoted to placing the plant in a stable condition, which was achieved* within a few minutes.
At this point in time, there was uncertainty in the minds of t_he -operators* as to whether the reactor trip alarm wa~ a valia signal.
. Personnel from' the Inst~umentation and Centro! ()&C). Department wer_e_ dispatched to examine' the annunciators~ instrumentation, and protection system' circuitry.
The shi.ft SL!pervisors wait.ea until.they w.ere sure that there hQ,cl been ~
- failure of.the automatic reactor trip system to function properly, as_deter-
- mi ned by,,the.I&C testing,._ before declaring an Alert.and making the associated notification... This delay in classifying the event as an Alert-had no con-sequences Jor the surrounding population.. Per the Sta ti Of! EmergencyProcedures and Federal,Regulations,:there are-four.~lasses of emergency action levels (EAL).
These. are:
-Unusual Event, Alert~ Site Area Emergency, and General Emergency.
The rationale for the notification associated with Unusual Event
~nd Alert classifications is to provide-early and prompt notification of minor
-events which could lead to more serious -consequences given.potential operator e~ror*or equipment failure,: or which might be indicative of more serious con-ditions which are not*yet-ful*ly realized.-
A gradation 'is provided to a*ssure full er respons*e preparations for more seri otis *i ndi ca tors.
Events
~ nvo 1 vi ng more serious:*pl ant degradation would include other contro 1 room i ndi cations of
'reactor and plaht parameters or radiation levels that Should have enabled the
.* -:operators to promptly. c1_assi fy the event ; n accordance with the predetermined '
EALs appropriate* for the. situation.
These would also require proper *.
notification 'upon reclassification~'
~-~.
- .~
,!'.-* ~-
i*;*_!_
Tf1e. f ai] ur_e. of. _the }eactor'.tri p breakers'. repre~enfed" ari actual, substant i a 1 degradation pf. the
- 1eve1 o"f safety of' the p 1 ant in that an important safety
?YSterri. had-fafl ed -to.* operate as designed.
- By definition, this". event is' cl as-
'si fi ed as an °Alert" even tho'ugh the reactor was in a stable, safe condition.
- By definition, for an event classified at this _level* it is unlikely that an offsite hazard would evolve and a necessary prerequisite for such classifica-tio.i:i is a determination that the.situation can be corrected and controlled by the pl ant s faff.
- However, as-a precautionary step for Alert-type events, advisory level notifi-cations to the emergency response organizations of Federal, State, and local authorities are made.
Thus, the only offsite effect of the delay in classify-ing this event as an Alert was a delay in making such offsite advisory level notifications.
12
e*
Please detail t~e history of resolution of the unresolved safety issue of Anticipated Tr~nsient Without SCRAM and state the significance of this event in its eventual resolution?
Response
The possibility of a transient with the inability of the _reactor protection sys-iem to function was.first raised in the late 1960s.
The reactor manufacturers performed studies and submitted them to the Atomic Energy Commission, Regulatory Staff. in 1970-1971.
The Regulatory Staff issued WASH-1270, 11Technical Report on Anticipated Transients Withoui Scram for Water-Cooled Power Reactors in September 1973, that contained a_ Licensing Position on ATWS.
The reactor*.manufacturers felt that the.costs to implement the Li_censing Posi-tion were-too expensive for what was considered to be a verylow probability event;. The AEC*Staff and then the NRC Staff continued to evaluate the proba-bility and consequences ~f AT"h'S with more studies supplied by the reactor manu-facture~~ and by the Electric Power Research Institute (EPRI).
In 1978-1980, a four volume Technical Report, NUREG-0460, 11 Anticipated Transients Without Scram For Light Water Reactors, 11 was* published with recommendations for re sol u-t ion of ATWS.
A proposed rule was presented to the Commission in October 1980 (SECY-80-409) to have each applicant perform an evaluation of their plant with respect to
- prevention and mitigation from ATWS.
A Utility Group of 20 electric utilities was formed in the summer of 1980, because they felt that the NRC requirements would be prohibitively expensive.
The Utility Group submitted a proposed rule on September 16, 1980 (PRM-50-29) to install certain har~ware on pl~nts by vendor type as a r_eso l ut ion to this issue.
At about this same time Commissioner (and later Chairman) Joseph Hendrie was searching for a new approach to resolve the_. logjam between the NRC staff and the.industry.
A plan to develop and implement a program for relia.bility as-surance, plus prescribe certain hardware fixes, was presented to the Commiss.ion on July 16, 1981.
The Commission voted unanimously to publish the 11 Hendrie 11 rule and the 11 Staff Rul e 11 (based on SECY-80-409) for public comment i*n the Federal Register and to include the Utility Rule as a third alternative.
Thirty-nine public comments were received, with the majority of the comments recommending no rule or preferring the Utility Rule.
The NRC Staff presented a revised* plan in SECY-82-275 to the Commission on July 13, 1982' to resolve ATWS by-forming a Task Force and Steering Group.
The proposed Task F~rce recomm~ndations were presented to the ACRS on October 22, 1982 and t6 the Committee to Review Generic Requirements on November 3, 1982 and again on January 25, 1983.
At the time of the Salem 1 incident the new proposed rule was about 95 percent complete.
Currently, the rule is being evaluated in light of the Salem event to determine whether any changes are warranted~
13
Question 12 The rep'ort.of this event indicates really two separate events; was the restart of the plant after the first event adequately* and properly justified?
Response*
In retrospect,.it is clear that 'restart after the *February-22 event was not adequately nor properly_ justified.
The February 22 *event~ the first event*, involved several problems including the deenergization of a 4-KV bus, a loss of control power and* indication for the operating *main ;feed pump, loss of a reactor-. coolant pump; arid normal*
.lighting. As a result of these factors, control over *the steam generator feedwater system was lost. The control room supervisor apparently recognized
- .the situation. and.ordered a manual reactor trip at :about the same time* that the reactor protection system called for ~*trip. Based on interviews with plant operators, ~it was not apparent to the operators that the reactor.had
'failed, to trip upon receipt of 'a low-low *steam generator water level signal.
Based on available informatiorl'it appears that the'operators preoccupation with *the numero't.is. other alarms in the control *room and a problem with the manual scram switch'may explain why they failed to notice the automatic trip
- failure.
The manual reactor triR was in fact inserted 3.6 seconds after the reactor.. :Protection system called for. a trip. * *
... :.) ' :
~-..
~
A.Post :event.. review*:*cff the-:Fehr*u~ry 22".eve~t was conducted by plant staff 'in.
an attempt to determfne what*:had.occurred and. to resolve* any.equipment problems detected. The sequence of events (SOE) computer printout was the best* evidence
. available which co~ld have revealed that the reactor trip breakers had failed
- ....
- to open-.-when the reactor protection system called for a reactor trip. The SOE printout was examined.by plant.staff members but no attempt was made to sort out the precise* timi.ng of each recorded event and therefore, it was apparently
- not recognized that the reactor trip breakers had*f~iled to* open from the low-low steam generator water level signal, a reactor protection system (RPS) signal.* The individuals reviewing the event concentrated on the othe*r *problems identified. above *. Specifically,' the_ information provided. by the plant computer was* apparently used* only* to* verify.the sequence of events and *not the time -.
intervals between. events.. Later in *the afternoon of February. 23, the Assistant General Man.ager of the station, convinced that the problems.of the previous.day*were.understood and corrected, g~ve approval to restart the plant
- In retros.pect, it is apparent that the post trip review of the *February 22 event
- was not conducted in sufficient detail to disclose the malfuntion incurred by the reactor trip breakers since the information {i.e., the computer SOE printout) was available and if properly reviewed, would have revealed that the reactor trip system had malfunctioned.. This problem apparently occured becaused the.
existing.procedures for post-trip review did not expliCitly require anyone to examine, evaluate or interpret* the timing of events.
14
.1
,,,._ \\~.
- Question 13
.. ".J.
WRat circumstances explain how the undervoltage_ trip breakers which are considered
- 11SafetyGrade Components" could have been *mislabeled during recent maintenance?
-Response The mislabelling, or incorrect classifications, of the maintenance activity as 11non *safety-related" was due to personnel error coupled-with.inadequate administrative reviews.* In. this case, the classifier,* by ins.truction*, should.
have contacted the* Engineering Department for the* classif*i cation; he did not *
. As a result, there is a need to better understand the man_agement controls that allowed this situation to* develop.
As can -be seen in the staff's Restart
- Evaluation Report, this specific item was considered in resolving the Management Issues.
- This issue demonstrates the need to examine more carefully the management control systems (procedures, audits, etc.) associated with maintenance activities.
for *other. than the reactor trip breakers.
-~. :.
~. - '*. *-
,. *~..
15
. ~
_,.~
- ~
-~'..
Question 14
'How i~. it possible that company officials were unaware of a 1974 safety circular from the vendor explaining special maintenance procedures?
Response
The licensee has indicated that they were unaware of the_ existence of the
. vendor 1s (Westinghouse) 1974 technical service bulletins that provided preventive maintenance recommendations for the reactor-trip circuit breakers.
There were no administrative requirements to ensure proper distribution and adherence to vendor technical bulletins. Therefore, even if Salem had received the circular,
- they might not have followed.it.
Westinghouse has established an interdivisional task force to review current methods for distribution of technical information within the Corporation and methods for dist~ibution of this information to utilities. Additionally, Westinghouse recently has provided (after the February 25 event) the Salem
- station with all technical information for equipment supplied for Salem.
Ap-
. parently, communication problems exist between the Nuclear Services Division and other Westinghouse divisions..
Finally, it should be noted that Salem had in their possession vendor manuals for
. breakers which recommended a preventive maintenance program.
The recommended maintenance was never implemented, however, from the time the breakers were installed*in 1976 until January 1983.
-~ :
t'
- ~*,' --
- ' ~ - '
.. ~. *.
. ;., ---.. - **..... -~*-... ;*-. *.
16
..... ~
O;Jes ti ~n 15 What remedial steps can the Commission outline that woul"d prevent present and future notices from vendors going unimplemented?
_Response The NRC has established a Task Force to review and evaluate the broader impli-cations of the Salem event.
The report from this task force is forthcoming.
'The issue of vendor notifications to nuclear utility customers is being
- addressed.
' : ~..,...
.. ',.. ~.
17
ATTACHMENT 1
(
DRAFT ABNORMAL OCCURRENCE Reactor Trip Breakers Failed To Open On RPS Trip Signal Date and Place:
On February 25, 1983, Public Service Electric and Gas Company reported an event at Unit 1 of the Salem Nuclear Generating Station, a Westinghouse designed, pressurized water nuclear power plant located in Salem County, New Jersey.
- Nature and Probable Consequences:
At 12:21 am on February 25, 1983, a low-low water level condition in one of the four steam generators initiated a reactor trip signal in the Reactor Protection System (RPS).
The reactor was at 12% rated thermal power at the time preparatory to power escalation after a recently completed refueling outage.
Upon receipt of a valid reactor trip signal, the reactor trip circuit breakers which supply power to the react6r control rods failed to open (opening of either circuit breaker would have caused the reactor to trip). About 25 seconds later, operators manually initiated a reactor trip from the Control Room.
The reactor trip circuit breakers opened as a result of the manual trip signal and this resulted in insertion of a 11 contro 1 rods and shutdown of the reactor.
Fo 11 owing the manua 1,
trip, the plant was stabilized in the hot standby condition. *All other systems functioned as designed.
Later that morning when the cause of the failure had *been determined by the licensee, the plant was placed in cold shutdci~n at the request of the NRC.
Investigation of this incident on February 26, 1983 by the NRC revealed that*a similar failure occurred on February 22, 1983, at Salem Unit 1.
At 9:55 pm on February 22, with the reactor at 20% power, operators were attempting to transfer the 4160 volt group electrical busses from.
the station power transformers to the auxi 1 fary power transformers, a*
routine evolution during power escalation.
During the transfer attempt, one of the 4160 busses deehergized resulting in the loss of one reactor -
coolant pump and power for the operating main feed pump control and indication.
At 9:56 ~m, a low-low level condition occurred in one steam generator (due to the loss of the main feed pump), initiating a reactor trip signal.
Due to the abnormal conditions created by the lo~s of the 4160 volt bus and in anticipation of loss of steam generator water levels, the opera~or was directed at about the same time to manually initiate a reactor trip.
It was understood by plant personnel and was reported to the NRC that the automatic reactor trip signal due to the low-low water level in one steam generator had, in fact, caused the reactor to trip.
On February 26, 1983, as a result of NRC queries, the sequence of events computer printout for February 22 was again reviewed and it revealed that the reactor trip breakers actually opened in response to the operator 1 s manual trip signal.
Consequently, it is now evident that on February 22 (as on February 25)~the two reactor trip breakers 1
, I
- f.
failed to open upon receipt of an automatic trip signal from the reactor protection system.
Since the operators initiated a manual reactor trip shortly after receipt of the automatic trip signals on both February 22-and February 25, no adverse consequences occurred and the reactor was in a safe condition.
Cause or Causes:
On February 25, approximately 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> after the event, the cause of the fail~re to trip was determined by licensee instrumentation technicians to be failure of the undervoltage (UV) trip mechanism associated with each 6f.the two reactor trip circuit breakers to function as designed.
The UV trip mechanism consists of a relay and attached mechanical latches; upon receipt of a trip signal from the Reactor Protection System (RPS) the UV coil is deenergized and the mechanical latches cause the trip breaker to open.
Opening of either circuit breaker causes a reactor trip.
(A manual trip signal operates both the UV trip relay and a separate shunt trip relay within each breaker.
The shunt trip relay is energized upon a manual trip signal.
Either relay is designed to cause the circuit breakers to trip; and in the February 22 and 25 events, it was the shunt trip relay which actually caused the reactor trip breakers to open.)
The failure of the UV trip mechanism was determined by the licensee and the vendor, Westinghouse, to be excessive friction on a mechanical latch lever in the UV trip mechanism.
The cause of the excessive friction is still under investigation.
The circuit breakers are Westinghouse Type DB-50.
. Previous failures of a reacto~ trip breaker have occurred.
Following a DB-50 reactor trip circuit breaker malfunction at the H. B. Robinson Nuclear Power Station in 1973, Westinghouse issued Technical Bulletin NSD-TB-74-1 in January 1974 recommending certain periodic maintenance
- measures, including lubrication, to improve the reliability of DB-50 breaker_s.
In February 1974, Westinghouse issued a 1 etter (NSD.DATA tETTER.74-2) which, among other things, specified that a dry or near dry molybdenum disulfide lubricant should be used in the UV trip mechanism.
- I~ appears that no preventative maintenance was conducted on the Salem Unit* 1 DB-50 circuit. breakers until January 1983.
Additionally, the lubricatinn recommendations of the Westinghouse 1974 Technical Bulletin and Data Letter were not implemented during the January 1983 maintenance, sine~ personnel performing the maintenance (including a Westinghouse service representative) were not aware of this information.
There have been two previous events at Salem Unit 2 involving a failure of one reactor trip circuit breaker to trip.
On January 6, 1983, a reactor trip occurred due to a low-low water level condition in one steam generator and only one reactor trip breaker operated.
The second trip breaker finally opened 25 minutes later, although the reactor had already tripped from opening of the other reactor trip circuit breaker.
The failure of this trip breaker was concluded by the licensee to be due to dirt and corrosion interfering with proper operation of the UV trip mechanism.
As a result of this event,~maintenance was conducted on all 2
I
.J.
Unit 1 reactor trip circuit breakers in January 1983, under the supervision of the circuit breaker vendor, Westinghouse.
All breakers were satisfact-orily tested after maintenance.
Licensee Event Report (LER) 83-001/03L dated January 27, 1983, provides further details of the January 6 event.
On August 20, 1982, during surveillance testing of the Reactor Trip System on Salem Unit 2, one reactor trip breaker would not trip.
The cause of the breaker malfunction was concluded by the licensee to be failure of the UV relay coil.
The affected coil was replaced, and the breaker was satisfactorily tested.
LER 82-072/03L, dated September 8, 1982, provides further details of the August 2 event.
Actions Taken To Prevent Recurrence Because of the generic implications of this issue, the NRC issued IE Bulletin No. 83-01 on February 25, 1983 to all pressurized water nuclear power plants to inform them of this event.
For all pressurized water reactors having DB type reactor trip circuit breakers using UV trip attachments, certain actions were required.
These actions included prompt surveillance testing of the breakers, ensuring that preventive maintenance programs on the breakers include the recommended Westinghouse program, and reviewing with operators procedures to be followed in the event of a failure of the reactor to trip on receipt of an automatic trip signal.
With respect to Salem, the NRC staff met with the licensee at the site on February 26 and in Bethesda on February 28.
The licensee has proposed certain actions with respect to these breakers incl~ding implementing quality assurance requirements, augmenting surveillance test requirements, developing a maintenance program, incorporating the Westinghouse recommen-
. dations, and revising procedures to require the operator to employ a manual trip whenever an automatic trip signal is received.
The NRC is reviewing these actions to determine whether they are sufficient to correct
- the deficiencies.
An NRC task force has been assigned to review and evaluate the implications of this event.
A Region I task force was assigned to collect facts and data on-site to provide the bases for the generic review: *Additional
- corrective actions may be required at Salem and at other power reactors as a result of the task force review.
r.,
3
e e
/Htl~Me~T.:2. *.
- 1
-1 NRC Fact-Finding Task Force Report.i on the-A TWS-Events at*
Salem*. Nuclea(Gf3nerating Station,***.****.*.*.
. Unit.1, on*. February 22 *and 25,
- 1983* <..
~*.. __... ~*-... *...
. *:."*. *?~i
~ '
..*.:. - ~-1
.* 1*:
.*.U.S.. Nuclear Regu_latory~..,-
- Comrilission
- __,*
- Regio*n 1 *Task Fore.a -* - _.--~./.:
~-- -
. *. *This title su'persedes-th~- -title of earlier-d~afts, *. : -
- which refers only to the February 25, '1983 event. - _*
- References to the earlier title in the text and
. bibliographic. data sheet have not been.*... *.*
- corrected in this printing.. * *
-r
. _ _. ' - ~..
~* *: ~. ': --..
-~.. : '
- '* : ~.-- -.. : : _* -~ *-.*
,* -,;::::::1
. - *.. *-:*,-***-1
- .. :*:.;_: ~*>-~j'
~ ' *. ;. :... :-* *-:
- '_-~:'.~_:_:~-:;:::Vfa-~~0:*;{,:~(~~}:\\.x/:~ i*
... : _-*~'.;_.
--:~:",:,:<~j/~L:-::~~~'.*'*, -, __ --:*-:*
.* :-.':-* :-_ -*. *-*= ;'~.:;-
- .. I *.
.... ~
.. -*. :.-"- -~.. -'.
.-;~-~-~--......
.. :~*-- -'-~* /. ~f I
~---~~~~--- - :.*:* *_.
. *:.*_.. *.*:--*...-*~*
l' e:.. : ::--
.. -.. ~ *~*
.* +
NRC SAFETY EVALUATION RELATED TO PLANT RESTART..
PUBLIC SERVICE ELECTRIC & GAS COMPANY SALEM NUCLEAR GENERATION STATION UNIT NOS. 1 AND 2 DOCKET NOS. 50-272 AND 50-311 I\\*
- Date:
April. 28, 1983 I.,
ATTACHMENT 4 SALEM DOCUMENTS
- 1.
12/17/71 Westinghouse p~blication NCD-ELEC Repl~cement of Undervoltage Attachments on Breakers in Reactor Trip Switchgear.
- 2.
12/2/77 Document from PSE&G, re Project Directives.
- 3.
12/16/81 Operating Reactor Events Briefing.
- 4. 1 2/22/83 Regional Duty Officer Log *
- 5.., 2/22/83 Letter, Midura (PSE&G) to Haynes, re Reportable Occurrence 83-005/03L.
- 6.
2/23/83 U.S. NRC Region I Morning Report.
- 7.
2/24/83 Region I.Morning Report..
- 8.
2/24/83 Regional Duty Officer Log.
. 9..
2/25/83 NRC Duty Officer Log~ re initial notification of NRC of Salem Event *
. 10. 2/25/83 Reg~on I Morning Report
- 11.
2/25/83 Operations Center Cbmputer:Log of 2/25/83 Event~
. 12. Preliminary Notications of Unusual Occurrence, PNQ..;I-83-10, PNO-V-83-08,.
PNO-IV-83-07.
., * :r3*~* :2/25/83 *L~tter ~ 'Midura (PSE&G) to Haynes re Reportab1 e Occurrence 83-011/0lP.'
- , ::* ~-* -*:...
- 14. 2/25/83 Preliminary Notification PN0-1-83-11, re Reactor Trip Breakers Failed to Open *on Trip.Signal (ATWS).
15-.
2/2.5/83* Preliminary Notification PN0-1-83-llA, re Reactor Trip Breakers
,~
- Failed to' Open on Trip Signal (ATWS).
- 16.
2/25/83 U.S. NRC IE Bulletin No. 83-01, re Failure of Reactor Trip Breakers* (Westinghouse DB-50) To Open on Automatic Trip Signal.,*
i7'; Slides presented by licensee (Public* Service Electric* and Gas Company) during 2/28/83 meeting with NRC staff.
- 18.
2/28/83 Letter, Hal B. Tucker (Vice-President, Nuclear Production, Duke.
Power COmpnay) to Denton.
. *1
2
- 19.
2/28/83 U.S. NRC Regional Daily Report, re Region II and III reported plants that are affected by IE Bulletin No. 83-01.
- 20.
2/28/83 Memo, Dircks to Denton, re Evaluation of the Implications of the Salem Unit 1 Event.
- 21.
2/28/83 Letter, Midura (PSG&E) to Haynes, re Reportable Occurrence 83-012/0lP.
22~ 3/1/83 Letter, R. Uderitz (Vice President, Public Service Electric and
- i Gas Company) to Eisenhut, re Reactor Trip Breaker FAil4re No.,1 Unit Salem Generating Station Docket No. 50-272.
23: 3/1/83 Letter, E. P Rahe (Manager, Nuclear Safety Department, Westinghouse) to Denton.
- 24.
3/1/83 Letter, J. J. Shephard (Chairman, *Westinghouse Owners Group) to Denton, re Salem RT Breaker Incident.
2s: 3/1/83 Note, J. T. Beard to Aolahan/lppolito, re Breaker Failures at Robinson.
- 26.
3/1/83 Note J. T~ Beard to Holahan/Ippolito, re LE~ Search/Review.
- 27. J/1/83 Letter, H~ B. Tucker (McGuire) to Denton, re Additional Information on reactor trip circuit breakers.
- 28.
3/1/83 Memo, Dircks to Commissioners, re Salem Unit Event.
- 29.
3/2/83 Corrrnission briefing slides on Salem Event of 2/25/83, pre-sented* by Eisenhut 3/2/83..
- 30.
3/3/83 Board Notification 83-26, re Failure of Reactor Trip Breakers to*Open in Trip Signal.
- 31.
3/3/83 Task Interface Agreement No. 83-20, indicates certain NRC responsibilities regarding evaluation of Salem 1 restart.
- 32.
3/3/83 Questions provided to PSE&G thru Fisher (P.M. Salem 1/2) concerning Sal em 1&2 Q-L"i st.
'"33.
3/3/83 Merila, Stello to Eisenhut, et. al., re 1E Bulletin No. 83-01:
Failure of Reactor Trip Breakers (Westinghouse DB-50) to Open on Automatic Trip Signal (Dated February 25, )..983).
- 34.
3/4/83 Note, Capra to Mattson, re Comments Regarding Items Included on Salem 1/2 Q-List.
- 35.
- 3/4/83 Memo, Houston to Knight, re Propbsed Scram Breaker Test Frequencies at Salem Unit 1.
- 36.
3/4/83, Page from 11The Energy Daily 11
- 37.
3/4/83 Memo, Houston to Lainas, re Salem 1 Restart SER, Definition of Safety-Related OSI (ICSB) Input.
- 38.
3/4/83 Memo, Houston to Knight, re Proposed SCRAM Breaker Test Frequencies at Salem Unit 1.
39*~ 3/4/83 ~~mo, Silver to Mattson, re RRG meeting on Salem Ev~nt *.
. i
- 40.
3/5/83 Telex from Lee Catalfamo.
- 41.
3/5/83 Starostecki 1s proposed outline of "Findings Report 11
- 42.
3/7/83 Letter, Starostecki to Uderitz.
- 43.
3/7/83 input to Region I Salem~ Restart Report froni Kennedy
- 44.
3/8/83 draft of Salem Restart Actiori ~lan, prepared by NRC Region 1 (Dircks to Commissioners, unsigned).
- 45.
3/8/83 Memo, Denton to Dircks, re Evaluation of the Implications of the.Salem Unit 1 Event.
46~ 3/8/83 Memo, Haynes to Heltemes, re.Possible Abnormal Occurrence -
Salem Unit 1 Failure of Reactor Trip Breakers to Open on Trip
- Si gna 1. *
- 47.
3/8/83 *Letter, Uderitz to-Starostecki, re Reactor Trip Breaker Failure.
- 48.
3/9/83 Letter,- Gary Toman :to Noonan/ * *
- 49.
3/9/83 Letter, Uderitz to Starostecki, re Confinnatory Action Letter CAL 83-02.
- 50.
- 3/9/83 Memo, Miraglia to Eisenhut, re Use of DB-50 Breakers in RPS at Ginna and Haddem Neck.
- 51.
3/9/83 Memo, Fisher to Varga, re Interim Draft Salem Restart Report.
- 52.
3/9/83 Memo, Knight to Lainas, re Salem Unit 1 Restart Report~
- 53.
3/10/83 Memo; Fischer to Denton, et al, re Daily Highlight.
- 54.
3/10/83 Memo, Eisenhut to Vollmer, et al, re ~ongressional Subcormiittee
- Request.
- r.
4
- 55.
3/10/83 Memo, Mattson to Management Oversight Members of the Salem Generic Implications Task Force.
Enclosures:
(a) 3/8/83 Memo, Denton to Dircks, re Evaluation of the Implications of the Salem Unit 1 Event.
(b) 3/9/83 Memo, Starostecki to Mattson, re Report of the Region I Task Force on the ATWS Events at Salem Nuclear Generating Station, Unit 1.
(c) Draft Outline for the 4/18/83 Task Force Report.
- 56.
3/10/83 Memo, H. Silver to Mattson, re Meeting Notice of INPO Meeting on Salem Generic Implications.
- 57.
3/10/83 Memo, Heltemes to Denton, re Potential Design Deficiency in Westinghouse Reactor Protection S.ystem.
- 58. Westinghouse publication I. ~. 33-850-3D, effective May 1970,. re Instructions for Types DB-50, DBF-16 and.DBL-SO Air Circuit Breakers.
- 59.
- 60.
- 61.
- 62.
- 63.
- 64.
. 65.
- 66.
- 67.
- 68.
- 69.
- 70.
- 71.
Plant status summary of Salem Units 1 and 2 from 1/3/83 to 2/25/83.
Itemized list of Westinghouse domestic plants using OS Breakers.
Unconfirmed list of Westinghouse domestic plants using DB Breakers.
Itemized list of Westinghouse domestic plants and Reactor Trip Breakers being used.
Sal em Unit 1 p 1 ant computer printout during event of 2/25/83.
- Inventory of control room instrument recorder strip charts for 2/25/83 Salem Unit 1 event.
(Obtained during plant visit.)
Control room instument recorder strip charts for 2/22/83 Salem Unit 1 event.
(Obtained during plant visit.)
Salem electrical drawin~ #240148 B 9654-0.
Salem plant procedu*re IPD-18.1.004 Solid State Protection System Reactor Trip Breaker and Permissive P-4 Test -
Tra~n A.
Salem plant procedure IPD-18.1.008 Solid State Protection System Functional Test - Train A.
Salem Emergency Procedure EPI-1 Notification of Unusual Event/Significant Event.
Sale~ Unit 1 Alarm Procedures.
Certificate of Conformance accompanying UV trip attachment 23A9019G61.
.~..
.J
_J 5 -
- 72.
Copy of Work Order No. 925774 for Reactor Trip Breakers.
- 73.
Salem plant procedure IPD-18.1.009 Solid State Prot~ction System Functional Test - Train B.
- 74.
Salem plant procedure IPD-18.1.005 Solid State Protection System Reactor Trip Breaker and Permissive P-4 Test - Train B._
- 75*.
Salem Emergency Instruction I-4.3 Reactor Trip.
' i
- ~
- 76.
Salem 1 Restart Report.
- 77.
Salem 1 Restart-Report with-Cases comments-dated 3/7/83.
- 78 *. -Sal em Restart Report. (_3/9/83 markup)
- 79.
Telex from FRC regarding Mainten~nce Procedures.
- 80. _ Jnstructions for Itemizing E~Llipment for MEL. --
- 81.
One page _to Fi_sher regarding Salem Nuclear Generating Station Reactor
.Trip Switchgear Operational* Verification Program.
- 82.
Master Fne.No~ 255 from PSE&G (2 page excerpL)
_ 83. *General Conclusions ~f 3/4/83 Visit :to* Salem Site (3/4/83 draft).
- 84.
Conclusions onOpera:torTraining and Procedures: (3/7fB3)-*.
)
- __ 85.*
Input to Region I Salem Restart Report' (3/7/83).*. *_ *
~.
- *
- 86. - Infonnaticm Paper on Sa 1 em* Res ta-rt Action Pl an. (3/8/83 draft markup).
- 87.
Summary of Li _censees' Re~ pons es to I EB 83-01..
_-.88. *.staffs.*camments on Salem Restart (3/8/83).
89._ *1nfonnat_ion Pap<;!ron Salem Restart Status-Report (3/10/83 draft).
- 90.
Equjpment Specification Cover Sheet for Reactor Trip Switchgear
. {3/~/83); -.
9L SSPS Train "B*"; Reactor Trip Breaker UV Coil Functional.Test.
- 92.
Maintenance *Procedure A-11 (Rev. O).
93; Maintenance Procedure A-11 (Rev. 1 draft).
- 94.
Maintenance Procedure. M3Q-2 (Rev. 1).
_-.~
- ~.'
.. ~.
fl.-
. t
- 95.
- 96.
Record of Maintenance on Breakers.
- 97.
Reactor Trip and Safety Injection of 2/22/83 (dated 2/23/83)
- 98.
Salem Generic Implications, Agenda for RRG Meetings, March 8-11, 1983.
- 99.
Draft NRC Region I Task Force Charter.
100. Region I Summary of Actions Taken as Result of Salem 1 *AnlS Event.
10~. Sale~ Restart Report - SECY-53-98A, March 14, 1983 102. Salem Restart Status Report - SECY-83-98C, March 29, 1983 103. Salem Restart Evaluation SECY-83-98D, April 8, 1983 104. Salem Restart Evaluation SECY-83-98E, April 11, 1983
. 105. Letter PSE&G to R.
Starosteck~ - Additiona1 Information on Corrective Action~, March 18, 1983 106. Letter F. P. Rahe to R. Vollmer, Informkation on Field Service on Breakers, March 24, 1983 107. Letter from E. P. Rahe to R. C. DeYoung, Information on Trip Breakers,
. March 31, 1983 108. Let.ter PSE&G to D. Eisenhut, Supplement to Corrective Actions, April *3, 1983
.. ~
109. ~etter E. P. Rahe to H. Denton, Information on Westinghouse investigation of malfunctions, March 22, 1983 110. Letter PSE&G to D. Eisenhut, Additional details of independent management diagnostic, April 11, 1983
- l1l. Letter V. Gilinsky to W. Dircks, Salem Breaker Testing, April 11, *1983 112. Letter to D. Eisenhut from PSE&G, Completion of Short Term Action Items, April 13 ~ 19.83 113. Summary of March 14, 1983 meeting with PSE&G & Staff, Restart Status, April 18' 1 983
.114. Letter PSE&G to D. Eisenhut, Commitment to independent management diagnostic, April 4, 1983 115.
Lett~r PSE&G to R. Starosteci, Vendor Manual Program, March 23, 1983 116. Letter W. Carrington to S. Pandry, Contract Work Scope (FRC),
April 22, 1983 117. Letter W. Dircks to Commissioners, Verification of Actions Performed by PSE&G, April 22, 1983 118. Letter PSE&G to R. Starostecki, Additional Information and Comments on NUREG-0977, April 22~ 1983 119. Letter PSE&G to D. Eisenhut, Responses to 10 CFR 50.54(f) letter, Apri 1 -22, 1983 120. Letter PSE&G to 0. Eisenhut, Beta Corporation Reports, April 28, 1983 121. Letter PSE&G to U. Eisenhut, Clarification of Response to 10 CFR 50.54(b) letter, April 2Z, 1983 122. Lette_r PSE&G to D. Eisenhut, Corrective Action Summary, April 28, _ 1983 123. Sal~m Restart Evaluation, A~ril 28,-1983
,\\..
~-
~'
A:lu1yses of the Salem *Jc1~ar Piant for Postul c.te d Fcedr.c.ter-P,o. ifunct~ on ~ tho1,1t
- Autooiati c Reactor Tr1 p-
~STI~H3U~ ELECiRlC CORt'O~TlOtt M. P. Dsbcrne
- 7ransfent Anz1y~i~ *
~i;l ear Safety Department
_-..... *t'.-
., Approve~r:.YT Jh~ I' I.:-.*
. -~..
- .ir:.'.
. ~. : -'.
. ~
'\\,.. -
!*,;1 -... 'I l.1tt e:,
na~er
- ~rans1ent Analysis _ _
l(Jc;1ear safety Department
. ; : ~ -..
-:*.. - t":.
- ~ *..
-~.
ATTACHMENT 5
- SCD?E
,.. e......
Jn *light>1of the recent f~ilul"i!s of the rte:tor ~np b;e::k~ t.o
-autO-;";";ctk~i1y function at the* Saiem pial'it~ the- ?Ui7DSe cf t~1~s !.tudy ~s to rea1i ~tit;a11y predict the consequerrCH of a fai1ur~. t.o "t.ri~ fer 1imiting plant transients whi1e the plant 1~ at fuli reactor po~er. The
. transients an~1yzea. speeifica11y for the Sa1em p1ant. 1 ~re b partial 1oss of st<?a.m ~enerator ma1 n feedwe.ttr f1ow*due to the trf p of a sin.9le main feedwater p~-np and also a complete loss of m;ain "feeawate:~ flowr dut to t_he loss of both main feed~ater pumps..
The h.ttert less pr~ab1e,, '*
. event is that pres~nted in the Sal~~ pl~nt fS~Rr As stated previously,
- the purpose of this.study is to ~aiisticai1y pr~dict. the "tes-ponse of the p Lint _to these ~vents* and~ as $Ueh, the plant' syst~s. ar-~ assu~d -to**
fun,tion nonna71y with the so1~ ex,epticn ?~ing the C?iimcn mode f~i1ure of th~ reactor. br-eab~rs *to.autorna.tkli lly funetion as "'~~ ~xpi>r1-enced. an
- feb.ruary 22 and 25,-* 19S3~ *It shoui d be not.ed that the spuri e>u.s steam generator level trip_generated on 2/25/63 was*.as a -r~$.l.11t of normal ex1>e.cted fe~dwater control syst~m difficu1ties experienc'id at 1ow n.u.)
_power* 1eve1,. lt.'also shou1d. be noted that the los~ *of a. fe~dwater pump
_on 2/2'2./BJ was due to a nonna i
- manuev~rb'ig of an e 1et.tr1ca T bus \\!:'hi 1 e configuring the plant* in preparation for a po~r esc:~1~tit;rn.
Both ~f.
these *events are not nortli~11y expetted at ful 1 power and thus one shou1d.
- tonsider more credible event~ suth a$ A feedwat~t ~~ter drapaut-rather than the more 1imiting and* much 1ess freq~~nt fe~.dwater pum9
- ma1functions.
The study corisidl?r~ *a thirty scic6nd operator t!sponse t~trtE> for.t. manual
- * ** r-eactor trip foliowing: the automatic protection system "demand ~igna1,. a
~imui atj on 'of the actual *response t~me of the* February 15,, 1~B3 -event.. *
- The sti.Jdy a1 so c.onsi der.s *a more c:cn$ervati. "\\'e operator resp::ms~ of five **
minutes in ordt:r. to detemihe*the sensitivity of.the plant resp'Dnseto
- operator action~.
OCSCfHPllOH OF.TRANSIEITT £F'f!ClS _:
-~neric*:.:~tudies.:(WCAP 8330 <Westinghouse Anticipated "Transient~ ~ithout
- 7dp'*A."la1ysts) :of-fat.lur.t.to *trip events previously submitted-1.o the NRC have
- 1 den ti fied the -1 imi ting lull *power events to b.e *ma 1 fJ.Jnttions. - *
~ff~cti ng *st~am generator main feedwater flow.
- The reduction :in main
- feedwatar f1 ow affects; :the* *overa ti heat remova 1 ti!p~bfii ty of the $team*.. *
. 9ener~to,r$ and,.as a result of the misrri~tch between tn~,p.rf~r.r ~ide.
heat *genera ti on,*and the $econdar;y side heat ~mov9 \\ iproduc~.s ~ he~tl.fp of t.he primary: system <;:oolant. If the n;a:::t<Jr-. is tri??ed pr~tly. the *
- auxiliary 'fee*owater system.pr~vit!es s1.1ffic.ient heat removal capai>iHty*
to remove decay heat.. H:>wever if feed~*ate r f1 ow to tf\\e. steal;( generators
- is r.educed or terrn'>nated wtthcut subs~qt;~nt reactor 'td p the seconcfary syst\\2m* ;.;'i 11 be unable to remove *a 11 of the hea: t *t.t-i~ t is ~~n~r.&t.et1 i ri the core..
- Thi~ heat buildup in the prim?ry ~ysteai *is. a. fl,!nctiqn of the amcunt of the feeowatE!t reduction and is indic~ted by rising* ~li.:tor coolant system* t.eruperature and pres:suret.!nd by inc~~stng pressuriz.er
~ater level due to the insurge of the expanding rea~tor cQ~l~t~
- Water 1evei in the ste~rn gel'l!!rators drops as the remaining ;nventor:t in the steam-gener~tor~ 1 s boileo off due to in?dequate spp1y_ ~f f~e~.ater..
Wh~n*the steem generator water: level fa11> ~the point ;;heT"E' the steam generator tube bundie is uncover~d tind pr"imbry to se:c<H'ldarY. heat transfer '\\ ~- reduc~d, ~attar ~~o 1 ent syst~ pr-e$sUr-e cl'ld tempet<~tUN!
.,~~
. ~ j
.... ~
.' *.i:-:c,,-ease at~ greater r!te:.
This ~rtib.~r rate of*te~p~r.ati,;re,~i1Z
. * ~re~su~e 1i n::rea.s~ ; :s rr.ai r.tatlJ.~ ~:: the: pre~:suri -z.e r f 1 l'l s *. p1e te1y ?nt5
\\>.'~ter is,discharged through tt1e pressurize: re11ef ~l'\\c.s.i,_..,,~ v~1v~s.
~~z¢tiYity fee-db6ck~ due tc* th~ !"s~gh prk1ary system ie:iiper<<t:Jre,. ri;:dtrc~~
core po~~r.
- As a result the system pres:.rre begins. to d~crease ~nd -~
steam ;:p~ce is t9dn fom.ed in the pll:~sLJdzer.
'The* limiting criteria* for the ~ostu1 ated tr~nsient~ i:.; tt..at recctor
. coolant pressure be maintained* sufficicnt1y below 'the ~~~ur-e corre~ponding to the ASt-E *Code Service Leve1 C (£.tt.ergencyJ stress Hmitt;p. F(lr the.feattor coolant system, tne co:rreSpQt.d~r:g pr-essure is 3200 p~ia *. *.
Although *t.lie reactor ~ s ptevanted from tripping a1.rto.mati<;~ 11.!f by the
- c~n tn:ode *f~Oure. of the reactor trip b.~at;~rs, there are ~aru r:ontr\\ll
. r-oom'indii;aticriS and alanns*~h{ch ar~ g!!Mratad durin~ the* t-ans.ient which would serve to alert the.operator that the event nas tttken* phce.
-The!~e. fndications in addition to etnergens:,:r proc~duresf whi¢h rtqu1re the
- ,*eri fication of a*. succ.essfu.1 rea,tor trip b-efore a 11 other a.etHms *..
wov.Td s.t,rpport the *mitigation of the consequences* of the transient~
For ~. 10$~ o-f riormal fee-dwat.er event, in ~ddition to '1~~1 process.
~on.t:ro 1 ai tms {pump trip. temperature. P°l"'essure*, 1eve1 and f1<>". * *
... _<ieviat~_on alanns_for both primary ~nd ~~ondQTj':.systems}t*the ftil1cnirrt~
~udible alarms would be g~n~rated: *.-
L
- s*~ai"n/fe~dwater how mismatch an*d low-level (e;!~h st=~.~m!rlitOr}
- z..
Overt~o~r4'iure Delta.. T.,t.urtiine rvnbat.k' - *.
3..
- Overt~p~r_ature t>elta-T rfiactt;>r trip __ d~nd* ::.
- 4. ** * ;Oi~rpo'?rer Del ta* T tur-hi ne rt.inback.
- 5.
(fy~rpoW.er Delta-I reactor trip demantf<. *::
' s.* :High pressunier pre~s-ure'*retictor **tti p 'd~and **.
7... -High *pressurizer 1 eve l. T"e: aetct. trip dti""l'land
... 8~ *Steam,gener~.tOr" 1 O'ff...:low 1 eve l -re~ctor trip dem~rttf -..,... *.* *.
9~' tow steam pressuN safety i.njettion {in co$nt~de~a *~ith h1gh flow).
"> lO. *Low :reactor*cco1ant loop ~lQw reactor* ~rip_ demand.
.. *_ -iab*i_.~s '1"-~~d*2' ~h~-:'the t:~me seciuendes;for'.ttiese *a'tarms;.*;**<*:,* ~.
. As P*~rio*f-th~_procedu~s i,h~*operator is r-eq~irf!d*t~~xf.rci.se~_fo~lowing any "reactor_ trip dem~nd~ the operator is re~1re~ne rur.btd:. action "is the most irn~o\\'"1..ant, ~ft r~!:.::tor trip
- cannot be obt~ined r.':~nu~11y, to terminate the st.e~'ll flo~* <k~ar.o from the
~teem gener.at.prs t.o pr{?s~r-ve steam genentor foveritory.
St~~m pres.-s.u"N:
~ r.d henc:e primary syst.era temperatur~ wii i be contro 11 ed by means of th~
steam cut:lp control system. steam generator relief and/~r saf~ty valve~.
other means out~ide the rnain control rotXll are avai1ab1e:-
1 *. Loca~ rnanuaJ ~r~p of aJ'ly reactor trip brteker
- 2.
local ~~nua1 ~r1p of the rod contr-ol.sy~tem motar-9enr~tor sets 3~ Local manuai _trfp -of -~e turbine iRANSl(~T:*s*lMULATIOH Ana1rises ~re Perfo~d to sfrnti~~te botJi a partial a:ntl ~vmp1ete: 1o-s:s of
~iin "feedwater. These analyses are bas~d upo~ previous m0del~
con~i stent with previous subm1 tta1 s to thr? ff?.C by 'Westinghouse ~n -ATifS
{N.S-W.A-2182., T. M. Aidl!r1on to Dr. S. Hanauer> 12/30/79} b-ut ~lso a~
~difie4 to more accurately *model th~ Sa1em Pli!mt..
. ihe fo11c~ing "e.on'diticns were: ass~ci for. both analyse"s:
l.. lnitia1 r.omal fun power op:rat'iQn at b~ginning: of core 1 ffa.
1ht~
- c~rresponds to the -~urrent c:ondit.ion of the ~)~m rt ant an~ is a1so the ii>>i~ ting conoitfon si rn:e the moQerator te~e:r-at~r~ ct-eff1ci~nt ii-at it~_ least r;eg~t'ive v~iue. Ava1ue of -8 pcm/~f> which i~
"--'-=*-:
va1id for*;~si-_of core Hfe,*was ~ssUF.ed~ *
~'.
~* *..
- 2.
&~th the p;..es'$urfa~r ~1. ief ?nd-safety vaives-ere a-s~~o t~.*
fut.ct1on.
- 1h~re ~re two ~1ief* and thr'.ee.saf~ty..-a1't~s..
Pressudfer he*ater~ and _spra.Y _a1so !uni;:tion automati~~11y *. :
- .. 3.,
Thg -:aut"omatic: turtine _runbQc~. on either Overtemper&ture *_r>~ Over-po~~
. te'l ta-T. s.i gna 1 s 'is 'o?erab1 e.. The "runb.dck setpoint ~ s-3~ be1 ow the..
-trip set-point~ *.The turt>iiie tunback operates on a 30 $econd cy<;:le.
1urbine load *is-first reduced 5: fri LS seconds. - If ~t the end Qf the 30- seconds the ru.nback signal stii1 ex1sts. th~ load is further reduc~d anoth~r 5t and so on.
The load reduction has a ~itigating *
. ef fec:t on th~ transient and he: l ps ~duce peak prim~r:r systE;>m pressure~-
. 4.
The rod control system is asstrmed tc ~e in the rnanu~1 trt0ii~
consister:twith ictuC?1 pr~t1ce.
f..ut~atic action of the t"Od control sys~c..m \\6uld *cause rod ~n~~rtion when pr.iraacy tew.peratu~
1ncrcases and would be iess*consetvat~ve.
- 5. l'he 1.teain oump contro1 sys'tem is l!Vaiia.bie.-
ihe c:ap.scjty cf the*
s'team dump is 5(1';. of nominal steam fio'tt ~t *fu1l power *. *
~. *Auxiliary feedwater flow {1760 QpG) begins at io se~onds fo11o'tri~
r-ece~p~ of the ic~-iow st!:!~m 9ener~tor 1eve1 signa1..
Tnis ~sponse t)rrn: is based upon ~ctual test data f'f"DI~ the Sal~ -?1tnt#
- 7.
Dper~tr;rr. action is asst..rmed to irsitiat~ a successful ~t'.J:lal trip..
Turbirie_~rip is 1niticte-d via the reactor trip brea~er opening.
- ~f.* *. For the com?.1 ete 1 oa.ee:h*ate~ tr ::.n~i ent, ihe: *tall~-~~~:-:-~ ter.
._, pt..r.:-;p,s 2re ~sst.J-rned to coe.!t6o""n to zero fio* in five '$_... tind:;.
for
. the ic.:;.s of a sing1~ Pt.'T.1?. On-:1 pi.:;;-.p i:s. a:ss.\\.r.71$d to ccast.~s~-r1 to zero flo'l:i in five secor.ds; ho,..*e\\*er*. the rern~ining pump ha.s rgted flo'Pt' ce.pt::ity of 7Wr. cf nominal fui1 power feect*F;*c.ter fio'irt.
1h~refo~t the. second p:imp (the Sa i ~m Pl ar;t has t!fD pumps} 'tii 1 l 1 nc:--eG.s~ its f1o¥i' to* i*~ flow.* *The ie$pons~ tims for the second pUi.ip is 20 seconds.*.*
~
<9..
Nom'ina1 c;ontrQ1 end protection sy~tem i-~tpoints were assumed..
1?.AHSlt:Ni RESULTS
.-~- -
- \\*.
- l"he sequence of event~ for hoth a SO sCond and 300 s-?Cond deiay*of hianu~i reactor trip a~ shor
- n 1n Table 1.-
The transient primary pt'essLtr-e t:~_iculations are :s~wra in f1gu~ i.. ihe 1ow-1ow -ste~
<gener~tor 1eve l s:etpoi nt is r-eached at 99 second$; auxi 1 i~r,y
. feedwater is automatica1'ly initiated,.
Ten sec::ond~ l~'U!'t, *-aux'iH~r,y feedwat~r begins to be_ de1iv~red to th~ steam generators..
- .. 30 Second fie 1 ay.
. i For the *case -.;here tbert: 'j s only ~ 30 second. Q.elay,_ t.here are_ no.*
- . :subsequent reactor.tri J:f si gni\\!1 s generi;ii;eq..
There.; s *no 1 a rge be~tup *
- ._.of the re~i;tor coolant be?caus~ the ste~ Qer,erator tube bund1e doe$.
. not uneover. 'Thus there is always ada~uate secondary siaa heat" remev~l.. The pea$:. pressure of ZZ86 psi a ~h.ich occl.lr> at 30 second$>
is on1Y sH ght1y M>ove the* pre~s1,1-re_ at whi c:h. *the presuri.z~r sprays
- are b.ctuated. **'
~
For ~thi~.transie.nt; th~ r~attor coolant *system inteS"ti.tY is not ChA71En9ed.
~
~ -
~ -: : :
. For the case where operator action is *d~\\~y~d 300 seconds (5
. * :-*.*: :mi niites}; the rear;tor coolant system temperature 1 ncreast!s~ ~achin,9 the Overpo"Wer Oe1ta-T, setpoint for tvrbine rl.!nback *at 190 second$.*
This.si ~al i-s ~intai ne:d and thus turbine power* c:ontim.ies *to *rechte~ *
- " * <. 5~ every *30 seconds.until the turbine-load 1 s* at-7~... *At this **
-' * ' point. the su~ of the main f~dJtater Tiow from one p1.m1p pha t~
. ~ *.,, au.d1i*ary feedwater flow is equal *to the turoine s~am fk~~.:* '* '
- . *..
- Jherefore's _steam~ seneratot level does not continue dec~a~frt~ and
- sta~i li zes.' : The ope-r~tor i ni t"i at~d reactot" 6:nd turn1~ ~d p at' 399
- . seconds occurs ~fter the steam and f eed~ater fi OJ{ have :natthed.
The
- peak prirn.ar.r sy:;:te-m press1.rre of Z33D psi a at Z.67 seconds occur-_?.
- before the. steam ~nd feed flow ar-f:.i:r...::tchea.
This,P1"ssu\\' 1~ bel;;>W the relief valve setpoint {2350 psia)..
Th~ pr-essu"J*fzer sprays,_
t:tn-nt>ined wfth-the effect of reduced turnin£: 1oa_d pre"vent am-si gn1 ficant overpre$surizaticnr A~ain; reactor.coolant priessure stays be1 O't( Servi c:~ L~ve 1 C 1 i~its of 32:00 psi a..
~-
/.
' ~'.
i:'* -r>:.
.. 2.
Loss or A11*~ain Feed~~r e
ihe seQUcnce of events f.or thi~ tnnsi~nt are present~d ~n T~ble 2.
The tr!nS1Ent pressure ca1c~l~tions ~re d~pict.ed in FSBur~ 2.
The 1o~-~o~ ste~u generator* level setpoint is re~ch~~ at 3~ seconGs*
10 seconds 1Ater,.auxiHary feedwater is de1ivere<1 to the tt~~m ge l'l<!r'ttors. :
30 '.Second ner~v An autG.':latic turbine* ruriback due to an Overpo~r De1t=-T is initiated at 4S se-corids ancl turt1n~ 1oad fs reduced ~. The
' pressurize,-
~eJief Valve~ Ope:f\\ and mai1'tAin pressure.at the ~etpoifi.~
- j value (2350) until ttte operator tr1 ps the pl ~nt at 63 seconds.
. i Steam dl.lmp is i ni t:i ated and reduces the.primary ten-.pe.ratv~. *to t~
- : no load value of S47°F.
For this tran~ient the ~actClr (:'1-01.ant
~*-* *s.Yst~ pressure is. we11 below 3200 psia..
5 Minute De1ay As in the previou~ case:r the heatL.rp of the primary *ct>o-lant. caused b turbine runbaet.fr;~tiateG by an Ov~rportE:r Oe1t~5 i :si~r+a,.
The.
turbine 1oad is reduced t""ice in 5~ increme*nts until thE Joa-G*is 90"b
-of '~ina1 load,.*
- Stea~ pres:surc* starts _to crop au~. t'O the :'ho_i) off. -.
of 'f{atf!r in the steam 9~nerators, generating tl ic*w \\s.te~. pres~ure -
a lam..
At thi $ time primary pre~sure ste:-ts to im:re~.s.~ a~d thert f s
- a.n in~1,1rge into the pres~uriler-, c~v~ing beth. prE?$~"1.lM~~Y' higf:i 1e11el. and pressure trip alam:s.to be actu~te-d..
The steam ~-enerat-t.'r tutie *bundle ~$gin$ tq *uncover, C:iiiUsing 9 1~rger rate of inc:rease in pdtnt:cy pressutt -arid temper-ature *. "f"nii> p~~sur.'{zer fins.an<t the
_peat pte:ssu~ ~~hed is 3491 ps~~.
Hud~at-po~r hds d~~~s~d ~t trti :S 'po1nt to aDout 3~ of norni na 1 ewe to t.he negative moderawr
.' temperat.ure reac.tivity.feedback. -As the relief rate of llC~ter.
- *>.ttiroug_h the re.lief. and* safety valves im;:r-eases, the.prim~ry SJ$t....~
pressure statts to oecrease ~nd the ~af~ty ano r"e1ief va1 ves clos~
about:3 ~:seconds afte-r. the-time of pea~ pre~sur-e-.: *The operator:
td ps. the reactor.-.. tilanua lly at l33 seconds *.
- CO~LUS IOHS.
Los~ of* a M~in fee~~t.Er Pume The ~su:1ts*pl4esent.ed he-r-e deronstrate that for th-.'! 1oss of Onf= ;i\\!1n
,..{.
.i.. --.
.... ' ~.. c:. $ i
. i. ~
o&
i... "rl t"l -t t '
~
te:e.. :~ai..er p!..1!Up,.i.;nere_art;'.~..... '::-... -w. 17. ~o~ _a om.~ ~ii......,t.....
'::.t;-~ ~~-
- o"-h~... ~
"'----~..,..,"' *o _... * "'!!>
-p.--*-*~.....,..,... t"".c
.1:...... * *... £.... -*".z: ** --,o..
\\.n. ~',:.
~t:1J:1*~i..-':\\.i.
~
~ ':l \\.o
'~*t-V t;:lQ.L-""'-*
i.;."'\\r*
- I~-
- u~t-Vt\\'C.\\.I '3-
~'Q~!~\\~-.i,..
.1 ht~ occured.
Fur,J-H:~re, ~Vert for.t}1e =~ent_ ~ith a five m~rmt£ oela~
- 1r. ~ea.::tor trip. aut...~tir; turbine runback reduces stea;u fl.ow Ui... ~atch
. the Cli.p~'iiity of the auxi1iar.y 'fe-edwater.
For thi$ even~ ~i"-e t~ r:-o threat of overpressurizat1on ~n t.~at t~ pr~ssuri~er relief val¥~
setpoint.is not ~ven re~ched.
r...
- r-(-.
.[_
f.~ -
i;;
~:_:, r-.
r.:=*
~-*
g_:
=-
t~;
~
1'
- £i
~-
- !F
'.,i
.E--
F:--
~
$-=
- .. *-~
-~*>E::.
J,
'£
..r-j
ComDiete Lo~~ of.Hain Fetdwa~
ro r the cdr.:p 1 et~ i css of f e~*i:1..-~ter*) ope re: tor action cor.s i ~tent ~i th th~
~ction tir.!~ tar.et\\ at the pl~nt en *the Febr.:.:H)' 25, i98.3 eYHit ~s
~ufficient to prevent overpressuriz~tion of* the reactcr cooi~nt system.
?ea_kprirnary.system pressure results on1y_~n pressurizer ~lief ;r~lve actuation without the actuati o.n of* pf-essunzer saf~ty va1 ves.
Fur:thermor~, there ~re 3 mb.jor-aia~s which are ~ctuated-in ~dd~tfon t.t:-
the steam generated 1ow-1ow level alarrn to* alert the operator t~ take ar;ti on..
r' f..s dis:,ussed earlier, 1t is a :atJor r~duction in pr~tntiry tn seconda11 heat. t:ransfer capab111ty \\ltirtch caut~s th~ prima~y system h~atup.and..
p~ssut-e increase~ A turbine trip r-educes. the amount of ~te.t:t f1ow tnd the *r.atti: *Qt "'hi ch the level *in the steam generator d;op~.. lf the *.
turt>~~~.is.tripped before the~ is a significant loss ~f' st~~ 9~
.. n~r~wr
- ~ *inventory, the. tubes will.not. uncover !:nd the pMmary syst~ \\rl.11 not overpressurize.
Based upon th@ results di~cuss~d in the pt£:viro.ts sec'tion. operator action to.trip the wrbine at or bE;rore Oc:{e to one anc
- a ha*lf minutes. fol.lowing the_ 1ow-1ow 1eve1 tdp ~nd a1~m ~~1d pi"event
- cverpressurization.of the Nactor. cco1ant tyst!SP b~ottd 320(} p$1a..
It shov.1d be noted that-the core nuclear characteri sties* (t ~o~r~tor reactivi t.Y coefficient of -8 pcm/*F) used are r-.ot representat"\\"~ of the
- * ~ctu~ l*,, core. de'S:i gn. for".the S~ l eiit Plant. Pnvi OU'i> A TiS ~na
~,i'5-!S h1H'~
- $hown
- tn.e pea.k *pre$SUre to be " strong function o.f t.i)e.
.cc.efflc~'f!nt and
- .the?"'!: ts a i(jOpsi reductior'i for'~very l pcm dei;:r-.ea~e in tbs coefffef~nt *... Th~ ~iem cor~ {s aes1'gn$6 to oper<Ite su:;h th~t by the-
- tirr-re the plan.~ ~ached fu11 po~r it w~u1d h~v~ ~ cc~rncient ~f...
- 1ops * *
.p~m/.,.Por'2~5 pcm less *tnan:the *ceefficient in the st.urly..
'This
. > *: c~efficient. "P."OU1d be *redutE?d even furt.her:.by. appro;;i~ately z p~cF per.
- . ::. :ro.nth of *opefat~on _(see r~gur~ 3).
- Th~ 'J0.. 5 pt~ coeffici-ent ~;;uits in
, *a peat~pr-essure*fcr the limiting ease of f1ve-rninute tipraror *.actfcS"i of
,. JZ4l p:;i.s. (a,250 p$ia reduction :from 3~91 psi a} which i$ '!i:it:h~n:-ti-,e*...
- cai.culatton~l* band of the.ASt-E Stres~ Level C Hmi t.. The~fore~ the
.ca$~.-~pr~s~nted *'fn_F*igure 2 would not exee~d the tci:~ptance ~-riteri~.
.~
- , ~
~
- .. '. :. ln tone lusit>nt
- th1:$ $tUdy h~s
- demonstrate-d the ability of the ~lem *
- '
- f!Jtl ear' p'\\irrt. to wi t"istan.o th~ eff~ts of po!;;tu1 ~t~d gicSs f~e~w~ter.
ma 1 func-tichs With out ti: actor trip ~t fuli pow~r". \\ii th ~n a.rtiff cf ~11.Y *.
- *. ** 1ong d.e'lay for operat.o.r action. The r-esu1ts snow act~?til>)& respons.~ **
.....t.iich i:s..rithin ca1C::ule;tio.nai unterteinties of the IS~ St~ss Leve1 C 1~n~t~*.
'ii;~se r~su1ts are*ftirther 2:.ffected by the iol':' probah11'\\ty of tJ"iese. ev~nts * ~cur-i ng at fu 11 po~r in e.odi ti on to the ~xpect-e~ * * - *
- i nc~_a::i ~1y l;enefici ~ 1 nuc1 ear chart:.ct~d sties cf td+-;e p1 ~nt c~?r co~
~1fe..
- J
- c:.,.
- 1-._
TABLE 1 Sequence of E~ents tvent Lo~s of one pump {al ara)
- Remaining pump de11vers maximum f1ow lo-w.:Jo'W SS lev~1 setpoint {altirml;.* *
! auxiliary fee&water signa1 * (aiann) k.!x"fTi a ry f eechilater begin~
. *~Operator trips reactor and turbine
- O? 11 T rt).O~ac:k ~etpoi ~t {al arm).
turbine load reduced St Turbine i oad reduced 5%
OP li T trip setpoi nt *(a 1 arm j Turl:lin~ 1otd reduced Si Peak ?ress:ure Occurs.
. lurbi n~ loa1f r~~uced 5~
- Turbine 1 cad n:duced **5$
H1gh pressurizer level setpoint.(alarm).
.. * *operator trips reactor and turbine :
(1}
30 sec;:ond de 'It:)' before man~a 1 trip
- (2) 300 second da 1 a)* before roaooa i tdp 0
20 99
. 109
- *iZ9
~-
3 alarms Time2 D
20
.99 100
... 190
~33.
220
- iso 267 { 2350 ps1a )'
ZBO
- .*..:no 311.
399 pr-ior to *
- prior. t~ * *
- trip
- trip -
~-
iABLE 2 Sequence cf Events C-ornplete Loss. of P..ain Fe~dwater Event Loss of main feedwater P'tlmPS {alarm)
Low-jow SG l~vel setpoint (~1a~);
, Qu-x,i 1i~ry f~dwa~r ?i gna 1 generation O? b Ti runtiac~ setpo'int !alarm}
furbi'ne load reduced si OPb T trip setpoint (alarm}
Auxiliary feedwater besins Pressurizer reH ef valves open.
Cper:atQi trips reactor/turbine T1.1r6in-e load red1,.11;:~d ~
- High pressunzer 1eve1 tdp $etpoint {aiam)
Low steam pressure S! (a1arrn}
High pressurizer pressure setpoint.(a1llrm)
SG tubes ~gin to uocover; Ste.!i.m. f101' drops pressurizer safety v~ives open Pres~uri zer fill$
Peak pr~~su~
_Pressurizer safety valves c1ost Pressurizer. relier vaives c1os~
+.
Low.RC flo\\it' setpoint (alarm)
Operator trips reactor/turt>ine.
{1) 30 second delay before manua1 trip (2°}
JOO seeonti aelay before manual trip Time1 0
33 34 43 43
- 55.
¢13 4alarms prior to trip Tirne2 fj 33 34 43 4S
-55 114 ~ 3491 psiz:) -
1?Z 155 *.
1~9 J33*
1 aian:~
p-rior to trllf
tte I *
.J:
t.\\
- J 1 I
- \\ *
<L*
c. '
'I I\\
I I I * :e vi c.:,..
- U*
\\.. :l j)"
I' C1 -.
L.
- 01 L-(l
- *II "'
ol.* ' ~
u c::
IJ ll
- 11 1.11 C.:t 0: I
~.
II lJ B
~..
- f..
n
- 8...
- ~)
(")
g
\\,l u
1 I v:
- II
- 1 I n
~-.
Ot
.. (.)
~.,..
\\ j
' -IV
. *: 44)
II>..-
_ V', *_;
(.)
l
.'&II
'O
.~
- ~
1:l '\\*'
lt i * ! ;.
~ i ~
- Ill-.
- Jl
': i i
~ l
- 'i
- .\\:'1:
. i * '.
I
,i 1_,
.*M
- -~--r-7-r-....-'~~~ "
,i
' \\
' \\
.. *1*
(~
H
<")
<I....
f ~
~*
.~
I
(' i 0
(*:,
g II LI I *
'* ~
h)
I.
I I
J
. \\
j
~.lljQ
~ -
~
.A --
noo
! r
- I i lOC 30 Seconds 0~1ay 2.00
.lOO J.0-0
~Otl
~DO t\\ll'lf (a-;'
300 Sr;COt'"id$ De1ay SefcN: Tr1µ
. ('
- '..5
~ -10
-30 0
FIGURE
~
MODEAAiOFi ID.?i;:R,:..11.J?.E COEFF!C!ENi OURIHG CYCL~ 5
. AT Hf?., k~O, E.QUIL!'SRIW: X8iOH CONOiiIONS"'
- lQ
,. 12
~~
1CAP 10242,"The Nuclear Design of Sa1em Unit One Power P1~nt Cych: S't
~.
Distribution: Ltr for Sen. Joseph R. Biden, Jr. fm Chairman Palladino dtd 6/2/83 Central File w/incoming NRC PDR w/incoming Local PDR w/incoming EDD Reading ORB#l Rdg OELD OCA (3)
M. Bridgers (EDO #12879)
Program Support Staff, NRR M. Jambor D. Nottingham D. Eisenhut D. Fischer w/incoming C. Parrish W. Di'rcks T. Rehm J. Roe V. Stello Reg. I Administrator R. Minogue R. DeYoung_
G. Cunningham SECY OPA HDenton KJohnson #21879