L-85-347, Changes,Tests & Experiments Made W/O Prior Commission Approval for Jul 1984 - June 1985, Per 10CFR50.59(b).Related Info Encl
| ML20141N256 | |
| Person / Time | |
|---|---|
| Site: | Turkey Point  |
| Issue date: | 06/30/1985 |
| From: | Williams J FLORIDA POWER & LIGHT CO. |
| To: | Grace J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| Shared Package | |
| ML20141N255 | List: |
| References | |
| FOIA-85-654, RTR-NUREG-0737, RTR-NUREG-737 L-85-347, NUDOCS 8603060013 | |
| Download: ML20141N256 (61) | |
Text
{{#Wiki_filter:'
. l e
4 l TURKEY POINT PLMT UNITS 3 MD 4 DOCKET NlM8ERS 50-250 MD 50-251 CHMGES, TESTS, MD EXPERIMENTS MADE WITHOUT PRIOR CG NISSIOM APPROVAL FOR PERIOD , JllLY 1, 1984 THROUGH JlalE 30, 1985 IN C(MPLIMCE WITH TITLE 10. SECTION 50.59(b) CODE OF FEDERAL REGULATIONS e-1 am e 8603060013 060106 PDR FOIA DOHERTY85-654 PDR
q INTRODUCTION This report is submitted in accordance with 10 CFR 50.59 (b),* khich requires that reports of: '
- 1) changes in the facility as described in the FSAR ii) changes in the procedures as described in the FSAR, and 111) tests and experiments not described in the FSAR -
which are conducted without prior comission approval be reported to the Comnission at least annually. This report is intended to meet this require-ment for the period July 1: 1984 through June 30, 1985. This report is divided into three sections; the first, Plant Change /Pbdifica-tions, coverink changes in the facility as described in the FSAR; the second, Procedure Changes covering changes in the procedures as described in the FSAR; an- the third, Tests and Experiments, covering tests and experiments not described in the FSAR. - Appendix A to this report is a list of safety and power operated relief valve actuations, which is submitted in accordance with FPL's comitment to comply with the requirements of Item IIK.3.3 of NUREG 0737. This report covers the period from January 1,1984 to June 30, 1985. . J I Appendix B to this report is a sisnmary of the findings of the Steam Generator tube inspection performed on Unit 3 during the report period from July 1,1984 i through June 30, 1985. 1 i 6 i m. I l
- - - - - . , -- -- ,_. n .-,p.- - - . ,- . , _ , . - , . _ . , . , . , . , _ - , n-- n,n - . . ,-mw_ ,. ---,---,m---,nn-,mm.,,-
I PLANT CHANGE / MODIFICATION 83-88 PC/M CLASSIFICATION: NS Y UNIT: 3 TURNED OVER DATE: 02/19/85 03/01/85 4
SUMMARY
DATE: REVISION: 0 DELFTION OF FLUX RATE IMPUT TO TURBINE RUNBACK . Summary: This modification consisted in deleting the flux rate input to Turbine Runback - Logic and. Modification to ascertain that the reactor may be operated only on manual. The Turbine Runback System has been prone to spurious runback which has subjected the plant to unnecessary transients and inavailability. Saf,e,ty e Evalaation: This change i s nuclear safety related with no unreviewed safety question. s Pased on. the following: the probability of occurrence / consequence of an
' accident or malfunction important to safety, previuvaly evaluated in the
/ F.3.A.R., has not increased, No possibility for an accident or malfunction of a different type from any' evaluated previously in the F.S. A.R . has been created by this modificati~on.- Additionally, the margin of safety, as defined in the basis for Technical Specifications, has not decreased. _
4 hw 9 1 e 4 1 a e m I e* _ . - - - _.-,._n., , , - - - ,, ,n-- n,- _ , , , _ , - _ _ , , _ - _ - . _ , , - , _ _ , _ - - . - - - _ - - , , ,, , , , , - - - - - , - . _ . _ . . . + - .
c e , I,, *
\ UNIT: 4 l TURNED OVER DATE: 12/14/84
SUMMARY
DATE: 12/20/84 REVISION: 0 DELETION OF FLUX RATE INPUT TO TLRBINE RUSACK e M. l
~
This modification consisted in deleting the flux rate input to Turbine Runback Logic and Modification to ascertain that the reactor may be operated only on manual. The Turbine Runback System has been prone to spurious runback which has subjected the plant to unnecessary transients and inaiallability. Safety Evaluation: This change is nuclear safety related with no unreviewed safety question. Based on the following: the probability of occurence/ consequence of an accident or malfunction important to safety, previously evaluated in the F.S.A.R., has not increased. No possibility for an accident or malfunction of
- a different type from any evaluated previously in the F.S. A.R. has been created by this modification. Additionally, the margin of safety, as defined in the basis for Technical Specifications, has not decreased m-9 .
\ n b
/>
FLoaica eowsa a Liowv covPauv [ ,
,EP .a. ass L-85-347 Dr. 3. Nelson Grace Regional Admini:trator, Region 11 U. S. Nuclear Regulatory Commission ,
101 Marietts Street N. W., Suite 2900 Atlanta, Georgia 30303 3 /)s /QE///g;,-
.j Desr Dr. Grace #U g
*/
Re: Turkey Point Units 3 and 4 C ;,CQ Docket Nos. 30-250 and 50-251 + 10 CFR 50.59 Report Florida Power and Light's Report on " Changes, Tests and Experiments Made Without Prior Commission Approval" for the period July 1,1984 through June 30, 1985 is attached, s Very truly yours, L-
- b ':: Wil6ams,
/" 3.'W. . : W Jr.
Group Vice President Nuclear Energy Department 3WW/PLP:dkw Attachment i cc: Director, Office of Inspection and Enforcement Harold F. Reis, Esqusire PNS-LI-85-309/1 PEOPLE sERkiNG PEOF6E
. ._ \
e i 11 si $ t II [41I o' . I s .
.7, ,
IN [ll?ll%u._l
!- s- g , ,
l !i i Al !- > I I ~!7 'Q' i I
- i g
tt II is I i' r . i i li P .I 1 e i - - g i g4 1 I .e 88 5 : a , g i M ij - I ' A i
, -C i : 8 tr id l h h
-u c
t gIl . 18 .. .._ M qh 1 .. :, 3 e l,
.._t . ..
,'E7
.] s g ~~' L .
e l ilL tnt E lt o ir 1]-
- M 's kJir)
I Ev' I lli[- - H!
<!g!!
til 8
,, , s !
I , ij ..,i I;n -
. i.
i
- u .
.g 3
.L :
l 1
- I T
,3_[ t I
es: . 1 .. .. .
>I o i
l ' 1 1 _ __. _ _ _ __ __ __ l j
'I '
s, -. s ; i at - ' . T 8 i
.g
. i
. 4 is -
6... _. g 3 rdtyg 3 _ m
. .7 -
i
, ,v,;3%,
-o . l ,! . *,.T..',., .
k ,e ' ,, 8 1."" O '
- ed
' l " 'h l -g (rb ' 5' ""'7""
T
- 5 -
o T 7 ' 1 Ill "*.,".";;='l~,J-
?- -
il l ' f, g
.I i1
==3.
,! iit,r. $v. ir d,,r
, l
[r: : r I *r5 =
. 1 3!i.
1 1 i, . . . . . . 8 11111111 lif Il 11 .. I
- " I q Q '".."."...l
~-
, - . - - - , . - --- ,.-.,...-ne
.', */p u
1 ! TURKEY POINT - TURBINE RUNBACK LOGIC MODIFICATION SEPTEMBER 12, 1985 (P. MILANO) PROBLEM i , MODIFICATION OF TURBINE RUNBACK SYSTEM AT TURKEY POINT 3 REDUCED REDUNDANCY / DIVERSITY OF ACTUATION FOR DROPPED ROD INVALID BASIS FOR MODIFICATION SUBMITTED TO NRC j - OTHER PLANTS (E.G., INDIAN POINT) MAY BE CONSIDEPING ] SAME MODIFICATION 1 j - SAFETY SIGNIFICANCE - FSAR ACCIDENT ANALYSIS ASSUMES TURBI~NE RUNBACK TO PROTECT CORE FROM DNB, INVOLVING REDUNDANCY AND ] DIVERSITY DISCUSSION FPL EXPERIENCING SPURIOUS RUNBACKS DUE To NIS NEGATIVE
! FLUX RATE SIGNAL WESTINGHOUSE ANALYSIS SUPPORTED ELIMINATION OF NIS l
SIGNAL BUT PROVIDED NO HARDWARE / DESIGN RECOMMENDATIONS , j (ROD BOTTOM SIGNAL WOULD REMAIN) 1 NRC CONCURRENCE BASED UPON LICENSEE INFORMATION PEGARDING ADEQUATE REDUNDANCY / DIVERSITY WOULD REMAIN i ! M2-
9 2-AFTER MAKING MODIFICATION (INSTALLATION OF', BYPASS SWITCH), FPL REALIZED THAT DIVERSITY OF RUNBACK ACTUATION ASSUMED IN FSAR WAS LOST; I.E., REDUCTION OF T-G GOVERNOR AND T-G LOAD LIMITER l PLANT OPERATED FOR ONE MONTH WITH MODIFICATION IN PLACE FPL INSTRUCTED OPERATORS NOT TO USE NIS/ RUNBACK
, BYPASS SWITCH (FEBRUARY 1985); NRC NOT NOTIFIED ALLEGATION MADE BY FORMER WESTINGHOUSE EMPLOYEE THAT: -
ROD-ON-BOTTOM SIGNAL NOT REDUNDANT; THEREFORE, NIS
~
SIGNAL SHOULD NOT HAVE BEEN ELIMINATED FOLLOWUP REGION ll AND RESIDENT AT TURKEY POINT. REVIEWING AND
~
EVALUATING EVENT AND CAUSES TO DETERMINE SAFETY I SIGNIFICANCE l VPB REVIEWING WESTINGHOUSE TRANSMITTALS TO FPL AND 4 COORDINATING WITH REGION 11 DETERMINE WHETHER OR NOT OTHER WESTINGHOUSE UNITS 4 WERE SIMILARLY MODIFIED POTENTIAL ENFORCEMENT ACTIONS BEING EVALUATED WITH , RESPECT TO FPL AND WESTINGHOUSE - 4 i I
-+-.-..-w.,,~_v-. --nng.--,-ww~ -,.-,,onw-.n, r_., wn ng m , - ,, - -,n,w,, g, - , , ,,,---m-..,-g . , -,-n,yn--, .n, , --weawn,ww .n m . . m . .
M FIGURE 1 TURBINE GOVERNOR CONTROL SYSTEM - Twasma mas atranteca esoucTeow A i -- -- _ 4 OvtRTEMPsaATWas, owEmpowna 8 8 tel% ROD es av(%) = ax(%) Oft 0P ti&NAL (1/4) 8
$4AltiA) i j
t i
, t T I M t. ',
DELAY T.ME DELAY ' ac6Av l-i cycuc) LO< R(ELAY .
-.i H a ss sec. -*
so-so sac. f E LU y- -
~
m r... ;.c , _ _ I _ i s i ,4 . l l 1 1. 1 i 410JCE LOA 8 REFERENGE av moo % / naa.
. l -= - - -
1 g arnumeAvT / f FSAR TURKEY POINT UNIT 3
- - - - - - - - - - - - - , - - - - - . , . , - - , -, -----,,,..-_-,----n~, . , , _.,--..e
- *., --e.,--,,a-------,,--,, -e-a,--n--_.
FIGURE 2 TURBINE LDAD LIMIT REDUCTION PROTECTION SYSTEM-WAtl4E, LOAD LSettT REpuCTIOel . A
~~
8eII ~ SOD Rop~ M yon DROP SteadAL Se6NAL (b) ( M ROD) ( (tuin S) l l (;) . I - TURStat POWEst >y'& (F4R47 STA4E TV8testet, pittmRE) l 5* 5 1 l g i - e i s RCDVCE LAAD UmlT I -_ .. __ k e FSAR TURKEY POINT UNIT 3 E
b e REPORTABLE EVENT NUMBER 02062 . FACILITY : TURKEY POINT DATE NOTIFIED : 09/12/85 ' UNIT : 3&4
, TIME NOTIFIED : 19:53 REGION : 2 OATE OF EVENT : 09/12/85 OPERATIONS OFFICER
- BYRON HANSEN CLASSIFICATION : 10 CFR 50.T2 NRC NOTIFIED BY : W. SCHINKUS .
CATEGORY 1 : UNANALYZED j , COMPONENT :
. CATEGORY 4 :
a, , ' BOTH UNITS AT 100% PCWER. IN LATE 1984, THE LICENSEE DETERMINED THAT
~
-THE' RUNBACK LOSS (ONOF HIGH A SINGLE NEG. FLUXVITAL RATE). AC INVERTER WOULD CAUSE AN NIS INITIA SPECIFICALLY, IF AN INVERTER WERE LOST, THE f;IS ASSOCIATED WITH THE INVER"TER.WQULD GENERATE A DROPPE0 F00 SIGNAL (I RECOMMENDATION ROD OROP/ TURBINE RUNBACK THESIGNAL.LICENSEE A INSTALLED A' SWITCH WHICH BYPASSED SUBSEQUENT REVIEW OF THE METHOD EMPLOYED
! TO DEFEAT THE NIS INPUT TO THE TUREINE RUNBACK LOGIC' REVEALED TH C0fJDITION PLACED THEM RE3UCEO IN A THE. DIVERSITY OF THE TURBINE RUNBACK PROTECTION SYSTEM AND , CONDITION NOT CONSISTANT WITH THEIR FSAR. THE SAFETY , SIGNIFICANCE CONL ONJ . OF
/
THIS EVENT IS CURRENTLY UNDER INVESTIGATION. NOTIFIED R2 (T.
==
4 9 O e Na
RUEIVtU i /- OCT 2 41;63 ,, Q [Q M,L w F. P. O L CO. Tun.Dese..rna W [ AUG 261983 Westinghouse Electric Corporation Water Reactor Divisions g;sy4s, j FLA. PWM. & LIQHY Co. sex 2ns TURKEY LTINT PLANT PatsDurgh Pousyfvama 15230 FPL-83;-627 s O# August'23, 1983 pf g Ref: HS-TA-83-357 ) K. N. Harris -1 Power Resources Nuclear P. O. Box 529100 1 Miami, Florida 33152
]
Florida Power & Light Company Turkey Point Units 3 and 4 UNREVIEWED SAFETY ISSUE FOR DROPPED ROD ON TURBINE RUNBACK
Dear Mr. Harris:
This letter is to notify you of the potential of an unreviewed safety question concerning the safety analysis of the dropped control rod event presented in plant FSARs. This accident is analyzed as a Condition II
, event, an event of moderat'e frequency. Conservative assumptions are -
made in all accident analyses to ensure that the cases presented in the FSAR will bound all potential conditions under which the accident could occur. Protection is provided by an automatic turbine runback. The acceptance criterion for this event is that no fuel failures occur. This is verified by demonstrating that the departure from nuclear boiling ratio (DNBR) remains above a limit value for the plant. A dropped rod causes an initial reduction in nuclear power which corres-1_ ponds to the reactivity worth of the rod. A dropped rod typically does not cause a large enough reduction in nuclear power such that the nuclear
^
power exactly matches or falls below the turbine power at the runback setpoint. However, with sufficient reactivity feedback, the nuclear power will decrease to match the turbine power and the plant will stabilize at the runback setpoint. No direct reactor trip is expected to occur as a result of a dropped rod. This is the scenario presented in the FSAR. If there is no reactivity feedback, nuclear power will stabilize at the power level corresponding to that caused by the d opped rod. Primary reactor power will be greater than turbine power, resulting in a heatup
= of the primary coolant. If the overtemperature delta-T setpoint is reached, a reactor trip will occur and terminate the event. In another case, the steam generator safety valves will open to accomodate the mismat reactor and turbine power. No reactor trip will occ '
stabilize in this condition. Both of these scenarios igj - than the transient currently presented in the FSAR. g{ Mr
v K. N. Harris FPL-83-627 August 23, 1983 Since the transient presented in the FSAR is not the limiting case, this must be considered an unreviewed safety issue. Confirmation that the DNB design basis is met must be made for the more limiting conditions. Westinghouse will demonstrate this on a plant and cycle specific basis. Westinghouse will keep you informed as to the status of the new analy- - tical methods. Resolution of this issue is expected within 6 months. It should be noted that actual plant operating safety is not affected by this issue. Sufficient reactivity feedback exists such that the severe heatup predicted on a safety analysis basis is not expected to occur. Therefore, it is expected that the DNB design basis will be met under actual operating conditions should a dropped rod occur. If you have any questions, please contact this office.
\ .
Sincerely, . WESTINGHOUSE ELECTRIC CORPORATION J* !b: = J J. C. Miller, Manager M Operating Plant Projects, South 4
~ Il- ~- TFS/pmc
_ cc: R. 4. Acosta S. Shepherd
- H. N. Paduano A. E. Siebe R. A. Kaminsky S. A. Verduci i H. E. Yaeger J. K. Hays - l D. W. Haase
- l J. J. Quinn i i
9 O O 6 8
. .' N i: " 4 194 1 s
*- ., . . , . : n 1.
n4 . ............
\
Westinghouse Water Reactor . ea re +m Electric Corporation Divisio,ns j # ' ', p' m 3,3, yc c f' PmsDuc Pewytvama 15230 April 6, 1984 Mr. Steve Craig 84FP*-G-046 - Florida Power and Light Company P. O. Box 013100 KEYWORDS: FPL Miami, FL 33152 RSE ,
Dear Mr. Craig:
FLORIDA POWER AND LIGHT COMPANY TURKEY POINT UNIT 4 CYCLE'10 REDESIGN RSE Enclosed are five copies of the Turkey Point Unit 4. Cycle 10 Redesign Reload .. SafetyEvaluation(RSE) report. The dropped RCCA analysis was consistent with the attached Westinghouse notification to Florida Power and Light Comp'any of a potential unreviewed safety issue. Specifically, the accident was analyzed from an initial condition af hot full power with a moderator temperature
. coefficient (MTC) of 0 pcm/ F. A positive MTC is not pennitted above 70% ~
power. Credit was taken for the turbine runback to 66% power (nominal set-point minus 4% uncertainty). A range of dropped RCCA worths was considered to ensure that the DNB design basis is met for all possible dropped RCCA's. The results of this analysis are acceptable. If you have any questions, please call me. . 1 __ Very truly yours. W0.Y B. A. Pearson
/5 Project Engineer NFD Fuel Projects ,
i cc: R. J. Acosta I J. R. Bensen I D. C. Bradford D. Mantz W. E. Coe S. K. Mathavan l R. A. Decker R. Mende - ~~ J. A. DeMastry J. E. Moaba .- D. W. Haase H. N. Paduano C. Vi.11ard - R. D. Hankel D. C. Poteralski C. O. Woody .
. J. A. Handschuh G. A. Rowan H. E. Yeager V. A. Kaminskas* 5. H. Shepherd C. S. Kent A. E. Siabe .
E. R. Knuckles F. H. Southworth C. U. Laisure
- w/ Attachment S. Verduci
, _ . ~ . . , ... . . _ _ . . . . . . - . . . .
N 1 W'estinghouse . Water Reactor ww Fw asen Electric Corporation DMslons . g,,
.', - wwp Pemsmne 152m
"' (, e May 21, 1984 84FP*-G-067 Mr. Steve Craig ~
. Florida Power and Light Company KEYWORDS: FPL P. O. Box 013100 .
TURBINE-Miami, FL 33152 RUNBACK SAFETY-
Dear Mr. Craig:
ISSUE FLORIDA POWER AND LIGHT COMPANY TURKEY POINT. UNIT 5 3 AND 4 SAFETY ANALYSIS OF DROPPED CONTROL ROD EVENT Westinghouse has completed its dropped control rod analyses for Turkey Point Units 3 and 4, based on the following transient scenario. This accident is analyzed as a Condition II event, an event of moderate. frequcncy. Conservative assumptions are made in all accident analyses to ensure that the cases presented in the FSAR will bound all potential conditions under which the accident could occur. Protection is provided by an automatic turbine runback. The acceptance criterion for this event is that no fuel failures occur. This is verified by demonstrating that the departure from nuclear boiling ratio (DNB't) remains above a limit value for the plant. A dropped rod causes an initial reductio'n in nuclear power which corresponds to the reactivity worth of the rod. In addition, a turbine runback to a pre-
- set level is actuated by a rod-on-bottom signal. A dropped rod typically does not cause a large enough reducticn in nuclear power such that the nuclear power exactly matches or falls below the turbine power at the runback setpoint.
However, with sufficient reactivity feedback, the nuclear power will decrease to match the turbine power and the plant will stabilize at the runback set-point. No direct reactor trip is expected to occur as a result of a dropped rod. This is the scenario presented in the FSAR. If there is no reactivity feedback, nuclear power will stabilize at the power level corresponding to that caused by the dropped rod. Primary reactor power will be greater than turbine power, resulting in a heatup of the primary coolant. If the overtemperature delta-T setpoint is reached, a reactor trip will occur and tenninate the event. In another case, the steam generator safety valves will open to acconrnodate the mismatch between reactor and turbine power. No reactor trip will occur and the plant will stabilize in this condition. Both of these scenarios are more limiting than the transient currently presented in the FSAR. Westinghouse has perfonned an analysis for Turkey Point Units 3 and 4 based i on the transient scenario described above, which Westinghouse has verified to l be more limiting than the FSAR. Consequently Westinghouse has revised their
~
84FP*-G-067 Mr. Steve Craig May 21. 1984 procedures to be consistent with the new transient scenario and all future
. safety evaluations will be based on this.
The Westinghouse analysis for Turkey Point Units 3 and 4 shows that the DNB design basis is met for this scenario, as well as the FSAR transient scenario. Thus, no unreviewed safety question exists and Westinghouse considers the issue resolved. If you have any questions, please call me. Very truly yours, ,
. MO
/s B. A. Pearson *
. Project Engineer -
NFD Fuel Projects h -w e e . l l l G l
\
~~ ~^ -- ^ - - - - - - -- -
~ .~i ~ { j~^ -
i ', . M
- f. m. L, .a i..? ,
. < PL;'t.
-Fn p wa., ,
- k. . .
on . .--~ ,
- "*T"**p'**
Westinghouse WaterReactor ** Electric Corporation Divisions euna l ""*"' W** " i j}j ,, y ) > f f March' 18,1985 FFL-E-586
' NS-0PL5-0PL-5-119 -;
j j C. M. Wethy i Vice President j Turkey Point - Nuclear Plant Florida Power & Light Capeny l F. O. Box 029100 l Miami, M orida 33102 i FLORIDA FOWS & LIGHT COMPANY , TUREY POINT UNITS 3 & 4 ; l
! pomncmas in tunnist nUNBACK SYSTM
Dear Mr. Wethy:
4
- In 1982, Westin$ouse provided to Florlds Pouer & Light a enrety evaluation for .
l a modification to the turbine rmbeck syste at Turkey Point which concluded
- that the deletion of the NIS rod drop (or num rate) signal input to the ,
l l turbine rebeck was acceptable. This ev.nlustion addressed the impact of I deleting the num rate signal on the socident, analysis. NRC approval was l obtained in January 1983 and the change was impimented at Turkey Point 3 & 4 in late 1984. Recent discussions with FraL personnel at the alte revealed that the means by which FraL deleted the NIS rod drop signal ruoved diversity from
-- the turbine reback protection syste. This is inconsistent with thopter 14 of l j the F3AR, which states that there are diverse moons for obtaining the reback.
l Besed on a Westin$ouse recommendstien, (telecon on February 20,195), the NIS rod drop signal input to the turbine reback was reinstated until an saceptable j means of deleting the signal can be socomplished. The following provides a background discussion and formal recommendation concerning this issue, i The outematic turbine rebook feature at Turkey Point' Unita 3 & 4 provides protective action in the event of a single or estiple ' Rod Quster Control Assembly (RCCA) drop. Detection of single or multiple dropped RCCAs (including dropped RCCA banks) oocars by either a rod-ort-bottaa position indioetion signal or by a change in neutron nun as seen by the encore detectors (NIS rod drop signal). The rod-on-bottam signal provides asperate indication for each RCCA in the core and ont, signal is sufficient to initiate the turbine rehack. Also, a change in nun as seen by one of the four escore detectors will osuse i j the turbine Iced to be redwed. The turbine Iced is reduced to a preset l value. At the anne time, automatic withdrawal of the control rods is prevented i by a rod withdrawal block. This acenario is discussed and analysed in section (
~
14.1.4 of the Turkey Point FSAR. .
i.. . l l[- FMW2-586 15-0R.5-0FL-E-119 March 18,195 hge 2 , 1' As stated in the F5AR, diverse actuation of the,* turbine runbeck occurs by ., either of the follwing:
- 1. Reection of the load reference setpoint of the turbine governor speed by a preset moet. 1his is econsplished by reecing the I
i changer (See Figure 1) l setpoint at a constant rate for a prese@ time. ;
~'
- 2. Reduction of the turbine Iced limit to a preset value. De load limit '
1 (a relief valve which limits control oil pressure) is reduced s til i turbine thermal load as sensed by either of two turbine first stage l l pressure channels is below a preset value. (See Figure 2) j ~ i 1 Although the turbine reback syste is not anfety grade, the syste is designed l such that no single failure could prevent a runbeck from occurring for a single. l l J cr multiple dropped RCCA event. l The design of the automatic turbine reback is prone to spurious rebecks ! l rebacks not comed by an RCCA drop) boosuse there is no coincidence (i.e., logic med in the initiation of the rebook. The, a single failure of an l j electriosi component (e.g. burr >out of a rod position indioster signal, failure
- of one escore detector, etc.) instead of acting to prevent protective This action semes will in fact, cause a turbine runbeck when it is not needed.
} m necessary plant transients and results in a significant loss of operability j and availability. Operating history at the Turkey Peint units shaus that t spurious rmbecks have occurred over the years. 1 l The majority of the spurious rmbecks have resulted from failures in the flux l } rate input to the rebook logic. To alleviate this proble. Westin$ouse t
} performed a asfety evaluation in 1982 to show that deletion of the f1a rate (NIS rod drop) signal frem the rebook logie is soceptable. The evaluation .
! 7 __ considered the effect of the change on the dropped RCCA enelyses as presented l 1 } in Section 14.1.4 of the FSAR. I l
~
In any accident snelysis, a limiting single active failure for that transient is assined. With the change, it is apparent that for a single dropped RCCA a l
' failure of a rod bottom signal would result in the failure of the turbine ! re back. Thus, the single dropped RCCA esse was reanalysed asstaning no turbine rehack. The results shou that the static RCCA misalignment accident bounds the dropped rod if the plant resins in menus 1 rod control. It was also stated that the dropped bank (multiple dropped RCCA) analysis was not affected since multiple dropped RCCAs generate several rod bottom signals, and only one is l required to initiate the turbine rmbook.
} Houever, the evaluation did not consider the means by which the deletion of the i NIS rod drop signal would be schieved. Art modifiestion to the systen should Merely l be such that the single failure criterion een still be' met. 1 disconnecting the NIS signal effectively defeats the turbine governor control syste as a means of rmning book the turbirs. As noted above, the original j l 1
FE-85-586 NlbOPLS-0PL-O-119
. March 18,1985 Page 3 system provided diversity in that the turbine rmbeck is achieved by either the turbine Imd reference rek: tion or the turbine Iced limit reduction. The NIS rod drop signal provides input to both means of turbine rebe6k, but the rod bottom signals provide input only to the Iced limit. Thus, with the NIS rod drop signal disabled, a single failure in the turbine Iced limit reduction for the protection train could prevent a turbine rmbeck from occurring, even drop of multiple RCCAs. In addition, discussions with plant personnel indicate -
that the cabling between the point where the rod bottom signals connect and the l point where the turbine rebeck trains diverse is not redundant. With the NIS rod drop signal deleted, a single failure in this cabling could prevent a rm beck from occurring. l Westin$ouse recommends that the turbine reback systen be modified such that no single failure could prevent a reback folicn#ing a multiple dropped RCCA J event when the NIS rod drop signal is deleted. This could be socceplished by j ohanging the logic so that a rod-on-bottom signs 1 provides input to the turbine governor control system. (Figure 3) Also, the cabling noted above could be i made redundant. In addition, the effects of a loss of power supply to the rod l position indication system should be evaluated. (i.e. Are there redadant power supplies, or is a loss of power fail-asfe in that it generates a rmbeck?) i l With the single failure condition met the 1982 anrety evaluation and subsequent - I
~
NRC approval would remain valid.
! If there are any questions, please contact the undersigned. , t Very truly yours, )
WE3TINGHOUSE II.ECTRIC CORPORATION 1 __
. . kM D. J. Richards, Manager l NSID Projects
!' - JJD:pj . Southern Area oct R. Acosta, IL, IA
- 5. Shepherd, IL, IA R. Diaz, IL, IA i
I i . l l 1 l l l l l 4
- -.. --. ,,..n. .,---,-...-n..n,-,---
, , , , , - ,_ _,,___._-- ,--_w.- ,_,.,- ---,,--_ . -----.
1 I i FIGURE 1 i i TUR81NE80VERNORCONTR6i. SYSTEM -
. I l
twes.ma tone maessa.ca ensue. view I i -- --
* -)
- . . .,_ s l SvtRTEs9PERATWEE, SwtaPwsa l toept &T(93) 8 ,st s 2 0 0 emSte47(%)
8 DROP ti&N AL. (t/4) I DUM 4 TeME
- 8
! Tioga!ba6Av i . SELAY l
., RELAY 4(aLhv cycts)test .
* ~
, --( b-- a se sat. e-no asc.
l
/) 3 Er LK1 t- -= y' l
1 I a - ) .mm ,._ 0 i i 1 e 1 l 3 a!DUCS LOAD REPsetsNfA
= soog/nns. .
l == - e l g arnunniuv / S i e
- e* *
. ~ .' l , FIGURE 2 TURt!NE LOAD LIMIT REDUCTION PROTECTION SYSTEM TUS Sitet. Le nt 4004 1 R4 mpr.Teest .
A I \
. 001E M ROD hTO9% l DROP SetedAL es&d&L 8
('h) (Assy ROD) ' (inun s) l l e e (FsRST S1a4L Tuse RE)
' O E I I . T ;
i ~~__ ' () r h e
~
e REDUCE. LSAS 44217 ; l -- ' --
-,.Z,.oo,/ I I
e a _- -e--w
. u
, FIGURE 3 REVISED TURBINE GOVERNOR CONTROL SYSTEM ~',
Tuate m EgygggeeC5 RStutTeted . A
' ~~
OWERTEMPG5A b i OvtRMA MGM&T(%) Med M($'J) 8 % 8 DROP N A L. ( ) UD WM i SICAIA L. e hWh (4Alf d4D) o m o .
. T I M f. TIME SEbAV -
DELAY RELhf LDS.C l ..
., RELAY (CYCLC) - -
, I F-..is ut. ,o-en uC.
El Er ~_ U y- - -
- ,- . y e. ,
_ , I O l. g at9 WEE LOAD REFERt8M4 ATSOOS/MSL , l -~
. . . o f.T /
b w-2---.--,.-e.---enr.-- , ~ - - - - - - m--,.,w-,-c , .,
l
. . 1 m
, FIGURE 4 REVISED .,
TURBINE LOAD LIMIT REDUCTION PROTECTION SYSTEM ; TURSitet. LOAD LtMIT REouCTsoni A
<- __ - __ s S# __
Dec' ROD M TOM NAL a#6NAL (AMTROD) f (ta m e) l l y b ' TutStWE M WEm )y4 (FsR87 STAet TV4 tant Mts&uRE) e ~
. r..L o
i e \ ' ' 8 REDUCf. LaA0 ums? I- - . .. .. O hk 1 W'
' ,9 s -
Arrecweer No. 6 L ',#lE9'} 'A 4h#W
- NE..
.\ J'w.. .)__ .,Ts %.s. upo
$.Ti w-2o gos 'bgDJ /.RA ' . . -
pdd #a.m pe w 4" t . . . . rn sea, vestincome provided te n. ride tower a 1,fght e asfety ev.1 a modif16ation to the Lufbine rmbeuk system et furWey Point whian concluded that the deletion of the N3 red drop (or flux rate) 885nel input to the ~ turb,ine rmbeek was soeeptable. Since the turbine rmbeek provides protootien for a dropped ACCA/ bank seelder.t, the thenge was reviewed with 'rdspect
.tooident.
With the change, it is apparent hist for s' single dropped ACCA a eingle fa!!ur
. of a rod bottom signal would result in the'talavre of the turbine rmbeck.
l - Thus, the drepped red ease was reanalysed sesming no turbine rmbeek. The resulta shew that the etstle IICCA m1611snsent seoident bounds the droppsd rod. It wu also stated that the dropped lenk analysis was not effected since e dropped bank generates multiple rod bottom signals, and only one is requir to initiate the turbine runbeck. ! Hewever, the evaluation falled to sensider that the deletion of the N18 rod drop alsnal effectively dettata the tvrbine vsener emmecontrol systen sa a tr.eens
._ px
- 2. - .. : Wr wy.
or rw an g b.ek the turbine. pe. omm; m.u ... x.: namrr, uo - m .m; eo=a . om-eg
- n -a - . , .
1 ga4:ng t,urbine rmbeck, ? ' $8 r 2 f a iS5 % P ! W a u' ta . deer. thus, a single failure l'n the turbine lead limit reddotion i protection train ocu3d prevent a turbine rmbeek from occurring, even for the drop of a RCCA benk. ' l
.I i
Bince this possibility was not evolusted, Westinghouse reconrands that the N13 . Pod drep signet input to the turbine runbeck be reinstated, mtal additional tysluotien of the problem een k siedv. REC'O B Y i
. rELECOPitR l'
pgo PLT ENg ure. jpEQ
._ -_. - _= - . . _ . -- _-. -
~
q/ E RAT-PTA-C-244 o '. NS-0PLS-0PA-85-293 Nuclear Safety Department t i FROM: Risk Assesment Technology ' WIN: 284-4707 .. DATE: August 5,1985 E8 JECT: P-9 Evaluations Keywords: Safety Analysis $tandards/ P-9 Evaluation /PI-85-013 ; Loss of Lead . Ref: MS-RAT-PTA-85-216 ' / 1 10: PTA personnel
- CPA personnel .
4 oc: J. L. Little : R. A. Wiesemann j D. G. Bevard
- D. Misner/WN1 eruereAs M. Cantineau/WNI Brusels In the referenced letter, a ocamitment was made to write a letter for inclusion .
in the Safety Analysis Standards (SAS) to provide clarification of the asseptions in the P-9 analysis (deletion of reactor trip on turbine trip belcw 4 the P-9 setpoint 505 power). The evaluation to support the asfety analysis guidelines for performing a N9 evaluation will be docmented in a cale note by l j September 15, 1985 and at that time, the letter of clarification will be i i included in SAS neber 6 (Loss of Lond). In the interim, the guidelines ' l outlined in Calc Note Nunber CN-RPA-80-91 should be folicwed. Please contact i , the tmdersigned befora besinning any F-9 evaluetions. l .
=-
i d'L. - M. A. Grace l! ,
, Plant Transient Analysis '
i b . ~ l+ N P.A.Lof6is,M ] Manager M. P. Osborne, Manager Operating Plant Analysis Plant Transient Analysis . i i l , i l i I
- 3. Dropped Roo Analyses for Turbine Runback Plants background on the Allegation 1he 411ecation reports an error in the analysis methodology for the dropped rod event for plants using the turbine runback feature for matsoation of the event. It states that the error was incorporated into several analyses and that thm analyses were not corrected when the error was found by Mr. Heberle. Mr. Heberle's analysis shows that the results are non-conservative only for those cases in which a reactor trip occurs via en overtemperature delta-T signal. The other cases are not impacted.
The analysis procedure was developed in August 1990 (NS-TA-0* *65) as a result of a separate P! on turbine runback plants (PI-S'-210) which was closed out in August 1994 (NS-RAW-94-580). S t a ty s,.and . Sa f et.y. I mp ac t. , At the. time the error was discovered (February 1995), a review was made of affected plants and it was determined that sufficient margin eHisted for the plants pending re-analysts of the accident. The following list contains the impacted plants (different than the list in the allegation) and the status of the reenalysts. For those plants not yet reanaly:ed, the availabliity of margin has been reconfirmed inf ormally by NFD. In addition to DNb margin, there is margin since the unanalyzed plants ~are not at the beginning of the cycle, which is when the transient is limitina. The reenalysis is therefore required prior to startup of the neNt cycle at the latest. Reanalyses will be done prior to or as part of the Reload Safety Evaluation.
~
Turkey Point O&4 Reanalyzed CN-TA-SS-6'(Error discovered here) Indian Point 2 Roanalyzed CN-TA-85-65 Indian Fotnt 3 Reanal yz ed CN-T A-05-91 -- Surry 1&2 Anclyzed correctly for uprating CN-TA-95-47 (Prior analyses are Vepco scope) Point teach 1&2 Reanalysis in progress . G1nna Not reenalyzed yet San Onofre Not reahtlyzed yet Deznau 1 Per E-Mata from WN!-Seussels (attached) ' I' - -- none of the cases trip on 070T. Therefore, no impact. Zorita ENUSA scope. They may or may not use the procedure. All other turbine runback plants have non-Westinghouse fuel and the 1980 methodology was not applied since the analysis was outside Westinghouse scope. Ac 41,or). t e _ Rate,91,ve, ,
-!ssue letter to SAS holders formally documenting change to procedure by 0-9-05.(Osborne)
-Complete reanalysis for Point teach by 0-30-85 (Loftus).
-Perf orm reenalysis f or Ginna by S-30-85 (RSE dus 11-05) (Lof tus) .
-Perform reenalysis for San Onofre by 11-10-05 (RSE due 11-85) (Lof tus) .
-Notify ENUSA of change in methodology (same letter that changes the procedure). -
1 i
=
l i l
1 d, '
%= ?- ~
e i I I
; , NS-RAT-PTA-85-238 i
Nuclear Safety Department
- FROM: Risk Assessnent Tednology L
! WIN: 284-4357/4303 1 DATE: July 30,1985 E8 JECT: IPP Modification to Turbine Ambeck system !
l l l )' To: J. Gasperini R&D 18 Keywords: IPP ! Dropped Rod ! oc: P. A. Loftus MNC 4-09A Turbine Am beck l
! G. P. Roulett MC 4-09 Indian Point i j 0. Meewis R&D 18 FI-85-013 :
i J. L. Little MC 4-17 ' R. A. Wiesemann MC 4-01 ( L K. Figenbeun MC 4-01 ' l In 1984, Westin$ouse provided to Consolidated Edison a asfety evaluation for a ' modification to the turtiine runbeck system at Indian Poirt Unit 2 which I l concluded that the deletion of the NIS rod drop (or nux rate) signal input to j the turhire reback was acceptable. This evaluation addressed the transient i analysis impset of deleting the num rate signal on the socident analysis. ! 1 j However, precautions must be taken in the hermare implementation of the nux ; i I rate signal deletion such that diversity is not rammed from the turbine runbeck ! protection system. This is consistent with Diapter 14 of the 75AR, which states i
!
- that there are diverse means for obtaining the reback. Westin$ouse recomends l that the deletion of the NIS rod drop signal input to the turtune rmbeck be ,
i implenented in such a manner as to mairtain consistency with the F5AR i
) description of diversity, which secounts for certain sin 61e failures. The i
follwing provides a background discussion and femal recomendation concerning 1 i this issue. The automatic turbirm rmhack foeture at Indian Point Unit 2 provides protective h action in the event of a single er multiple Aod Cluster Control Assembly (RCCA) ' j drop. Detection of single or multiple dropped ACCAs (including dropped ACCA banks) occurs by either a rod-on-bottom position indication signal or by a change in neutron nux as seen by the escore detectors (NIS rod drop signal). f l The rod-on-bottom signal provides separate indication for each RCCA in the core l and one signal is sufficient to initiate the turbine reteck. Also, a change in ; nur as seen by one of the four encore detectors will came the turbine Iced to ' j be reduced. The turbine Iced is reeced to a preset value. At the same time, automatic withdrawal of the control rods is prevented tiy a rod withdrawal i block. This scenario is discussed and analysed in Section 14.114 of the Indian i i Poirt FSAR. 1
_ ~ . _ ~ -___ _ .__ _ _ _ _ _ _ _ _ _ .. 7 _.__;._..__ f, J. Casperini 2 Nb RAT-PTA-85-238 I l I As stated in the 75AR, diverse actuation of the turbine ratisch occurs by either of the following: t
- 1. Reection of the Iced reference astpoint of the turbine governor speed l l ohanger by a preset anomt. This is accomplished by rescing the ,
setpoint at a constant rate for a preset time. '
- 2. Reection of the turMne Imd limit to a preset value. The Iced limit ;
(a relief valve which limits control oil pressure) is reeced until L 1 turbine thermal iced as sensed tiy either of two turbine first stage ! { pressure channels is below a preset yalue. . l 4 Althoudi the turtiine ratisek system is not astety yede, the system is designed ! such that certain single failures could prevent a ratisek from occurring for a ! sin 51e er multiple dropped ACCA evet. ;
- The design of the automatic turbine retinck is prone to spurious ratiseks (i.e., l l retseks not catsed by an ACCA drop) boosuse there is no coincidence logLc use'd ;
- in the initiation of the rmhsek. Thus, a sin 51e failure of an electrical l j oomponent (e.g. tiurtwout of a rod position indiostor signal, failure of one l 1
encore detector, etc.) instead of acting to prevent protective action will in ! ] ~ fact, canne a turtline rateck when it is not needed. This seuses anecessary j plant transients and results in a significent loss of operability and availability. Operating history at Indian Point. Unit 2 shaus that spurious f ( retmeks have oocurred over the years. ~ f f The mejority of the spurious rehacks have resulted from failures in the flux ! rate input to the retsck logic. To alleviate this problem, Westingiouse ! performed a safety evaluation in 1984 to show that deletion of the rius rate j ! , (NIS rod drop) signal from the ratisek logie is noceptable. The evaluation i considered the effect of the change on the dropped ACCA analyses as presented in ! Section 14.1.4 of the F5AR. I In the dropped rod accident analysis, certain limiting single active failures t for the transient are considered. .With the change, it is apparent that for a single dropped.RCCA, a single failure of a red bottom signal would result in the : failure of the turbire ratsek. This, the sin 61e dropped RCCA esse was reanalysed assusing no turtsne rahsek. The resulta show that the static RCCA miss11piment sosident bounds the dropped rod if the plant, remains in manual rod , control. It was slao st,ated that the dropped bank (multiple dropped ACCA)
- analysis was not affected sirme multiple dropped ACCAs generate several rod l bottaa signals, and only one is required to initiate the turbine runteck. !
However, the evaluation did not consider the means by which the deletion of the NIS rod drop signal would be schieved. Arir modification to the system should be ( I such that the single failure criterien een still be met far the types of sin 61e failures considered in the origins 1 analysis. As noted above, the origirsi l; system provided diversity in that the turbine rebeek is achieved tiy either the turbirm load reference reduction er the turtline lead limit reestion. The NIS rod drop signal provides input to both means of turbine rebeek. Thus, with the l NIS rod drop signal diashled, e sin 6le failure (e.g. en open circuit) in the I nor>reendant channel that transmita the rod-on-bottom signals to the turhire i 1
- _ _ _ _ __ E -._ - _ _
,.' l i J. Gesperini 3 NS-RAT-PTA-85-238 e : i i l load lise,t and turbire governor control system will prohibit a turbine rmbeck. i i Therefore, protection against a single failure for either a dropped bank or a l multiple dropped rod event can only be justified if the failure ' occurs in the l rod-or> bottom sensor / transmitter and not in the non-redundant chiennel. Westingtouse recommends that the turbine rmbeck system be modified such that no ! , sin 61e failure could prevent a ratsek fo11 ming a multiple dropped ACCA event
- when the NIS rod drop signal is deleted. This could be socomplished by modifying the system ao that the rod-on-bottom protection is re&andant. (Figure 3
i
- 2) In addition, the effects of a loss of power supply to the rod position t indication systen should be evaluated. (i.e. Are there redundant power supplies,
- or is a loss of power fail-aste in that it generates a reback?) With the single failure condition met the 1964 safety evaluation would remain valid. !
i 1 Please transmit the above inferination to Consolidated Edison, to the attention , j of Art Ginsberg. Contact the meersigned if there are any questions. ' I N h = :- ,?" ! + la , ""f**' -
! G. H. Heberle J. Polevarapu j Plant Transient Analysis Plant Transient Analysis t '
i I M ! I . Approved: M. P. Osborne, Manager
, _ _ Flant Transient Analysis '
] ' .l I i i 1 ! I i i l i . l I L l .
. l i
)
I
FIGURE 1 TURBINE LOAD LIMIT AND TURBINE GOVERNOR CONTROL SYSTEM NotAESONEANT *
/ '
... , .c .comse.~,
C: tost Ovtt TEMR 47 (p)(4 mitt al) N'GM OvtR PDwtR l 900 D80P sieuAk OT (y4)(twit? 48) g (SNLET4) 8 pop seTTosa siswAL MCET 6) [ [ l l l a b ' TWA 8itti .
" THROTTLE PCSITICM esma mm a 70*90 2 CYC LI C 3/g CPEN -___ A t t. AY 9 -Wo.sset.
q - g l H '*d** Timt CELAY b RELAY q g,g.50 SEC. i
.M n
-D L. ,,
g mWM SACE l. l Of FI/bb Pee PSA M&. , 1 4 0
r a FIGURE 2 REVISED TURBINE LOAD LIMIT AND TURBINE GOVERNOR CONTROL SYSTEM 1 Notpeson m . REDurGAHr r_y-
. 2 - - .
eism ovtn tema nien m.1 mm l noe siemaa. noe sovvoan sie=AL. / AT(e)(ses Tn) av(v ,,(swarf n) g ( T (satti 6) l
- f , f 1 1 l .
g ll I 1 1 a b e i TW A SIM E o , , Tu AcTTLE POSITION
- e mu mm o 79 902 4
CYCLIC 4/g CPEN [
.2 --
AtkAT - He.sset. q
.M- - -
l
" - ! .Timt DELAY
,, , , g
**" " E L' AY d h1.5 50 SEC.
~~
ON O
~
l . i em ',' LEAD Llast? -
, nu uu ..
g Of FWk'. Pos PSA $4&.
- 4. NUI REPORIING APPAREN1 SAFETY VIOLATIONS 10 THE SAFETY REVIEW COMMITTEE
- 4. Analyses that Raise tho Reactor Trip on Turbino trip Sotpoint backgrourid on the Allegation the allecaton states that an error was discovered an an analysts to support the deletion of reactor trip on turbine trip f unction at or below the P-9 permassive setpoint for Comanche Peak. The alleQatton further states that Ms. Osborne was informed by Mr. Segletes on January *2 5 , 1985 l
via a memo (attached) of several other analyses which may have been performed non-censervatively. The sequence of events in a P-9 methodclogy referred to,by Mr. Sealetes considers a turbine trip, subsequent reactor trip due to a protection l signal actuation and loss of reactor coolant flow just prior to the time of trip. Status,and Safety.. Implication. ' i In June 1984, Mr. Grace identified a potential problem involving the reactor vessel inlet temperature assumptions made in the DNBR calculation for a special study performed for the Comanche Peak units. This study is l not a P-9 analysis as stated in the allegation, however, it is a sam 11ar j event. It was determined at that time by evaluation, that the Comanche Peak study was bounded by the FSAR Loss of Load / Turbine Trip and Lqss of ' Flow analyses. However, the formal revi si on to the calc note containinQ
~
the evaluation was not completed at that time. The cale note was cavised on 7-26-85 and includes the evaluation. Ms. Osborne became the manager of PTA in September, 1984. Since the Comanche Peak analysis is similar to the P-9 analysis, Ms. Osborne assigned Mr. Grace to investigate the P-9 analyses and to ensure that subsequent analyses did not include the potential temperature error, pending Mr. Grace's completion of the P-9 review. It was determined as part of Mr. Grace's reve4w that the loss of flow
~ 71 -_ assumption prior to reactor trip could result only from multiple failures. This is because of a turbine generator motoring feature which maintains power to the reactor coolant pumps for 30 seconds. As such, the
" loss of flow prior to time of trip (no motoring) was overly conservative and did not need to be assumed ih the analysis, although the assumption had been made in some P-9 anal yses. The appropriate assumption is to l model the loss of flow either at the time of the initiating event or 00 seconds after the initiating event. The transient is thus bounded in both i cases by the loss of flow analysis in the FSAR. The non-conservatism in the vessel inlet temperature modelling is more than offset by the ,
assumption of loss of flow prior to time of trip (i . e. , prior to 00 seconds after the initiation of the event). Therefore, the conclusions of the existing P-9 analyses remain valid and there is no impact on the plant l safety. This was the conclusion reached informally by Mr. Grace in ' J September, 1984. This will be formally documented and reviewed in a cale note by 7-31-85. The informal letter written by Mr. Segletes to Ms. Osborne in January 1985 was responded to by Mr. Adler saying that if the transion't is t:cunded by another, there is no need for a PI. Mr. Segletes did not takejout a PI at d 1
, - - - . . - . _ . - - , ~ , , - , - . - - - - . . - - - - - , . - . -.- -- - , - -
- - - - - - - - - - - - - - - - - - ~ ~ ~ ~ -i
s this time. e The plants with P-9 analyses are listed on the following page. All analyses performed after June 1984 have the correct modelling of the temperature as well as the overly conservative assumption of a loss of flow prior to 30 seconds of turbine conorator motoring. The DN8 deston basis is met for these plants as well. .. f Action to Resolve there is no standard procedure for psrforming a P-9 analysts. It is a combination of a loss-of-load and a loss-of-flow. The procedure for performing these two analyses is contained in Safety Analysts Standards 6 and 13. A letter will be westten for inclusion in the SAS manual providing the clarification of the assumptions for the P-9 analysts by 7-31-85.<%"The calc note formally documented the evaluation will be completed by 7-31-85 (Osborne). 4 [,y,, a a_ M S - RA T - FTA 244 A TT AcH aED e e
ge
h-e e
=
I: l Jhyg, l l\ f )2(s el de i s e E c o *.n ye rkr4sy ws hs 73 % /$1rek ba{osrt yHsis9 ca/c a.+c cAf - TM -9 r-t1 o.,,/ a . /ra A. A * / e ve.s . a)/m.eAsJ /D if. Yhe ito M eve s __.ons . r...evee h. hr t%/sja a.cLi49 bae. W A u <- . rome o e i:,_.h e .. yteaf . cAc<1 -he. . covte >2*fe. . Z ask d iser .k.__do fns je.en.r.e inn Kr .her_ rc sponsdll:b _ per. . )).s_ A+7_ Z q-3.... e d pg -Astr~ _ rcr- 3 (H. yze ti />
- stfis 5dapon d~f review <<) .
. . h e !/ N ... r s % J - _ % e*le.._ noh.-.M.h kee Ni'n= ho4..
_Y4eNi . ,ee i _ themlL.yo __ou.+ an.d__fdd_.ouy_owir_JxsOm . _ 7dn /$f<oklari, _ did . ched _ ike_Colc__ Dote ,_kui_),<. bee sa teh-aM+ sk< J jt k. - (m y o,,k f. fr tik. o.~K. r%/iWt Jesp .*i< Ev>w).
**> m . T ask.ad_.hr4 ..Q Ke._A-d it.u w .4u.t.L hse _._na L _._..
, .- . To n Gi d hs. U. ._ _*I AskeJ hS__ip_ Jue_.tensenkeed -
. . . .. wk+k i+ ...e.n.r ;<d. _ Tow,_.coU_ haJnd. . r a .r h a 4 - . h k _ . - -.
= . is he u .s 6 w n e.. . M 'L% _ Mg ec. (A sti k ._.th
^ ^^ au1._ co.,a)...
. wit._ sappos<d b a ssaks,-. th e. cria mn esubw . T o m . - _. _
Sld c. k^6L6 . 5 o st. _ ._4,l10 . h4enho.Ae d. .' fha 4_ .f5 Yhe ule.,+1n1% r <.- y a v. f _ h <. . w a e h :- / tv c~ < ky . lv 1.L _(xi.e Aoa4 a#/9 s0 -6 c m ,.. p ,.. . /m 9., s.r/ f. . s._ W ,c .. ckssk sr.. .
-e
- e. me .
._ me-.e- e ._e b
b/ / 1
. _ ~ . _ _ _ . _ . . _ _ _ _ . _ _ . . . _ _ _ . _ ,
~
, /1/ T ThuerAay 2//yl95 (
~1h,'s no+e. documts % conuteswhin Z had wi b,
//stih. Os b><n< i), he, ya e 1M AM. Malih ChIlad ms In+o her ofpii e **~ d cuan +d h $nou) M m ik knA r 4 by by seve.ro l me=<a. .C had sent L ee,
'Thss e u e t n o t.e d -
(U Le+< pr-fer- rne -9r-es 1 tAois itp rH clerihbis on thC Cwree + fr*Ctk *n o . ** f Ars e % r e y ..- + a p k .h4 ss,cedy co u <en -/c h e sap u y tevi<<J
\ ca m ,%.
l2.) A mk 44ed 1h/9r regner h9 went }"sAher% the r<. i.e pse nef .p. tlem ig tA <. oseinssi (s tos ) e4
- ; cen,. w + cycIe f.r h, s. 1+s.I in derermu. /%/ -
Metroal tepasi<wks-(A>s- tsr sa-o.% l3) A Rr'an Aerssres a1 qu. h h Accuram /eaci~, thef
.C hid her . ws .Chowld abode b y taAsn sendri> f ou +- +hsQ'Y?.%. fef.e.ew <.s fIs e + - $*-e 4 -sye f'so icsits i
.itfs k pore %ko I phce if spe< *fsrd co+s 5 tt. fide ~m~b i
ratet M shmld be p//en d uhen ka,*rm:Ikn.g fs lim;n a rey do cuwtsosis. Af) 14 riott AEP.$As & D. treka,Json do. k of In >.sl, If 9.t' pors b iy o n. + he p. t h.cl onc of 0"r Ct.shssers u.re e %f ain1% l no on< c Iled hsk hse 4 { reya e disig
- o. in ss+d9 h e w k-nl c,uds. (Ussiss9ou*sto)a. h p.tI suos th a,e.9se he origior a fly re.jsa <s M $b k<. colled .
/4e 4 +es ts, / ris. Messyh + sti e,e fn e o bev<
s<< re. ih k + / h ho.tra. c 4 her h e w g.e op yn e,
};bar perpiu ue o.ppeair6 I sAa. ha.d 9iuk m < .
.1 hid hse tm t Z w lI) tus t.ase t beese-r<. % h it-
-, . d,,;, .
., 9,, o, % m, u,.a x, u
d L43abh of. % c. Mcmo 36 d h, f P Q fas ha.4 .pv44
~~ t,us o s. o r. r - w as Pi-+ An.g r6 .W F1-+
4 rar vi b 4 r4 n ly re i . I %let hv Ihr ~2 -4 1 we c<1.v r +. 4M. frehw Q fac,hissI fL fsde < w br 2 M .s a.i. ..t h (ts +vi L &e - ht i.r, I u 4<J % pi1au/ % ! Corr M , puMe. . M li+v proce.t o/ k. 4eil m<. A+ ip I m + b c.<- \ a q .v.a j we. lit co n o k ,: sh e. weta( M Mrmia&4.c u d Wh ttw woeN ! cc hu N ., wat s 6< % ~ A 4.s w & , tw. *1 e e e e <- + a.t My pc N j yp Pet [a l of perf em e4. i Ta.n 3o, l819f . -
'E %14 4a.Wh. T w. 4A spv<
c.4. c. f4 I c.,s a. 44.A
-to b t- & p.c 4 4.. Ii by I
M <.) s h s c a %. + o.4y ptu* gu..c W.w or 1 h w+ x IkA si.~ t 4 be. h..A .
~
'- %i
, f(f " ~ " ^ l .' 1 '. . p t * ^ M v.c- b tl 9 ft. Sh.e. v a.n H.4 n. m e e e, w = 4 c..r (%t L i t .( % p) ker to
,r5 pea u.a a. wri % fa p.afe ) .
1 o,cE4d A4 4ik. ip. C k<. ka,( En rsa r Diesdeu , t'cvJc 4 m.g (pu.as h k atre takneh h,e C haj yru4
- s. ~ L 4
,r- e . < h .'< . r h a saia ehe k 4 i,. ti M tAa. u..nid do so bcpr.e She fcfuen<d frvm 2+s1
.T MJ hu I re m M d in e 9 Cveer Qparfunsh& Sullehis \
eAg Lf L p us o f.esI job: +ka+ a y aled fu ne e . A a.sks.A & Me. lo cerc.it +s,e jobt I m ags) 1 c n ig \ . fill o J 9&c it h her k
- ne tew<.u). E dal prodde he< wik b u lla t m o c S s tep, < +y,,,( \
Neliin u Led me // Z M 'fpliid ou hide
9 A Weehhhn.y . Z fold ha s hd not, %+ 1 u ,. y h,
.ree aJ,a./ deudop:.4 A nda MeAng A4< bQ.w r % <;,(
oc, hide.. /ta;,h. .ru.yyu w z n e e i, pied soon ~4 sico ragge, M Z %% to h*h h k . 2/Iy/or S*f . O h-
?
g i
\
\ l l
/ 9 2.C '
Y, E RAT-PTA-B-051 ECPLS-OPA-$-046 FROM: Risk Assessment 1;chnology WIN: 284-4481/4901- - DATE: February 12,195 ' ~ SJBJECT: Clarification fe Reporting Potential Safety Itans to the SRC l REF: ERAT-PTA-B-047 KETWORDS: Potential-Safety-Ite/SRC/ Procedure TO: J. A. Segletes . cc: R. A. Wiessmann MC 4-08 E. P. Rahe MNC 4-12 D. C. Richardson MC 4-15 J. L. Little MC 4-09A l ! 1 In the referenced letter, you requested clarification concerning management ! approval fcr the opening of potential safety issues. You are correct in - stating that safety concerns may be reported to the tsunediate manager, his l representative on the Safety Review Caanittee (SRC), or to the secretary of the l SRC. The information distributed at the OPA/FTA group meetings which contains ! a list of doctaments for manager approval was handed out informally as a l guideline for une by the members of FTA and OPA. The intent of this handout is to ' 4 help ensure that proper review by management is done fe those doctments - requiring review or for those doctaments fcr which the smdersigned consider review to be prudent. This latter concern is the purpose fe review of opening
. potential itens. This is also consistent with the procedure outlined in the
~ RAT Irutruction and Guidance manual, ERAT-E9, which states that an employee should notify his manager of a potential marety ita. Also according to ERAT-L9, it is the responsibility or the manager to determine if the issue needs to be reported to the SRC (e.g., approve a letter). However, abould an employee disagree with his manager, he has the right to contact the SRC directly. This is also stated in ERAT-59.
w6hm M. P. Osborne, Mansaer hispu P. A. Loftus, Manager Plant Transient Arulysis Operating Plant Analysis { /sh . b I
- 4. VIOLATION OF QUALITY AISURANCE PROCEDURES
'I A. Transmittal of Preliminary Draft Responses Bac k g.r ou.n d.. on the. Allegation Mr. Sealetes alleges that the Functional Requirements for the Italian Reference Plant were not marked preliminary, were not reviewed in-house.
and were not approved by a manager. This is alleged to be in violation of l procedures. S,t a,tus_ and Saf e.ty_Impac t, ! The preliminary copy of the Functional Requirements f or the Ccntrol and , 1 Protection System for the Italian Reference Plant was provided to SOPREN, the Westinghouse licensee in Italy, pursuant to the existing engineering consultation agreements. The cover letter of the transmittal (NS-TA-83-520) is attached. Contrary to the allegation, the transmittal 1etter did identify the " Preliminary" status of the documents. It identified that the documents were being transmitted to SOPREN for their
! review and comment and that could transmit them to other organizations if they so wished.
- Engineering consultation is characteristically provided to a licensee to I
supplement inf ormation which is normally provided under a license. In this case, Westinghouse was contracted on a time-and-materials basis to provide functional design documents which illustrate the design and verificaton of the Control and Portection Systems including special Italian plant performance requirements and systems for coping with Special External Events (SEE). SEE's are not part of US licensing requirements. That the preliminary Functional Requirements did net receive first-line management approval and wre not subject to normal in-house revi ew -i s l' explained by the limited scope of the order. Westinghouse did not protray the documents as having gone through a detailed review or as being final. Resonsibility for the correctness of the information supplied remains with the licensee. As stated in the transmittal letter, the documents are recognized to be working documents which SOPREN will periodically revise
~
to reflect evolving plant design. Should SOPREN desire that Westinghouse perform an independent review at anu time in the future, such would be subject to contractual negotiations at that time. There are no safety implications as a result of this issue. Action to Resolve ' None required. D e
, ,. ,r m w- i ._ _ ___---- - - - - - - . . - -.c. . . - - , ---~_-.-----,,n~--n. -r,.,. ,-,,-,--w,,n-~--,,-v
l
. l
# 2. )
1 NS-RAT-PTA-85-047 Nuclear Safety Department FRON: Risk Assessment Technology *- WIN: 284-4358 : DATE: February 7,1985 - SJBJECT: RB20E3T FOR Q.,ARIFICATION GI 1NE PROCEDURES FOR REPORTING POTDITTAL METTY tYstf7RNS 'ID THE METTY REVT5v COiiiu n d (SRC) KEYWORDS: POTDITIAL-SAFETY-CONCERNV SAFETY-REVIEW-COMMITTEE /SRC PROCEDURES 70: R. A. Wiessmann MC 4-08 CC: M. P. Osborne WC 4-09A P. A. Loftus MNC 4-09A
- E. P. Rahe MC 4-12 D. C. Richardson MC 4-15 Clarification is requested on the proce@res that one may folicw in reporting potential safety concerns to the SRC. This request is made because of 'an apparent inconsistency between the notice regarcing this matter that is posted on the main bulletin board of this building and information that was recently distributed to OPA and MA personnel,
'i The notice on the bulletin board states that safety concerns may be
' reported directly to ones supervisor, or to the manager's representative on the WRD SRC or to the secretary 'of the WRD SRC. On
the other hand, Itan 4 of the attached sheet, which was distributed to all PTA and OPA personnel by M. P. Osborne and P. A. Loftus, requires that all SRC letters requesting the opening of a potential item be
-__ reviewed and approved by either of them. Since opening a potential itan is how one reports a potential safety concern to the SRC, Itan 4, in fact, requires all safety concerns reported to the SRC by PTA and r
OPA personnel to first have their manager's review and approval. This appears to be totally , inconsistent with the reporting of safety
, concerns to the SRC as posted on the main bulletin board.
Please provide clarification of this issue. An early response would be appreciated. (A.
^
J. A. Segletes Plant Transient Analysis Attachment i t i e
. b(Y
\
ITEMS RECCIE NG MANAGER EEVIEW AND APPECVAL S!CNA
- 1. Safety Analysis Standards (r.ew ar.d revised issues)
- 2. Analyses which are based en draf t standarcs or ncn-cenfigurec cctes
- 3. RSACs and changes to RSACs .
4 Saf ety Review Committee letters
" -request for PI
-c1csecut cf PI
-c1csecut of identified Safety Issue
- 5. Safety Evaluations .
-Neclear Safety Checklists-ESES and RTSRs (blue sheets from NT
-safety evaluation reports (e.g., weekend specials)
- 6. Nuclear Safety Department letters
-Fres Rahe letters
' -Ecb '.'iese= ann letters and proprietary content review '
- 7. Form 36 (WCAP release) .
E . T-She e t s ar.d C PADS .
- c. Change Ccr.tec1s (impact and ic;1ementatien)
- 10. Requests for Quote
- 11. Technical Descriptions for Proposals
~ ~ ~ ~
- 12. Shcp -to open Orders 5.0. (if not done by projects)
-to close S.0.
-to transfer charges between S.O.s(NTD Program Centrcl form) i
- 13. Ex;ense accounts and cash advances i 14 i Reconciliation sheets
- 15. Overtime requests
- 16. Audit response and closecut letters
- 17. Technclogy transfer ar.d licensee agreezerts .
- 18. Trar.smittal (written / vertal) cf accident analysis ~atabase d infermati l
1 i l
.e lS # 2e_
q.
,- p i; \l1c/st*
\ .
Meli.k - . 'I
\ "Eg th e, p ac.ssy . .g O.u t e= td$ Iw -cu l ca.ic. n o k, on ha eklMn at &
- c f*/ N P .__6 r\
b.ekne bp hol.ua . $~6 A.p w,T hkk. A Sa.f.4.h
. .. f to kl.s . e.d c F2 ln % .s.< 'f e S .. s % 4sL.s , a d @okIy o.tl . ssi)o.c...g .sk4t s %4 k we. b c.u .- d .
_ _ . . . . . ..Tb. p en b 1% .is % t . % 9 c .. c t w t.. + c,r . _ 'b + a g 12. __k o.r J, . _. o o .. ..hi + .' . l . .. [A j e.f . -{4 p = em he.< 5 (t=.o)
.__..._'Mtn[mur .._.D N t it. __. o es.a rs _. .. t "i".T '3 o . 54 c 4 4.s . la . . .
~
". 5.e4..._c s s__.wsh.
% e..__-ha 7.~ % (tsiA,) ^- 3.-f
-'_ __ oland g ._ %.c t1 h e ... p4rs o ' L. . b p a lei a c < ,, be~ __
.___.... rec.hhia.d__.b9 % r p r e i d y _. T A ..V l - a r _ d. a.. pnc.bA .
e +. ...t.i. d e___ h. T M 1.0 c . . _ _ _ . _ _ . _ . . . . . . . . . . . . . . . . . .
'T M ._. p c. 9. .
__ w 3 _.t J , i _ b . k J . b oy.. A .le Sr . . . .
' E % i k. h n*r..___. s w s J ___ b e. r.<- y a e+.<.J_.. .h .. .h e. _ _5 R. c ._.
T.- ._c's. c .po A.t[*.I _t k . ____ __._....._........_... . . . . __...____f35.~s.c._.p4._Ae.- .. 3 e a r_ _.c. % . t.f ... . o his ' l . % ^ * * n -
= . . _ . _ _ _ . _ . . . . _ . _ . . . . . . . . _. . ..
jr.
._ . . . . 0bd .. .
-i f .- -
y.kW A p . . . . . . . . . ... . . . . . . . .
.. _C: s.D.UE c.sk 4 den rw'9 E m w sc 'J h w.M s y uds pa-asji.stH.;y&au
. . . 7X;- a R.E E S. /,Yf/.2A,.
1 .. l i
~ ~
I Aif f
4' P-9 ANALYSES s 30 Sec. T-G Correct T, h r21e. Nete h Meterina Medellina.n c w ents
- 1. FL VFPL :
Turkey Point CN-RPA-77-98 4/20/77 .No No - (1)
- 2. ILW Beaver Val. 1 CN-RPA-77-171 8/24/77 Yes No (2)
- 3. 10R ICRI 1 CN-RPA-77-208 IV8/77 , No Yes (1),(3)
- 4. ANG ANGRA CN-RPA-78-52 3/1/78 .No Yes (1),(3)
- 5. NSP/NRP Prairie Island 1 and 2 CN-RPA-78-66 4/ 26/ 7 8 No Yes (1)
- 6. ALVAPR
. Farley 1&2 CN-RPA-78-74 5/1V78 Yes Yes (2),(4),(5)
- 7. SNP -
SNUPPS CN-RPA-79-12 V16/79 No Yes (1),(3) .
- 8. NAH Seabrook 1 CN-RPA-79-141 9/7/79 No No (1),(6)
- 9. WAT Watts Bar CN-PP-80-03 10/20/ 82 No No (1)
. 10. TVA/ TEN
_ __ Sequoyah 1&2 CN-RPA-80-91 6/ 5/80 Yes Yes (2),(7)
- 11. DW Beaver Valley 2 CN-PP-81-132 6/1/81 No No (1)
- 12. DCP Catawba 1 CN-PP-82-44 3/ 17/ 82 No No (1)
- 13. GAE Vogtle 1 CN-Tb83-67 3/22/83 No N/A (1),(8)
- 14. SNP SIAJPPS CN-Tb84-91 5/ 15 / 84 No N/A (1) ,( 8)
- 15. (CE ~
V.C. Sumner CN-Th84-63 5/ 15 / 84 No No (1) 16.1GX/THX ' South Texas 1&2 CN-Tb84-95 6/1V84 No Yes (1) . l l , 1 l
- - - . ~ - - n. -- - - , - - - - . , , , , - , - - - . , - - , , , . - - - - . ,~-,, -. - ,.-.- ,,, zm,-..-a_ - . -n--
i
- 2-P-Q ANAf YSFS (Continued) '
~
30 Sec. T-G correchT, g rile. Note g Meteri ng Mariel l ina*D Cemnants
- 17. DCP catawba 1 N TA-84-122 0FA (Not Checked) 7/18/ 84 No N/A (8)
- 18. NEW ML11 stone 3 bloop N TA-85-011 2/13/5 No Yes (1) ,( 9)
- 19. NEU Millstone 3-El N TA-85-14 2/ 15/ 5 No Yes (1),(9)
- 20. IPP Indian Point 2 NTA-85-48 3/ 4/ 85 No Yes (1)
- 21. TWP/TXP Maanshan 1&2 N TA-85-63 No Yes (1) _
9 D
*6 h-r j
4 e
-r -~ - --- . ,w,, . - - - - ee. , ----w <e - - - -----w---------------em----,---n,--e w----n.--,e -~~-,m---r- - ------
4 JEEi
- 1. The cases with pressure control assmed the full 30 secor.ds of reactor coolant flow.
2. All cases assmed the full 30 seconds of reactor coolant riow.7
- 3. A vessel inlet temperature corresponding to the time of RCP coastdown was used in the DNBR calculation.
4 The ILW analysis was evaluated and used for this plant.
- 5. This cale note documents that all of the 505 (P-9) power cases are less severe than the F3AR Loss of Load analysis. .
- 6. Only 1 case was evaluated for the DNBR.
- 7. Cases without pressure mntrol did not have a DNBR rm with THINC. An evaluation was made.
- 8. No DNBR evaluations were made using THINC.
- 9. Limiting statepoints were verified by T4H in NFD.
G d e Mw i e t 6
~~
s .
', W-PTP-64 6
7); c y b-(c
) p,.. . .
r i Westinghouse Water Reactor ' Electric Corporation 'ww c:mre Divisions enanmm:at 5: 3S . s P:ns:wg iPt1r.sf tvem 152 3
-r Mr. C. O. Woody, Manager Nuclear Energy Juile 22. 1982 Florida Power & Light Ccmpany P.O. Box 529100 Ref: 93,000-85525
Miami, FL 33152 DWA 93507 MI-28460/4
Dear Mr. Woody:
W-PTP-61 FLORIDA POWER & LIGHT COMPANY 1 TURKEY POINT UNITS 3 A::D 4 e Turbine Runback Svstem - The attached recort was prepared in accordanca with the recuirements of the referenced Florida Power & Lignt Purchase Orter. The report describes the impact of a proposed modification to the Turkey Point Unit 3 and 4 Turbine Runback System wnich significantly reduces the pr:bacility of spurious.runbacks The report concludes that plant safety is not cem;rcaised provided that the plant is in the manual red control mode of o; era:icn. . With this submittal, Westinghouse has ccm:Isted Phase 1 of a 3-Phase prog Phases 2 and 3, which will be procesed to Flcrica P wer & Light in the near . future, of cperation will c:ver increasing in autcmatic rod the runback ;:.ver level to SC'I and evaluation centrol. We r ::
- __
- .a: :nis sutaittal meets year nee:s. Sh: '.: you have any g.esti:ns or c:mments en the attached report, please c:ntact the Westinghouse Projec:
Office. Very truly yours, W"7INhC"S" ELECTRIC CORPORATION s , , GJM:rst bh l = Attachment Gw Y Murray, P sect Engineer FloridaPcwer1/ightProject J cc: C. O. Woody, FPit , 2L , 2A H. H. Paduano, FP&L, IL, IA R. J. Acosta, FP&L, IL, IA . . . . H. E. Yaeger, FP&L, IL "-'* R E M E R G Y J. K. H1ys, FP&L, il . y
,;gy.o5 1987, R . L . Whi tney , ~W , IL , l A.
E . V . Eu r.l edge , '.1, IL
., , _ _ . _ , , . . _ ,, _.___.,_.____,__.,_.mm_ ., _
, _ . . _ _ _ __. , _ _ , . - ,.._m
. r . n . . . . . .. . e . . . . . . . . . . - v, 4 4 .
. s t
MODIFICATIO.1 0F THE TURSINE RU.13ACK SYSTEM The automatic turbine runback feature of. Turkey Point 'Jnits 3 and 4 is designed to provide protective action in the event of a dropped RCCA or dropped bank. 1
; Detection of a dropped RCCA or bank occurs by either a roddn-bottom signal ~
device or by a change in neutron flux as seen by the excore detectors. The rod-1 on-bottom signal provides separate indication for each RCCA in the' core and one. signal is sufficient to initiate the turbine runback. .go,,achangeinfluxas o l seen by one of the four excore detectors will cause the turbine load to be redue j The turbine load is reduced to a pre-set ,value of 70".. At the same time, ' uto=a j withdrawal of the control rods is prevented by a red withdrawal bicck. This
, scenario is discussed and analyzed in Section 14.1.4 of the Turkey Point FSAR.
i . { The design of the automatic turbine runback is prone to spurious runbacks (i.e., j - runbacks not caused by a RCCA drop) because there is no coincidence logic used is i-the initiation of the runback. Thus, a single failure of an electrica.1 ccmponen-(turnout of a rod position indicator signal, failure of one excore detector, etc !
~
instead of acting to prevent protective action will, in fact, cause a turbine I runback when it is not needed. This causes unnecessary plant transients and l l results in a significant loss in operability' and availability. Operating hist:r. f at the Turkey Pcin: units sh:ws tnat seven s:uricus runbacks have occurred in {- :ne last two years. Mo run:acks occurre: cue to a droppec RCCA or bank. ( 1 {, The cajority of the spurious runbacks have resulted from failures in the flux ra '
- input to the runback logic. If this input cculd be deleted, operability would b i j greatly improved: Since the turbine runback system is designed to provide pro-l tection for a dropped RCCA/ bank, this accident must be re-evaluated in light of any changes to the system.
L In any accident analysis, a limiting single failure for that transient is assume . I
- In the event of a dropped bank (assuming the flux rate input has been deleted),
- between four and eight rod-on-bottom signals will be generated, one for each rod
- in the bank. A failure of an4 one signal has no impact, since ther,e are still
- several other signals available, and only one is needed to initiate the turbine 1
i runback. Th.:refore, the dropped bank analysis is not affected by this change j ..
... ..i- .- t
, , .:: :..2
.. .. r ine run:ack Icgic, and tne F5AR ana'.ys:s presentsc in Se::i:n l'.l.t remains applicable, However, for a single dropped RCCA, the failure of one red-on-botte. ' signal meant that no runback will occur, since the only signal generated, failed. (If the flux rate inout is used, this will initiate runback if a rod-on-bot:cm ' nal fails for a dropped RCCA.)
Therefore, this ac:icent must bi, reanalyzed assu.+irq no turbine runback occurs. - The transient & for a dropped RCCA is calculatec using the.same cetneds as describe inSection5f.l.4. f The LOFTRAN code is used to model the plant response. The LOFTRAN code is a detailed digital computer pr: gram which simulates neutron kinetics, the pressurizer and its re. lief and safety valves, pressurizer spray and heaters, rod control system, and steam generators and their relief and safety valves. Pertinent plant variables, including temperature, pressure, and power level, are computed. Most negative coderator anc do:pler temperature coeffician ! are used to maximize the core heat flux. Figure 1 illustrates the transient for a typical droppe: r:d wcrth cf 150 pcm in manual red control. The core heat flux initially dr ps and : hen re urns to the initial power. level due tc cesctivity feedback. Tec;erature and pressure drop to a lower value. New equilibrium conditions are reached with the core at full p:wer and reduced temperature and pressure. These c:nditiens are les.s If. .iting than
. thesE for whi:h the
. . _ _stat.ic
. full leneth misalic er: c f in R',;i. ' a_2_. a i ._: s t ' :. e
.ei:ac Sare:y tvalua:icn fcr esen cycie. In :nis ana, lysis, the reac:or
~ ~ - ~ is at ru,il power with nominal temperature and pressure (including uncertainties). At the same power level, the DMER benefit due to the reducticn in _temoeratura which occur _
in the transient case rcre than cetpensates for ;P.e CMER penalty caused by the e transient drop in pressure. Thus, the single :ropped RCCA analysis assuming manual rod control and no turbine runback is b:unded by static RCCA misalignmen A dropped RCCA in aute:atic rod control is not bcunded by this analysis since a power overshcot abcVe 102% could occur. I n s utma ry , '.. :U.te, house finds that it is acce: table _to delete the fl.ux rate portien of the _ turbina runback system provided the cients rer.ain in manual rod control. The detection and analysis of droppe: banks are not ~affected by this change. A .fe',pped RCCA which dols not result in a runback is bounded by static RCCA misa! , :nt.
.e . . .. . . . ..
-- -- ' - ' - '^ -
.. ., ..- fiO TU?.3!!!E .9U?!SACX s
9 i
~ 1.0500 i, , .
._= ..
=
f C .
= -
e1.0000 - i
~-
=
.t.o.,
u ! 1 C .9500 '
~ .
I I i i 1 I 0 N .
+
1.5 - 1
- a - 4 i e 1 1
t.
; 3.0 - W < .\ .
i ._. _ I i ' 4.5 - i 7 l i I 1 t , +10 ; i 0 . - - i m t 20 -- - ' 4 N +
- 30 -
,a -
l I
-40 O M 20 30 40 50 60 70
, Time (Seconds) - m _ _ _ . _ .-. _..._ .._,._ ..-- . - ~ _ . , . , _ _ . - . _ - . _ _ _ - . . . _ _ . _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ , _ , , . - , _ - , _ . . . _ , _ _ , , _ _ _ . _
j'
~
Ta ,_ < , , % , us, "f@ e d.w.c :3 g, 77
. l
, -r o :
O TMQ b re,Ls. S 4WabFay o S/ 5 St.ac> a % - Sck e o . O O o -
, O O. _
O-O Tf[/G, - T t:l w Segk 44.s - Lib g=A o. c.u (.( tm% 0 Eb hy n.h ce>~\ d Q) eld-s'b % q.) sM .. nne. J .-s. h' o -c . L.sR .< % L -/R f a l _L. 3 o O_
-r Q
L & h r e. . - L .: - :u.k
*~,__
3,,, ,,,Q (4 og%, g& co.-o . ~ L. ~ o '::t, m :0 pve,o.be e c. e, A #A _.
. a s t-b - ._p I : 4 m R s ec- d L.py, mA.M .
e' ~ o . s / q.. v. .. A a p/m--\, 4, 4,% JA. ,..M W. . , O O O O O O O - O
~
s C wwg hJCwS5e) thM t1, ., t$ t o o wa- M toe o.6% b Mk 4e
~ k e J as d [o.sM 7- -
C) s w e. j k & t9 % p d-oo .a- c;$sp
% 4 e,*w sk g __
cd w ass
, g /jg _ _ .
. '.~; m
~f 1 h_M bc bAhu) b *k % gws % gl 4A R e;G Va c.
~W, %gYs.- 14 w aPPs'<d A &
es2,_. A A p p/14. . - (50- 37T- 5 t 3 2:- uc. mi
-7dn Se l e I-s C e 'td wA - JIcd( m 35 e_d m(e
+%
~h W , 6sh 4A _
K e sc. . 4 me A 6.e E . e m
~
l l I I' I: ll 9 a!
f 4 % A vite ma%L us* ^ 9e"
=f
- s44: h c_ y, N 7[2 Sc344 c84 l cM 7/<o cJ h 4.% ~y owm 4A.dh Fox
- H b L:- m e. L 's vn., :A c/n/res f.ef6-z
- L<. m M ~lA .~ G/7 @i Y: sop-Q e sc. M c 1 + A h e e v-
-k Q msQ h ~
c.mg eV Lh pme d & (-os g pn~< k h is u . n w l - paha & prL n -e menil en
~Q Gre- Al-ag c ( i ,/tr s /* k - RvJ JL&t-A nat -
hv-l_J4 pea r le4% bNRC. [ -/c>
- wh L:~ % ~Cb reoca e-h Lo. A be lt
=.
LO j
, ma_,-,oea 4 - - . - - -----1.-- - -a-A -- ' _A _ as, tA.x4, _A-L_as a ,
O g t/n_ e.y w ot*-s J A e 3L 7 L
- .fN uJ6 4_ Ar , . /w A a.u u S
%.A. c i ,_4 pm / u3 f (c..-~= & P% A)
- c-Q4-sr a,..ab A4 ~
a e.m : \, 1
)
I i n 4 em I i 1 M =w I I O i 1 1
4 g$ I* s A list of rod stops is given in Table 7.2-3. Some of these have been ; previously noted under permissive circuits, but are listed again for completeness.
.- t Rod Drop Detection j j i
l Two independent systems are provided to sense a dropped rod, (1) a rod ; bottom position indication system and (2) a system which senses sudden -l reduction in out-of-core neutron flux. These systems are not reactor , protection systems. Both systems initiate action in the form of a I
. turbine load cutback and blocking of automatic rod withdrawal. This action compensates for possible adverse core power distributions and '
i permits an orderly retrieval-of the dropped RCCA. A dropped RCCA would be detected by the rod bottom signal derived for
~
j each rod from its individual position indication system. With the l ,position indication system, initiation of action is not dependent on - f location, reactivity worth or power distribution changes. l Backup is provided by use of the out-of-core power range nuclear de-tectors and is particularly effective for larger nuclear flux reductions occurring in the region of the core adjacent to the detectors. - i-
~~
)
i The rod drop detection circuit from nuclear flux consists basically of i a comparison of each of the four ion chamber signals with the same signal ! taken through a first order lag network. Since a dropped RCC assembly l !
'will rapidly depress the local neutron flux, the decrease in flux will j i be detected by one or more of these circuits. ,
l Such a sudden decrease in ion chamber current will be seen as a derivative i signal. A negative signal output greater than a preset value (approximately ] i 5 per cent) from any one of the four power range channels will initiate
- ~
turbine runback and block automatic rod withdrawal. l 9 I l 7.2-25 .(
1,- . j
. Figure 7.4-2 indicates schematically the Nuclear Instrumentation System, including the dropped RCCA alarm.
- Automatic' turbine load cutback is initiated by a signal from a dropped rod control cluster assembly as indicated by either a rapid decrease in nuclear flux or by the rod bottom on-off controllers. Load cutbajk is also initiated by an approach to an overpower or overtemperature condition.
This will prevent high power operation which might lead to minimum DNB _ ratio less than 1.30. The rod stop is redundant. Rod stop contacts are located in the rod control logic cabinet and in the rod speed control analog rack. The turbine runback acts by either of the following.
~
- 1. Reduction of the load reference setpoint of the turbine governor speed changer by a preset amount. This is accomplished by reducing
- the set point at a constant rate for a preset time. ,
I
- 2. Reduction of the turbine load limit to a preset value. The load limit (a relief valve which limits control oil pressure) '
is reduced until turbine thermal load as sensed by either of two turbine first stage pressure channels is below a
. __1 preset value.
The amount of the runback is to be determined by physics tests of dropped rod worths and hot channel factors during startup tests. The safety requirement of the runback is to preclude return to a power level that might result in a core damage because of adverse hot channel factors. . It is expected that the startup tests will show that , dropped rod hot channel factors will not cause a DNBR less than 1.30 even at full power, and that the runback will be set for operational requirements. That is, the automatic load reduction would be large enough such that with reasonable operator action, an orderly manual shutdown can be accomplished rather than a reactor trip on low pressurizer pressure. l 1 7.2-26
8. 14.1.4 ROD' CLUSTER CONTROL ASSEMBLY (RCCA) DROF Dropping of a full length RCCA could occur only when the drive mechanism is,de-energised. This would result in a power reduction and an increase in the hot chaa==1 factor. If no protective action occurred, the Reactor f ! Control System would restore the power to the level which existed before ~ the incident. This would Isad to a reduced safety margin or pfossibly , Init, depending upon the magnitude of the hot channel factor. i
-l
! If an RCCA drops into the core during power operation. (2244 left), it i j would be detected by either a rod bottom signal device or by the use of the j out of core ion chambers. The rod bottom signal device provides an individual position indication signal for each RCCA. Initiation of this .- j signal is independent of lattice location, reactivity worth,or power . i distribution changes. inherent with the dropped RCCA. The other independent indication of an RCCA drop is obtained by using the out of core power range channel signals. This red drop detection circuit is actuated upon sensing a rapid decrease in local flux such as could occur from depression of flux ;
"in one region by a dropped RCCA. This detection circuit is designed such' 1
i that normal load variations do not cause it to be actuated. l A rod drop signal from any rod position indication channel, or from one 1 i or more of the four power range channels, initiates protective action by , J {
- reducing turbine load by a preset adjustable snount and blocking of further [
~
- - - automatic rod withdrawal. Either action individually prevents core damage.
The turbine runback is redundantly obtained by acting upon the turbine load t j limit and on the turbine governor control systen. The rod stop is also ! l redundantly actuated. ; 1 1 i - [ L i i I i I l 14.1.4-1 Rev. 2-7/34 i 1 l l
{* . 2 . Method of Analysis l . 1 ' The transient following a dropped RCCA accident is. det. ermined, by a detailed digital , simulation of the unit. The dropped rod is assumed to cause a step decrease in reactivity and the core power generation is determined using. a point, neutron kinetics model. The overall response is calculated by simulating tha turbine -load runback and blocking of automatic rod withdrawal. The analysis, is. performed. for the case in thich the load. cutback nearly matches the power decrease, from the negative reactivity ~ for a dropped rod (-2.0 x 10-3 (k), and also for the case in which the load cutbach j is greater than that required to match the worth of the dropped rod (-1.0 x 10-3 CO, 1 In both cases the load is assumed to be cut back from 100 to 75 percent of full 10cc ct a conservatively slow rate of one percent per second. The actual amount of load cutback to be used will be determined during initial. startup esp.eriaants. and will be ! set to match the power reduction caused by the highest worth dropped rod. I
- The most negative values of moderator and Doppler, t.esperature coefficients of
- reactivity ,are used in this analysis resulting in, the highest heat, f,Lpx. durips the transient. These are a moderator temperature coef f,i.citatr of -3J z,1.0-4 6k/*F gnd. a, j Doppler coefficient of -1.0 x 10-5 6gjoF. A. control group, worth, oft 6 x 10-5 6k/in. is s
cesumed as equilibrium conditions are restored, Results I 1 n __ Figures 14.1.4-1 and 14.1.4-2 illustrate the transient, r, esp,onse following a, dropp,ed , rod of 2.0 x 10-3 Ok . *The coolant average temp,erature doct, eases, rapidly initially, then decreases slowly to new equilibrium condition. The peak heat, flus following the . l
' )
initial response to the dropped rod is 95 percent of; nominal. At the same time.the 1 l core average temperature drops by 20F and the pressure by 28 pai. l nResults were re-evaluated to assure that criteria are met for current cycle kinetics parameters and rod worths. Additional information can be found. in Appendices 14A, 145, 14D, 14C and 14H. Analysis has also been performed to: allow disconnecting flux ' rate trip on rod drop when reactor control is in manual; see Appendix.14C.
?
i .1 i i Rev. 2-7/84 14.1.4-2 j
'~, 4 Figures 14.1.4-3, and 14.1.4-4 illustrato tha trcasicet racponsa following a a dropped rod of 1. x 10" 6k. Again the coolant average temperature decreases initially, and then increases because of the negative reactivity feedba'ck and the load cutback. The equilibrium temperature will again be I achieved in about six minutes. For this case the peak heat flux following
- the initial response to the dropped rod is 96.5 per cent of nominal. At i
the same time the core average temperature drops by l'F and piessure by 40 psi. , I An analysis has been made of the amount of flux tilt that can be tolerated l without core damage for the maximum full power operating conditions (2244 MWt power; core water inlet temperature of 550.2*F primary pressure of 2220 psia); a more conservative condition than those mentioned above. The effect of the flux tilt was represented by an increase in the radial ]l heat flux hot channel factor. It was found that this factor could be increased j by 12 per cent before reaching a DNB ratio of 1.30. During initial startup l , experiments, it will be verified that the flux tilt caused by the most f j- reactive dropped rod, coupled with the thermal flux, coolant temperaturei and i l reactor coolant system pressure responses, will not result in a condition of DNB. J Conclusions , l
- Protection for a dropped RCCA is provided by automatic turbine power cutback
) - and blocking of automatic rod withdrawal. The magnitude of the power cutback j 3 is 'to be determined during the initial startup tests. As the analyses e presented show, the protection system, in conjunction with the load cutback, protects the core from DNB for a power tilt of 12 per cent at maximum full power conditions, greater than expected for the unit. At the reduced power , condition following the rod drop, this allowable tilt will be even greater. i i The power tilt will be experimentally determined and the protection system I set to maintain a DNBR greater than 1.30. I f, i I i ,.
REFERENCES:
Seetion 14.1.4 4
~
- 1. Westinghouse Report, " Power Distribution Control in Westinghouse j ,
Pressurized Water Reactors, "WCAP-7208 (1968), PROPRIETARY. The NON-PROPRIETARY version of this report is WCAP-7811. 14.1.4 3 i
- _ _ _ . _ _ _ _ - , . _ . _ , _ _ _ _ . . _ _ _ _ _ _ . _ _ . . _ _ _ . . _ _ ~ . . _ _ _ . . . _ . _ _.,_,._.,___ _ __ _ _ ,_.,_.,_ _ _, _ ___ _ _ _ ._.
T . AFFENDIX 14C '[ MODIFICATION OF THE TURBINE RLNBACK SYSTEM"
. ~
The automatic curbine runback feature of Turkey Point Units 3 &nd 4 is designed to provide protective action in the event of a dropped RCCA or dropped bank. Detaction of a dropped RCCA or bank occurs by either a rod-on-bottcc signa)
- device or by a change in neutron flux as seen by.the' excere detectqrs. ' The rod-on-bottom signal provides separate indication for each RCCA in the core and one
- signal is sufficient to initiate the turbine runback. Also, a change in flux as seen-by one of the four excore detectors will cause the turbine load to be reduced.
~ The turbine load is reduced to a pre-set value of 70f.. At the same time, automatic .
withdrawal of the control rods is prevented by a rod withdrawal block. This i
- s'canario is discussed and analyzed in Section 14.1.4 of the Turkey Point FSAR.
.,' The design of the auto-atic turbine runback is prone to spurious runbacks (i.e.," -
runbacks not caused by a RCCA drop) because there is no coincidence logic used in . the initiation of the runback. Thus, a sir.gle failur's of an electrical component (burnout of a rod position indicator signal, failure of one encore detector, etc.) Instead of actiRg to prevent protective action will, in fact, cause a turbine rvnback when it is not needed. This causes unnecessary plant transients' ano results in a significaat loss in operability and availability. Operating history 2 at the Turkey Point units shows that seven spurious runtacks have occurred in
-- the last two years. No runbacks occurred due to a dropped RCCA or bank.
The majority of the spurious runbacks have resulted from failures in the flux rate input to the runback logic. If this input could be deleted, operability would be greatly improved. Since the turbine runback system is designed to provide pro-tection for a dropped RCCA/ bank, this accident must be re-evaluated in light of any changes to the system. In kny accident analysis, a limiting single failure for that transient is assumed. In the event of a dropped bank (assuming the flux rate input hhs been deleted), betwee'n four and eight rod-on bottom signals will be generated, one for each rod
..n.the
' bank. A failure of any one.s.ignathas no impact, since' there are s.till several ether signals available, and only.one is needed to initiate the turbine runback. Therefore, the dropped bank analysis is not affected by this ' change
*" Turbine Runback System Safety Evaluation,"
FPL letter to the NRC, L-82-343, dated August 10, 1982. < 14c-1 i Rev.1-11/83
, , - ,-- - ---. ,- - - - - . ------- --.-.- - , - - - - - - - - -n-
- l , s i
1 t
. i 1
l
\ ..
3 %
'tc.the turbi~ne runback logic, and the F5AR analysis presented in Section 14.1.4 ,
remains applicable. However, for a si,ngle dropped RCCA, the failure of one rod-on-bottom signal etans that no runback will occur, since the only signal generated, failed. (if the it flux rate input is used, this will initiate runback if a rod-on-bottoat signal fails for a dropped RCCA.) Therefore, this accident must be regnalyzed assuming no t14roine runback occurs. The ' transient for a dropped RCCA is calculated u' sing the same methods as described
, in Section 15.1.4. The LOFTRAN code is used to model the plant response. The
. LOFTRAN code is a detailed digital computer program which simulates neutron kinetics, the pressurizer and its relief and safety valves, pressurizer spray and -
heaters, rod control system, and steam generators and their r:11ef and safety valves. Pertinent plant variabits, includin'g terrperature, pressure, and power level, are computed. Most negative moderator and doppler temperature coefficients are used to maximize the core heat flux. . Figure 1 illustrates the transient for a typical dropped rod worth of 150 pcm in manual rod control. The core . heat flux initially drops and then returns to the C._ . initial . power le, vel due to reactivity feedback. Temperature and pressure drop to a lower value. New equilibrium conditions are reached with the core at full power and reduced temperature and pressure. These canditions are less limiting than those for which the static full length misal'@ ' at of an RCCA is analyzed in the
.'.'keload Safety Evaluation for each cyc1r 1 t '.* analysis, the reactor is at full
~
power with nominal temperature and prtss ,g ( '. t. Jding uncertainties). At th6 same powir level, the DNBR benefit due to the rtdaction in temperature which, occurs
. J' 1.n the transient case more than co.npensates for the DNBR penalty caused by the transieIt drop in pressure. Thus, the single dropped RCCA analysis assuming manual rod control and no turbine runback is bounded by static RCCA misalignment.
A dropped RCCA in automatic rod control is not bounded by this analysis since a power overshoot above 102% could occur. In sumary, Westinghouse finds that it is acceptable to delete 'the flux ' rate - l portion of the turbine runback system provided the plants remain in manual' rod _ sg.01C9) . The detection and analysj,s of droept1. banit sar.e nqt.Af,.fectjrf by.this ' . change. A dropped RCCA which does not result in a runback is bounded by static RCCA misalignment. 14c.2 ' 1 Rev.1-11/83 -
- - ,. , - . . --- . - . - - - - - - - - , - - _ - - - - - - - - - - - - - , , , - - - - - - - - - - --- -,-- ,- ,,--w
l l.
. l I
, OROPPED R0D WJiuAL C0fiTROL l
.. NO TUR8!NE RUNBACK
. ~1.0500 . . .- _ . . .
7e -
%~ .
I w w w . . . . { y .1.0000 - ~ E g ;- - . o 6 g .9500 . . s .
* \ Q e , , e
- ?
y .
- E 1.5 - -
w
& N
-1
. 2 3.0 - -
=
4 4 . 4.5 .
+10 ; .
B 0- -
=
. 5
.O.- -10 -
i .
- u. m .
~ a- - "
g-5 .y :. b .
~. -40 0 10 20 30 to 50 60 70 Time (Se:snds)
FIGURE 1 14c-3 i
. .. Rev.1-11/83 ;
l i l
- - ~ - + + n-- avaw v, p ,-,- y, --w , - --- -,,<m---- - , , - , ----,-n- - - , - , , - .,-,--.,----ge-- m---,-u. - -
, aLonioA rowen a ucnr coueAuv August 10, 1982 L-82-343
= i Of fice of Nuclear Reactor Regulation e Attention: Mr. Darrell G. Eisenhut, Director Division of Licensing U.S. Nuclear Regulatory Consission QUAUTY CONTROL DEPT.
Washington, D.C. 20555 FIM COPY
Dear Mr. Eisenhut:
Re: Turkey Point Units 3 & 4 Docket No. 50-250 and 50-251 Turbine Runback System The Turoine Runback System (TRS) at Turkey Point Units 3 & 4 receives inputs fran either the ex-core neutron flux detectors or a rod-on-bottos indication
- s ig nal . It has been our experience that tne TRS is prone to spurious runbacks due to single failure of single electrical components. Tnese spurious runbacks have subjected tna plant to unnecessary transients and reduced .
availability. Therefore, we are proposing for your review and approval under the guidance of 10 CFR 50.59 a modification to the TRS which deletes the flux rate input. Inis modification will significantly reduce tne probability of spurious runbacks without compromising plant safety. The attached safety evaluation demonstrates that the consequences of a single _ __ dropped Rod Cluster Control Assembly (RCCA) assuming manual rod control and no turbine runback is bounded by tne previously analyzed static RCCA misalignment. The static RCCA nisalignment has been analyzed in WCAP-9272,
" Reload Safety Evaluation Methodology" which is the proprietary topical report submitted to you from Westinghouse on April 15, 1978. Tne automatic rod control system has been disconnected at Turkey Point Units 3 & 4, thereoy assuring operation in the manual mode.
Very truly yours,
,i
'l Robert E. Uhrig 8 Vice President '
Advanced Systems and Technology . REU/JEM/mbd , Attachments cc: Mr. James P. O'Reilly, Region 11 Mr. Harold F. R'eis. Esquire %i> 7
!i . MODIFICATION OF THE TURBIllE RU:iBACK SYSTEM l
~
The automatic turbine runback feature of Turkey Point Units 3 and 4 is desig to provide protective action in the event of a dropped RCCA or dropped bank. Detection of a dropped RCCA or bank occurs by either a rod-on-bottom signal device or by a change in neutron flux as seen by the ex. core detecters. The on-bottom signal provides separate indication for each .RCCA in the core and signal is sufficient to initiate the turbine runback. Also, a change in flu seen by one of the four excore detectors will cause the turbine load to be r The turbine load is reduced to a pre-set value of 70t. At the same time, au withdrawal of the control rods is prevented by a rod withdrawal block. This s'cenario is discussed and analyzed in Section 14.1.4 of the Turkey Point FSA The design of the automatic turbine runback is prone to spurious runbacks (i l runbacks not caused by a RCCA drop) because there is no coincidence logic us 1 the initiation of the runback. Thus, a single failur'e of an electrical ccep (burnout of a rod position indicator signal, failure of one excore detector, l instead of actirig to prevent protective action will, in fact, cause a turbin runback when it is not needed. This causes unnecessary plant transients'and results in a significant loss in operability and availability. Operating hi
~
at the Turkey Point units shows that seven spurious runbacks have occurred i the last two years. He runbacks occurred due to a dropped RCCA or bank.
~
The majority of the spurious runbacks have resulted from failures in the flu
' input to the runback logic. If this input could be deleted, operability wou greatly improved. Since the turbine runback system is designed to provide p tection for a dropped RCCA/ bank, this accident must be re-evaluated in light '
any changes to the system. In any accident analysis, a limiting single failure for that transient is as In the event of a dropped bank (assuming the flux rate input has been delete between four and eight rod-on-bettom signals will be generated, one for each
.. in..the bank. A failure of any one_s,.ignal has no impact, since there are s.ti several other signals available, and only one is needed to initiate the turb ;
runback. Trierefore, the dropped bank analysis is not affected by this chang
However, for a si,ngle dropped RCCA, the failure of one rod-on-bottcm signal that no runback will occur, since the only signal generated, failed. (If tt flux rate input is used, this will initiate runback if a rod-on-bottom signi fails for a dropped RCCA.) Therefore, this accident must be reanalyzed asst no turbine runback occurs. e . The transient for a dropped RCCA is calculated using the same methods as de: in Section 15.1.4. The LOFTRAN code is used to model the plant response. i LOFTRAN code is a detailed digital computer program which simulates neutron, kinetics, the pressurizer and its relief and safety valves, pressurizer spr. heaters, rod control system, and steam generaters and their relief and safe j valves. Pertinent plant variables, including temperature, pressure, and po-level, are computed. Most negative moderator and doppler temperature coeff ' are used to maximize the core heat flux. 1 Figure 1 illustrates the transient for a typical dropped rod worth of 150 p l manual rod control. The core heat flux initially drops and then returns to initial power le, vel due to reactivity feedback. Temperature ar.d pressure d a lower value. New equilibrium conditions are reached with the core at ful and reduced temperature and pressure. These ccnditions are less limiting t those for which the static full length misaligr. ment of an RCCA is analyzed ,
' Reload Safety Evaluation for each cycle. In this analysis, the reactor is power with nominal temperature and pressure (ir.cluding uncertainties). At
~
same power level, the DNBR benefit due to the reduction in temperature whic in the transient case more than compensates for the DNBR penalty caused by transient drop in pressure. Thus, the single dropped RCCA ar.alysis assumir l manual rod control and no turbine runback is bounded by static RCCA misalis A dropped RCCA in automatic rod control is not bounded by this analysis sir power overshoot above 102% could occur. I l In sumary, Westinghouse finds that it is acceptable to delete the flux rat portion of the turbine runback system provided the plants remain in manual contrp.1. The detection and analys.isaf drp.pped_ banks.are not._affecte,,d by
- change. A dropped RCCA which does not result in a runback is bounded by s' RCCA misalignment. .
Pace 2 of 3
.I -
s
-- 1.0500 i
m
=
x .-
= s
- O u = = .* ..
., s
=
u o1.0000 -
= =
O% e,a
.u -
o u u o
- u I. .
u.
.9500 ' . : .
l 0 -
- l a
= A o
u 15- I
.. -s - I , ,a
- nu. ,,; 1 r e-o #
;E' i 3.0 - -
u 4.5
. i
- l l
+10 ,
a 6: 0- l e i m *) U -
- c. - -10 .,,
I
,u .e '
so m -
- ~ =- )
i--
- u . - . ---
,g... __ _ . . . .
l s" -
. -40 -
30 40 50 60 70 l 0 10 20 ' l . Time (Second,) i "1 6
~
H_, A ,2 etcaics. coe.ta s tis-T cc ew.v Septe-ter 7, Ic52 L-E2-398 a Of f1:e c' ';uclear F.eactor Re;ulation A :ention: M.r. Darrell G. Eisennut, Director quAUTY CONTR01. gpy- _ ivision of u censing
'J. 5. '.s:Ica r '-e;al a:Ory Cox31ssion FILE COPY Wase. .;;on, D.C. 20E55 e r *' . E 5e .n :: l 1
te: . . ey : ;. i n .r.t:5 3 6 - l 7 ;; e.e: *).
. bb-2i.') aftC 50-251
.-: yJ.anoa: Syste.
Our ie;;cr i.-h2-20, c4:e: Aa::a s: 19, Pec trar.; :.;ec a ; gctes c.3:s fica ion to :P.e *st t r e Run:a:a. Sy s t e r.. ;.e las: Le-.en:e in tne 'inal ;;3regra;,n as l
.' ., c.,.:ic eij correct anc sn: 1: se replaces ..i:n :ne f::'.o :ng: _
l
*.e po.or r.is"latcfi i n;. o tne asto,atic ros conte:: circ.:tre r.es ::cen ci s:onne::e et iv ey I'oin U n ;; 5 3 !,
- c.: o; era: tor. 15
- :.2 anual .oce.
;ery .iy yours, M. ._ _ W6 1 :>oert E. Unrig l li:e President l
, A:sancei Systens and Te:hnology RE'J/JEM/.ed l
l A::achinents 1 c:: l'.r. James P. O'Reilly, Region 11 i Mr. Harold F. Reis, Esquire c ?!Wi' t t , a-
r- ..--- , I , , '5 /pse
- e/ r.:,'-~ 'N ,
. v -5 UNITED STATis ^
4 i y, y f* *! /i**Y . g NUCL2AR RiEGULATORY COM.'.11SSION W ASMlNCTON. D. C. 20!$ 5 U .
* ~ * *.. , y. . ~". i
$%[. ant-""lW.
- k...* - ./ S!.11 ~ HI January 5,1953 1.t1::::: .:: : 1is::::::
Occks: 'los. 5C-250 .- anc 50-251 - l Dr. Ecber- E. Uhrig, Vice Dresident Acvanced Systems and Technology Florida Power and Light Ccmmany
?:s: Office 5cx 529100
. Miami, Florida 33152
Dear Dr. Uhrig:
-g..._
6,: e ,....: ,..,. Oe..:
.... e..,JX r....
0:, ; v. - .m - L. . P L,_. . ., ..... .:. .....
. . . .w . i-
. ., c. .I:.
o the Se:: ember above7,1922, recues:.The relating staff has c:.1:leted i:s revie,. of your sur.i 1 [ated on cur eview, we have c:ncluded tha: "
.ar:ine s entack System is ac:e::atie provide: he ::r:;csed revision :o :he ,
-he reac :r re ains in manuai c:n:r:1 cece, ne e
The review. enciesed safe y Evaluatien ;r:vides the details and resul s :e cur 1 . Sincerely,' t t I .. .
~
~.t'. 4
,. .i
. ll(jljyl lt;_,;j%.,
i ' ,S tev en n.* *Va rga ;' chi e f Opera-ing Reac crs 5ranen it Division of Licensinc. Incie s~ure: - i As sta:ed - cc w/ enclosure: St:e next ; age em> 9 e g *. w, I 9 *1 - 4 _._J
]
% re . Rece.-: I. Uhri; y Florida 7cwe.- anc .' ign: C:::any , L cc: Marold F. Reis. Escuire Lowenstein, New=an, Reis and Axelrad James ?. O'Reilly 1025 C:nne -icu- Avenue, N.W. Regional Assinistra ce - Recien :'
~
Sui e 1214 U. S. Nuclear Regula:s. v Cc..miss'
~
Washing:en, D. C. 20035 101 Marie::2 I:ree - Suite 3100 A:lan a, Georgi'a 30203 Norman A. Coll, Escuire - Steel, Hec or anc Davis 1400 Scu:neas: Firs National Bank Building Mi ami, Fic.-ida 23131 Mr. Hen.y Yanger, Flan: Manager Turkey ?:in: ?1an: Ficric P. O. s,aexFower 013100 anc Lie.n:.Cem:an.v - Mia':1, Ficrida 23101 _ Mr. Jack Shreve Office of :ne Public Counsel '
. Rcom 4, McIland Building ._
.al,iahassee, .crita i acau,.
Acminis .ator - De:ar: men cf Envirencen al Regula:icn 7:=er ?!an: Si-ing Secticn Sta e cf Florica 2500 !!!'.- 5:: e Rcad T a '. '. a . ! ! ! a e , F '. : .- t :1 III:; Residen-}nspec:cr Turkey 7:in: Nuclear Genera-ing S:t-ten . U. 5. Nuclear Requia ory C::=issten Fos: Office Sex 1207 Momestead, Fler.fca 22020 e
. W S
B s e .
~. - _ - . - . . - = . . .- . . -
1 8 . . . O . ?.1
, , y ..
s .- - _
)m.--tg. t e. .=..m.4a g g e n -
- m.o w ....
- a. -t. : - . ....c. . sn . n..:. . . .o.
. r t ys m z a r,
. t. y e. . ...
rw V . JNTRCOUC7'C4 r By letter dated August 10. 1952 i Florida ? wer and Light C and clarificatica dated Se:: ember 7 . 1932 the for Turkey 7:in: Units 2 and;&ny 4 pr:;osed to m:dify the Turbi'nh Runba:k Syst em The m:dification would delete the flux rate input to the system in order to reduce the number of spurious runbacks. The Reac::r Physics Se: icn of the C:re Performance 3 ranch has Our evaluati:n follows. csalrevi I j In the red bank or single r:d dr:p ' event a turbine ranhack to 70 per:e { full power o::urs wnen either a red-on-bettom signal or a negative flu signal fr:e :ne :f :ne ex::re dete:::r channels is received. ' j A failure in the cate:::r :nannel then results in an unnecessary turbine runba:k. s If :ne flux .a:e input is removed fec: l 1
;r: e::f n agafest r:d dr:: the turbine rur. hack cir:ut the :nly \
- i. assumes a single failure :: events w'll te the r:d be::om signai. If cne.
; cf tre r:d 5: c::ur in :ne ;r:te:;f=n system (i.e.. the failure om signal) then a runba:k will no:
de::s. oc:ur wnen a single r:d i Fre:e::icn f:r a red bank dr:p will s:tli te pr:vided sin:e in.s:ri case ine*e are multi-le r:d 5::::t si:nals. t
. f..,..-ea...u j '7:___ The sin;'e r : cc:: even:
has been reanaly:ed for the Turkey .8: int rea: :r ' assuming :nat no turbine runha:k oc:urs. , ! . au:::a-1 The power .-ismatach input to :he t r.:e c:n:r:1 cir:uitry has been dis::nne::ed and the reac :rs are i c: era:=d in the manual me:e. Thus n: 3 re urns s ;cwer af ter the dr:;. eversn::t c::urs when the reac :r The return to ;cwer. de:urs as a result cf the rede: f:n in the average : dera::r tesserature Ib.e tem: era ture redue:f on , in turn, causes a redue:f n in reactor pressure. s The cora stabilizes at essentially full power with the de ; ped rod fully inserted. { This case has.
- been r
- d analy:ed for n:r=al tem;erature and pressure ' in misali;nten: the FsAk as th e limiting 11 its. case and shown to result in no violation of fuel ther=al The effect of the redue:1en in tec;eratdre and pressure 'in the dr: ;ed rod case has been analy:td 'liy standard 'destinghcuse n ysis methods.
a al i : Y1 J l hf f I 5 i J
- 4
. 's .
N. ,., . ..
-2 i
V Ihe increase in C.l;3,q Oue *.* '.5t .....y *.- eragypg 7pgg..sgp a " IU' de:rease cue :: :.ie :ressure chan9e. .ne licensee c:ncludes ..a. s,.ngle r:g dr-~~ ,~,... - . . . 5 bcunded by *b. -
.%.e
-- 3t-a..a.
-. . g r:d m,. salt
'# r'g"..'
ang thus has accep;aDie C:nsecuences, CONC'.US ION Base: on cur review ne cen:ur w*5 -ge licensee,s cenclusions hi C n~ r~e" e is ba s ed on the fac*' *'ha *'h* *~ransien; analysis is per<-- *a u, .s
-e same calcula:icnal me-5:ds .ha. were 'use"* **n '*ne raar. ana ysis and that the :=nse:vences are a -~~>%se.
ne c:nciute ~"a ~ -se pr :: sed revisten : ~ * - . .
~ ~ . .. . r.s e. n e ru n t a ~- k c '. . -.J '. .. y <. s a c - .' . .a ~ 3 u
react r remains in manual ::n:r:1 ....-s.e. Pi[ i'al'C:ntribu;;r: n.'-r::<s i . l I
.mm e L 4 e I
e l L 6 i e 0 4 l 6
l, - f i + 'o,, UNITED STATES 'j
![.. ... c NUCLEAR REGULATORY COMMISSION
{. wasM NGTON,0. C. 20555
%*****/.I January 6,1983 ..
Docket Nos. 50-250 ' and 50-251 Dr. Robert E. Uhrig, Vice President Advanced Systems and Technology Florida Power and Light Company Post Office Box 529100 Miami, Florida 33152
Dear Dr. Uhrig:
~
SUBJECT:
DELETION OF THE FLUX RATE INPUT TO THE TURBINE RUNBACK SYSTEM The staff has completed its review of your submittals dated August 10 and
-September 7,1982, relating to the above request. ~
Based on our review, we have concluded that the proposed revision to the Turbine Runback System is acceptable provided the reactor renains in the manual control mode. The enclosed Safety Evaluation provides the details and results of our review. i SIhcerel/,.'s] ~_ .I 3 1013 (StevenA'Varga{!b[k Chief
, . Operating Reactors Branch #1 Division of Licensing
Enclosure:
As stated cc w/ enclosure: See next page I
. I w .m
. _ . lit __5
? .* .
4 Robert E. Uhrig Florida Power and Light Company cc: Harold F. Reis, Esquire James P. O'Reilly Lowenstein, Newman, Reis and Axelrad Regional Administrator - Region II 1025 Connecticut Avenue, N.W. U. S. Nuclear Regulatory Commission Suite 1214 - 101 Marietta Street - Suite 3100 Washington, D. C. 20036 Atlanta, Georgfa '30303 Norman A. Coll, Esquire Steel, Hector and Davis 1400 Southeast First National Bank Building Miami, Florida 33131 Mr. Henry Yaeger, Plant Manager i Turkey Point Plant Florida Power and Light Company P. O. Box 013100 Miami, Florida 33101
- Mr. Jack Shreve 0ffice of the Public Counsel Room 4, Holland Building -
~
Tallahassee, Florida 32304 Administrator Department of Environmental Regulation Power Plant Siting Section State of Florida 2600 Blair Stone Road Tallahassee, Florida 32301
~ 12 -- Resident Inspector .
Turkey Point Nuclear Generating Station U. S. Nuclear Regulatory Commission
- Post Office Box 1207 Ho~mestead, Florida 33030 .
l l' I . em 4 '"
.' ENCLOSURE SAFETY EVALUATION i MODIFICATIONS TO TURBINE RUNBACK SYSTEM INTRODUCTION By letter dated August 10, 1982 and clarification dated September-7,1982 the Florida Power and Light Company proposed to. modify the Turbine Run'back System for Turkey Point Units 3 and 4. The modification would delete the' flux rate input to the system in order to reduce the number of spurious runbacks. The Reactor Physics Section of the Core Perfomance Branch has reviewed the proposal.
Our evaluation follows. In the rod bank or single rod drop ' event a turbine runback to 70 percent of full power occurs when either a rod-on-bottom signal or a negative flux rate signal from one of the excore detector channels is received. A failure in the detector channel then results in an unnecessary turbine runback. If the flux rate input is removed from the turbine runback circuit the only
~
protection against rod drop events will be the rod bottom signal. If one - assumes a single failure to occur in the protection system (i.e., the failure of the rod bottom signal) then a runback will not occur when a single rod drops. Protection for a rod bank drop will still be provided since in this case there are nultiple rod bottom signals. EVALVATION T._. __ The single rod drop event has been reanalyzed for the Turkey Point reactor assuming that no turbine runback occurs. The power mismatach input to the automatic rod control circuitry has been disconnected and the reactors are operated in the manual mode. Thus no overshoot occurs when the reactor returns to power af ter the drop. The return to power. occurs as a result of the reduction in the average moderator temperature. The temperature reduction, in turn, causes a reduction in reactor pressure. The core stabilizes at essentially full power with the dropped rod fully inserted. This cast has been analyzed for nomal temperature and pressure in the FSAR as the limiting rod misalignment case and shown to result in no violation of fuel themal limits. The effect of the reduction in temperature and pressure in the dropped rod case has been analyzed by standard Westinghouse analysi,s methods.
- q-~
g . 2 i The increase in DilBR due to the temperature reduction is greater than the decrease due to the pressure change. The licensee concludes that-the single rod drop event is bounded by the limiting rod misalignment event and thus has acceptable consequences. " CONCLUSION Based on our review we concur with the licensee's conclusions. This concurrence is based on the fact that the transient analysis is performed by the same calculational methods that were used in the FSAR analysis and that the consequences are acceptable. We conclude that the proposed revision to the turbine runback circuitry is acceptable provided that the reactor remains in manual control mode. Principal Contributor: W. Brooks , O 6 8 m M e & T e
=
e MD e e gg, W. f___.
,.i / \q,
^, UNITED STATES A [ j c NUCLEAR RECULATORY COMMISSION
; wAstumotom, o. c. zons
//,,
**.....**' July 29, 1985 MENORANDUM FOR: Hugh L. Thompson, Jr. , Director .-
Division of Licensing : Office of Nuclear Reactor Regulation FROM: Brian K. Grimes, Director Division of Quality Assurance, Vendor and Technical Training Center Programs Office of' Inspection and Enforcement
SUBJECT:
POTENTIAL BOARD NOTIFICATIONS ON ALLEGATIONS CONCERNING WESTINGHOUSE The enclosed document contains a number of allegations concerning Westinghouse, Water Reactor Division. (This document is in the PDR.) Westinghouse is aware ~ of the allegations and has stated in a July 12, 1985 letter (copy enclosed) that "Our preliminary review has revded m actual safety deficiencies as a result of the alleged incidents and practices."
- The Vendor Program Branch, IE, will conduct an inspection at the Westinghouse Monroeville Nuclear Center in August which will include a review of the safety significance of each allegation. An hRR participant knowledgeable in the technical areas discussed in the allegations and their potential effects upon licensing considerations, if any, would be helpful.
Please contact me or Gary Zech if you have any questions re'garding the enclosed document and whether you can provide an NRR member for the August inspection. Pending that VPB inspection, an NRR review of the allegations is recommended to determine whether appropriate board notifications should be made , this time.
/} L
~.! _
r- n c
/ Brian K. Grimes Director Division of Quality Assurance, Vendor and Technical Training Center Programs Office of Inspection and Enforcement
Enclosure:
Letter dated June 17, 1985 cc w/ enclosure: J. Taylor, IE R. Vollmer, IE G. Zech, IE _
# Craig, IE B. Sheron, NRR ^
V. Noonan, NRR
. c c d y (1 3 % 3 ~
N bh
lh . 9201 Wedgewood Dtive Pittsburgh, Pennsylvania 15239 June 17, 1985 .. To: Director, office of Inspection and Enforcement U.S. Nuclear Regulatory Commission Washington, D.C. 20555
Dear Sir:
Until July 31st of this year, I will hold the position of Senior Engineer in the Westinghouse Water Reactor i Divisions / Nuclear Technology Division / Nuclear Safety Department / Risk Assessment Technology section. Since joining Westinghouse in 1980, I have been located at the Monroeville Nuclear Center, Monroeville, Pennsylvania, where I primarily perform accident analyses. over the past year, I have either been involved in or have knowledge of incidents that I believe are violations of either nuclear regulatory law or Westinghouse quality assurance requirements. These incidents have categorized and are shown been as Items 1 through 8 on the attached sheets. It is requested that you investigate these incidences and take appropriate action where necessary.
~~
If I can be of further assistance, please call me. Sincerely, Q R . S 6-lk hn A. Segletes i] {h Phone (412)795-2795 h
- 1. LOST DIABLO CANYON SAFETY ANALYSES During a group meeting held in late 1984 for the Plant Transient Analysis and Operating Plant Analysis groups, it was stated by the manager of Operating Plant Analysis, Pat Loftus, that almost all of the calculation notes that support the Diablo Canyon Final Safety Analysis Report are missing. Apparently these supporting analyses were lost when they were to be put on tape in 1974 To the best of my knowledge, these lost analyses have never been retrieved and no attempt has been made to inform the Westinghouse Water Reactor Divisions Safety Review Committee, the NRC, or the customer of this situation.
Also present at the meeting were the Manager of Plant Transient Analysis, Melita Osborne, and approximately twelve engineers and technicians from both groups. I believe this is a safety violation since not keeping records that are required by a licensed condition is a violation of 10CFR50.71, Part C. 4 J Mb ew h l b l
.-- . -- -,_ -- - - - - - - - - . - - - - - - - - , - - . - - - - , - - ~ , . ,- - -
- - - , - - - - - - - - - - - - - - - - . . . - - - - - - - - - - - ~ - , - - - - - - - - - ~ - - - - -
- + - - - - - - - - - - - - - - - - -
. . _ _ __ - . - _ _ _ ._= - __ - .- .- -
- 2. NONDISCLOSURE OF AN UNSAFE PLANT CONDITION AND RETALIATION BY MANAGER ,
During November of 1984, I was assigned the
- task of performing an analysis to evaluate the impact of (emoving the flux rate signal device from the Indian Point 2 ' nuclear power plant. This device is used to initiate turbine runback to protect against departure from nucleate boiling in case a dropped rod or dropped bank accident occurs.
Redundant protection is provided by a rod-on-bottom signal ! device which also causes a turbine runback. The j rod-on-bottom device operates concurrently with the flux i rate signal device to provide the redundant protection.
)
l Before I started the Indian Point 2 task, I reviewed a ? similar study that was done for the Turkey Point units (se,e
- CN-TA-82-104). It immediately became apparent to me that deleting the flux rate signal device at Turkey Point violated the single failure criteria as specified in IEEE _
279-1971 " Criteria for Protection Systems for Nuclear Power Generating Stations". This is because the rod-on-bottom device, by itself, is not totally redundant. When I
. informed my manager (and author of the Turkey Point ~
. analysis), Melita Osborne, of this violation, she said
" John, do not disclose this information or we will be .
J sued." (I presume Melita meant Westinghouse would be sued I by Florida Power and Light (FP&L)). After some further I discussion on this matter, I dropped the issue because I i believed Melita would retaliate against me if I pursued it further. I { Independent of my finding, FP&L later recognized the same unsafe condition existed that I called to Melita's
- attention in November 1984 In early 1985 FP&L issued an j LER to report this problem. This time Melita did not 1 attempt to conceal the problem nor did she inform FP&L that I had previously determined this probita to exist. The l Westinghouse response to the'FP&L finding was documented in i Letter NS-RAT-PTA-85-091 which provides recommendations on
, how FP&L should modify the existing hardware to make the j system redundant. '
- On the 29th of January 1985, I had my performance j appraisal for the year 1984 and was informed by Melita
- Osborne that I was being terminated from Westinghouse on
! July 31, 1985. I believe a factor in my termination was
- retaliation against me for uncovering this faulty Westinghouse recommendation of which
~ Melita was the originator. 1 i i
l
. l
- 3. FEAR OF RETALIATION t
As noted in Item 2 shown on the previous page r I was l required to perform a safety evaluation for Indian Point 2, ! similar to the one that Melita had performed for the Turkey ( Point Units. In the Indian Point 2 analysis, I stated that i the rod-on-bottom unit by itself was not completely single failure proof. (see Page 10 of CN-TA-84-202). On the other hand, I did not disclose this fault in the customer report (NS-RAT-PTA-84-171) since disclosing it would result in either the Indian Point 2 and Turkey Point units having to undergo substantial modifications (to make the rod-on-bottom signal device single failure proof) or the flux rate signal device could not be removed from service, which would negate the need for the analysis. Furthermore,' based on Melita's response to my finding in the case of <the Turkey Point Units, I feared retaliation by Melita if I disclosed City. this fault to Consolidation Edison of New York I discussed this dilemma with two
- of my colleagues, -
Thomas Blackburn and, to a lesser extent, Mark Adler. Thomas Blackburn later checked my calc note. ,
'1 \
1 il } ) l l 1 i
,,--.--,n,-,_e..-n.._,.--...----- ---, -
n ._,,n_,.,----n- .~.,-n -,-- - , ,,,-- -.,--,---n- ~ - . - - - - , - - - - , - .
- 4. NOT REPORTING APPARENT SAFETY VIOLATIONS TO THE REVIEW COMMITTEE SAFETY A. Analyses That Raise the Reactor Setpoint Trip on Turbine Trip me In January 1985 a colleague, Mark Grace, point 5d out that a Coman:he Peak plant specific to previously checkei. (CN-TA-84-97) to study he had justify raising the setpoint for deletion its then existing value was in error. The error was the of reactor trip on turbine trip above result the of not transferring transient inlet temperatures to THINc3 computer, code where they can be calculate As a result, the initial the departure from nucleate boiling ratioused to (DNBR).
used to compute DNBR (constant) inlet temperature is - throughout the transient. This is unconservative inlet temperature rises approximately since in some cases analysed the transient 20 to 30 degrees Fahrenheit above the initial temperature by the time the
- minimum DNBR is reached.
a dozen I reviewedstudies our of this files type andhad determined been done that approximately in only one of these studies (CN-RPA-78-66) didpreviously and the correct inlet temperature history. THINc3 use I wrote a memo on January 25, 1985 to my manager, '- Melita Osborne, informing her of the 4 computational method currently being nonconservative 1 out that this error used, while pointing probably exists in several studies and that the problem should be reported to other Westinghouse Water Reactor Divisions the (WRD) Safety Review Committee (SRC). When I later spoke to Melita 4 the note, she criticized me for calling the problem regarding to her attention and said she would take care of it. No plan for resolution of this problem set-up as required in Risk was quickly ; Assessment Technology (RAT) procedure NS-RAT-IG-9 nor was the WRD SRC alerted of this { j potential issue within the'first two weeks as required by
' Item 10 of NS-MAT-IG-9. To the best of my knowledge, this problem has never been reported to the SRC and it has only 1
been corrected in two cases. i 8 Dropped Rod Analyses for Turbine-Runback Plants I In early 1985, Glen Hebele, while working on a study to justify an increase in the turbine runback setpoint ' Turkey Point Units 3 and 4 (see CN-TA-85-6), discovered for an i error to exist in the dropped rod methodology as outlined l l
4.(Continued) in NS-TA-83-365 which yielded nonconservative results. Specihically the dropped rod methodology calls for i performing the analysis at a turbine analysis limit 4% less than the turbine runback runback safety setpoint. However, the safety analysis limit should be 44 more than ; the setpoint value. This 84 error was incorporated into ' the following plant specific safety analyses. Point Beach 1 and 2 ! Turkey Point 3 and 4 Indian Point 2 and 3 Ginna Bosnau - Glen Neberle informed me he reported this error to Melita was osborne, but, no plan of resolution of this problem quickly set-up in accordance with RAT procedure NS-RAT-IG-9 nor was the WRD SRC alerted of this potential issue within the first two weeks as required by Item 10 of NS-RAT-IG-9
~
To the best of my knowledge, the problem was -
~ never reported to the SRC nor have any of analyses been corrected. the erronious 0 l
{ . --
! t i \
l i l 1
- 5. NONUNIFORM PROCEDURES WITHIN WESTINGN3USE FOR REPORTING
' POTENTIAL SAFETY VIOLATIONS i
Information provided by Westinghouse management shows three different sets of procedures that should be followed
> in reporting potential safety
! Westinghouse Water Reactor Divisions (WRD) Safety violations to the Review Committee (SRC). . In my opinion, the procedures that are specified by i first and second level managers (see B and C below) can, and do, lead to intimidation as discussed in Item 2 and
- burying" potential safety problems as discussed in Item 4 Also, I think it is the intent of the NRC that one, only one, set of procedures be used within Westinghouse and to report potential safety violations.
A. Posted in the main lobby of the Center, Monroeville, Penna, Monroeville Nuclear ) (1.) Report violation to supervisor, or-(2.) Report violation to Manager's Representative j the WRD SRC, or- on
, (3.) Report violation to R.A. Wesemann, secretary of j the WRD SRC. ] , B. Stated in the Radiological Assessment Technology -
4 t . Instruction Guidance Material . .
! (1.) Report violation to supervisor, then (2.) Get supervisors approval, then (3.) Provide plan for resolution of the problem to i SRC. the If the supervisor disapproves your request to report the potential safety item, you may
~~~
(4.) Report directly to the WRD SRC. C. Stated in undated memo provided
< r Transient Analysis and Operating Plant Analysis during to members of Plant meeting in late 1984 and also provided in a Nuclear Safetya Department handout to all members of the Department in early 1985 Nuclear Safety (1.) Get supervisors approval.
r Note: I have asked for clarification regarding in letter NS-RAT-PTA-85-047 The response to this issue ' i my request (see Letter NS-RAT-PTA-85-051) states the memo referred to l in Item C above was only intended to be a ' guideline', but there is nothing on the meno to indicate it was only intended to be a guideline. r 1 - i i I r
------,------,,-n--,----r----e,v- w-----,-- -- nm n,wnm.,,..- ---,-mm,_,-,-y_,-,_nwn,n-- -,,-e ..--,nn-4
i. 6 VIOLATION OF QUALITY ASSURANCE PROCEDURES A. Transmittal of Preliminary Draft Reports
- Letter MS RAT-83-036 dated Noveber 29, 1983 states the requirements to be followed within the Risk Assesment Technology (RAT) section with regards to preliminary draft reports outside the transmitting RAT group. The requirements are the following:
. (1) The transmittal letter should state the information is preliminary.
(2) The report should be stamped " PRELIMINARY". (3) First level manager's approval is required. ' On December 16, 1983, a preliminary copy of the Italian Reference Plant functional requirements were sent out (see Letter Ms-TA-83-520) without any of the above 9 requirements implemented. Note that requirements did not go through the normal these functional bat rather were in-house review to be reviewed by the
- (NIRA/SOPMEN). customer s.
- s. Assigning a Competant Independent ;
Risk Assessment Technology (RAT) Sectionverifier Within the Westinghouse Nuclear Technology NTD-DPP-38, Rev. 2 dated 7/24/81 and RATDivision procedure I section procedure NS-RAT-IG-2 state that the cognizant (or Appropriate RAT) manager shall assign an engineer to act as the independent i (or RAT independent) reviewer. This procedure is rarely 1 if ever followed in the Plant Transient Analysis i Plant Analysis groups. or Operating In fact, I requested that my manager, Melita Osborne, assign an independent checker 1 , check one of my calculations to ! difficulty in finding a n' independent (CN-TA-85-29) when I had l reviewer. Nelita returned the that I should find my own independent cale note later with an attached note stating 1
- reviewer. Tor.
l Blackburn did check and recolled seeing the note that Melita wrote CN-TA-85-29 when I asked hin to do so when I called the incident to his attention on March 25, 1985, f 1
~
i l i _ - . _ _ - _ . _ . _ _ . _ _ _ _ _ . _ . ~ . _ _ _ . . _ _ _ . _ _ _ _ _ _ , _ _ _ _ _ _ , _ , _ _ _ _ _ _ _ _ _ . , _ _ _ _ _ . _ _ _ . _ _ _ . . _ _ . _ _ ______
7 THREATENED RETALIATION FOR SENDING WRITTEN MESSAGES On Thursday morning, February 14, 1985, my manager, Melita osborne, called me into her office and told me she would terminate my employment with Westinghouse Vith two 't months notice if I continued to harass her. What she considered harassment included only the following items. 1 Writing Letter NS-RAT-PTA-85-047 which requested clarification on the correct procedure to use to report potential safety problems to the Westinghouse Water Reactor Divisions safety Review Committee. P 2 An informal meno dated 2/7/05 from me to Melita - asking why the normal in-house review and comment procedure was not followed for the Italian Reference Plant's Functional Requirements (Letter NS-TA-93-520). .
- 3. An informal memo from me to Melita stating that I
, planned to give the Italian Reference plant's Back-up Protection System Functional Requirements a PRELIMINARY i status until they were checked by the customer since this would conform with NS-R AT-83-036. f .' 4. An informal memo from me to E.P. Rahe and D.C.
~
i Richardson regarding a complaint by consolidated Edison of - i New York City that Westinghouse had never called back when they (Consolidated Edison) requested a meeting between Consolidated Edison and Westinghouse a month earlier. I also expressed my concern that our good business relationship with Consolidated Edison was being strained a because of this incident. 4 - - -
- ~ ~ -
Melita demanded that any future communication I have with her be limited to verbal communications. i )
~
~
i , l 1
. - _ - _~ . - - . . - - -. -. - -. ------_ __ . _ - - - - - -
, 4 POOR CALCULATION NOTE CHECRING WHICH RESULTED
- QUALITY ASSURANCE VIOLATIONS AND IN NONCONSERVATIVE COMPUTER
! INPUT DATA i A review was made of CN-TA-44-63, "CGE Deletion of i i Reactor Trip on Turbine Trip Below 50% Power (P-9)" by E. Kurt Nackman and checked by M.P. Osborne. Numerou,s errors ' were found to exist in the analysis. Most, if not; all, of these errors should have been detected by i reviewer. the independent i i ' The following errors were noted: I
- a. The Model 51 at'eam generator is simulated in this 1
study. It does not have a preheater, but the input data (NODEPE=1) indicates a preheater exists.(One or the other - ' of these inputs is in error.) f
- b. The buoyancy calculations were to be t
! turned off for
- conservatism was in (2 CORE =IRVO=38GT=ISGP=0) per page 9, but 3 CORE !
fact set to 120.0. (This error is nonconservative direction.) in the r ' c. The transient vessel inlet temperature, as the LOFTRAN code, increases with time but this computed by data never.
- got into boiling ratio.
the THINC3 calculation of departure from nucleate (This error is in the nonconservative direction.) . f 1
- d. The front page of CN-TA-84-63 is not completely out. (Violates NS-RAT-IG-3 procedure.) filled ,
z
- e. The checklist shows CN-TA-84-63 !
j to contain a purpose 3 , _, and results near the front. In fact, a purpose and results ! ! _ - _ _ are not shown near the front of the cale note. l j j f. The Introduction section (page 3 states four cases were analyzed, but only three cases are) shown. ! g. 4 The Table of Contents on Page 2 is not completed. 4
- h. Information that should appear in the " Analysis !
)
and calculations
- or
- Input Listing" sections (pages Method i 2 to
- 40) are actually put into the Introduction section. l t
- 1. No sample calculation is shown, but the checklist i shows the cale note to contain one. !'
1 j. j The checklist page is not numbered nor is the cale note number shown on the checklist page. (If this page were j separated from the cale note, there would be no way to identify the cale note it came from.) \ l
- t
0
; 8.(Continued)
- k. The in'put listing for the third case (violates NS-RAT-IG-3 procedure) is ni>t shown.
1 The microfiche identification numbers cover sheet. are not on the (violates NS-MAT-IG-3 procedure)
- m. The covar sheet requires a managers signature, there is none. bdt
- n. Microfiche identification numbers are anywhere in the cale note. (violates not shown
' NS-RAT-3 procedure)
- o. The P-9 uncertainty already .
l uncertainty. includes a nuclear flux It is not necessary to acccunt for this uncertainty twice as is done in this analysis. '
- p. The 24 uncertanty noted on page 47 j
uncertainty, not a LOFTRAN uncertainty, is a nuclear flux t I -
- q. On page 11, 5 lines from should be 4 degrees the bottom, the last ters l
Fahrenheit uncertainty, uncertainties. not 4%
- r. Use of GEND3 indicates a Model D3 steam generator should be used. The LOFTRAN input assumed the i Model 51 steam generator. (one of the two calculations is in error.)
- s. On page 16, DKSCRA=
a trip reactivity. .04 is not shutdown margin, it is t _ _ _ .
- t. On pages 28 and 29, the statement
; is made that modifications were made for 52% power, but they were t
fact made for 60% power. , in u. On page 28, no numerical value is given for NORDER.
- v. QFINTL requires an input for each loop. The proper input should be QFINTL=3*l.0, not QFINTL=1.0
- v. On page 56, third paragraphs *-- a rapid increase in coolant temperature
- probably was intended to be *-- a rapid increase in coolant pressure *
- x. On page 62, middle of second paragraphs the power operated relief valves are actuated, not the safety valves.
I l ) _ - . - - - - - - - ^ - - ' - ' ~ ' ' ' _ _ _ _. --
t 8.(Continued) i !
- y. On page 63, last paragraph '-- pressure -*
PORVS should be '-- pressurizer PORVS - '.
- 2. There is no indication where the two typos referred to in Revision 1 are located. They should be clearly marked by a bar in the right margin along with the appropriate revision number, but I don't see any such marking.
1 (violates NS-RAT-IG-3 procedure. ) A review made of analyses which justify raising 'the setpoint for reactor trip on turbins trip above the typical ! 104 power level has shown the independent reviewer, Melita i Osborne, has never previously performed analysis. Therefore Melita, or her manager,this type of should have 3 disqualified her from being the independent reviewer of t this cale note.
~
l . l I l l 9 ME +m 1 i e a 6
3 v Westinghouse Water Reactor Neendecnneret,omsron Electric Corporation DMslons se,333 Pms::urgh PennsyIvafta 15230
, July 12, 1985 )
Mr. James M. Taylor, Director Office of Inspection and Enforcement U.S. Nuclear Regulatory Commission Washington, D.C. 20555
Dear Mr. Taylor:
This is to confirm our telephone conversation of July 11, 1985 in which we informed you that on that date we first became aware of a letter dated June 17, 1985 and sent to your office by Mr. John A. Segletes, a former employee of Westinghouse Water Reactor Divisions at the Monroeville Nuclear Center. Mr. Segletes alleges in this letter certain practices and events in which he was involved while an employee which he - believes were violations of either " nuclear regulatory law or . . Westinghouse quality assurance requirements", and requests your investigation and appropriate action. Please be assured that we shall cooperate fully in this regard, and that we have begun an internal review of the factual matters contained in Mr. Segletes ' allega tions. Our preliminary review has revealed no actual safety deficiencies as a result of the alle=ed incidents and practices. Moreover,
- ~ ~ ~
we have found no reason to believe that Mr. Segletes or any other employee has been inhibited from raising safety concerns through the established channels as defined in the company's policies and procedures. Our continuing review will place highest priority on verification of the safety of licensed facilities, and our findings will be communicated as appropriate to affected licensees and to your office. Please at call me any time. (412-374-4868) if I can be of further assistance Very truly yours,
^wn s , -
f P. Rahe, Jr., Manager Nuclear Safety Department 1 h#sOfhf 850712 -][h/I CF f* n(p
/Y /0}}