ML20085L476
| ML20085L476 | |
| Person / Time | |
|---|---|
| Site: | Dresden  |
| Issue date: | 12/17/1971 |
| From: | Brian Lee COMMONWEALTH EDISON CO. |
| To: | Morris P US ATOMIC ENERGY COMMISSION (AEC) |
| Shared Package | |
| ML20085L460 | List: |
| References | |
| NUDOCS 8310310093 | |
| Download: ML20085L476 (10) | |
Text
- -
2.- 4x O O ,fo-a'/ 9
' $. . t s
Ch Commonwealth Edison Company -
our rinst warionAt ptazA
- C H I C A G O, ILLINol$ (
us.un.eu .
r Y ??- ,
Polf OFFICI $0E 747
- C M I C A G #) . ILLINOlt 40490 [
December 17, 1971 E " sj g, '"
l Dr. Peter A. Morris, Director ~
Division of Reactor Licensing ,
M
U.S. Atomic Energy Commission Washington, D.C. 20545
.Id
/ i
Subject:
Report of Safety Valve Operation following a Feedwater Transient - Dresden Unit 3 (DPR-25) !
Dear Dr. Morris:
This is to report a condition in which a condensate booster '
pump tripped, which caused a reactor water level. transient. This water level transient resulted in filling the main steam line with '..
water, opening of a safety valve, and pressurization of the drywell. 4 The following information pertaining to this occurrence is submitted:
pending completion of the investigation which is currently in progress. ,.
Summary
,A 1413 on December 8, 1911, as a result of a condensate 2
booster pump trip on,Dresden Unit 3, the reactor feed pumps tripped.
Tripping of the feed pumps resulted in a reactor water level transient. ~]
This eventually resulted in filling the main steam lines with water, opening of a safety valve for approximately 1 minutes and pressurization of the dryyell. Pressurization of the drywell.resulted in a high drywell signal which initiated starting of emergency diesels, l
low pressure core cooling pumps, and HPCI. " '
C1 1
During the transient, drywell pressure reached a maximum
/kl of 20 psig.
The maximum and minimum reactor pressures were 1050 psig f(, and 705 psig, respectively, and the reactor water level reached a l
g minimum of -20 inches and a maximum of +130 inches. With water level ,
Jp at -20 inches, there is more than 9 feet of water above the fuel. A gg detailed sequence of events is attached. ,
bl]
All safety systems functioned as designed. No significant radioactivity was released to the environment as a result of the
{ incident.
During post incident recovery, both the primary system and l
the primary containment were maintained in a " bottled up" condition until analysis of reactor water and containment atmosphere could be made.
, . 8310310093 720505 /
PDR ADOCK 05000249 f-i* S PDR G.Q'M . }
C,
Commonwe. tith Edison CompO Dr..* Peter A. Morris t
December 17, 1971 Damage Assessment A preliminary inspection in the drywell following the incident revealed damage to the following equipment:
1 - The rupture discs on all the safety valves showed cracks.
This may not be related to the incident since this condition has been encountered previously on normal shutdowns.
2 - The 3A electromatic valve was damaged by the steam jet from the "F" safety valve. One steam discharge rams horn on the "F" safety valve was directed towards the electromatic valve.
The cover on the solenoid assembly of this valve was blown,
. off. The holding coil portion of the solenoid assembly was found open. Wiring to a position indicating limit switch was also damaged rendering the position indication circuit inoperative.
3 - Miscellaneous thermal insulation was damaged and requires repair.
4 - The top coat of paint on the containment wall over an area about 3' x 3' was removed by the steam jet from the "F" safety valve impinging on the surfacc of the containment.
5 - Sections of ventilating duct in the vicinity of the steam jet were dislodged and require repair.
6 - The LPRM cables were found damaged. Thiscablemustbereplaced) in total.
7 - The SRM/IRM cables were tested and found to be good. Following the Dresden Unit 2 June 5 incident, this cable was replaced on Dresden Units 2 and 3 and Quad-Cities 0
Units 1 and 2 with cable having a higher temperature (302 F 10 hour1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> rating) rating.
8 - One containment cooling fan motor was found to have a ground caused by moisture in the containment. This motor will require drying out. The other six cooling fan motors were found to be.
in good condition.
Preliminary Conclusions The following conclusions have been reached regarding the Dresden Unit 3 incident of December 8, 1971:
's
Commonwealth Edison Comp]
- Dr. , Pete.r A. Morris December 17, 1971 1 - There were no radiological consequences since no significant release to the environment resulted from the incident.
2 - No compromise of the health and safety of the public resulted from the incident.
3 - All oper1tions during the 16cident and post incident recovery {
period were within Technical Specifications.
l 4 - All safety systems functioneN as designed including High Pressure Coolant Injection (HPCI) { Low Pressure Coolant Injection (LPCI),
Core Spray, Main Steam and Containment Isolations, Standby Gas Treatment System, Pressure Suppression System, and Standby Diesel Generators.
5 - Feedwater control system performance during the transient was deficient, in that, the control system locked out on low air pressure, probably during rapid valve movement. Previous experience has demonstrated the inability of the feedwater control system to automatically control level below the high water level trip point for main steam isolation valves during a system transient. This was the primary reason for the need to take operator action.
6 - Operator response was in accordance with operating procedures throughout the incident and post incident recovery N th_two__ i exceptions. The operator did not, reset the feedwater regulating
~
valve lockout condition when i_t ocE rrEd 7and he did not trip -
the feed pump when the water level reached +60 inches. Had he done so, the incident.may have been prevented. It is important to place these actions in proper perspective, and it should be emphasized that he did take a number of steps to control feedwater input to the vessel. The operator actions were:
(a) He reduced the master controller set point to minimize the error signal between the actual level and set point level.
This response was previously established on shift by General Electric during the startup program to compensate for the known overshoot which has been experienced follow -
ing scrams. While not specifically called for by the station opera ting procedure 600-ANI, it is consistent with the intent of the procedure to keep the level on scale.
'. ", . p Commonwealth Edison Compy J
Dr..' Peter A. Morris December 17, 1971 k c ,s He closed the minimum flow feedwater valve. p' Ag . ' "
(b)
(c) He reduced the manual output control potentiometer on the I
" manual-auto" controller to zero and transferred them to manual to terminate feedwater input. ,
1cb 7 The "F" safety valve lifted at approximately 1020 psig reactor C"';%
pressure. The safety valve set point is 1240 psig.
The lifting of this valve was probably caused by some mechanism resulting from the effects of feedwater flooding the main steam line.
The pressurization of the drywell could probably have been ' ! N'"
~ ~ ~ ~ ~
avoided if this valve had not lifted.
Corrective Actions The following corrective actions will be accomplished prior to startup:
p ec, p7
$ff 4 1 - Safety evaluations. 1P'pffF fpv' '
gpdb ,g O (a) Effects on fuel yhky.
(b) yessel internals (c) Performance of suppression pool (d) Effects of pressure, temperature and steam impingement on primary containment (e) Differential temperature on vessel 2 - LPRM repair 3 - Thermal insulation repair 4 - Replace "F" safety valve with tested valve 5 - Check operability of 3A electromatic valve 6 - Investigate reorientation of safety valve discharge -
l
~
l i,
5-
- - ~
~~'. OCommonwealth Edison ConyQ
- Dr.? Peter A. Morris December 17, 1971 7 - Check calibration of feedwater control system and verify feed-water regulating valve response to control signal. \oa C"'
8 - Increase torque setting on feedwater regulator isolation valve so that it will close under pressure and seat properly.
9 - Modify and emphasize procedure for handling water level transients.
Leave control system in automatic, reduce feedwater controller set point and trip feed pump at 45".
10 - Evaluate the need for non-destructive inspection of the main steam lines.
11 - Check calibration of head-to-flange temperature indicator. - -
12 - Test all electrical penetrations and main steam line bellows.
13 - Test all motors in drywell.
14 - Repair damaged paint on drywell wall.
15 - Perform functional test on all equipment exposed to drywell incident environment.
o 16 - Check system.for freedom of movement during startup. -- {
17 - Hydrostatic test of 1000 psig.
Additionally, changes to the feedwater control system to improve its operation are under consideration. When our investigation is complete, we will file a final report with you.
Very truly yours,
-s ,
3
-f f/ J i's en%')C7 i .. Ad -
Byrpn Lee, Jr. l Assistant to th President cc: Mr. Boyce H. Grier, Director
] Region III Ccmpliance t
' ~
t
)
\
SEQUENCE Of ETJNT_S DECEMBER 8,1971 INCIDEhr Initial Condic1e_ ns a) Load - 792 tGe; approxL.ately 2300 int; the baso load ceMition was being =sintained for daccruination of chiwy mouitor sas:pling syste:n plate out. ,
b) Ster = ficv - 9 x 10 6 lb/hr.
Feedvater flov - 9 x 106 lb/hr.
c) Equipmat in-service - 33, 3C, 3D condensate booster pu- ps (3A pu=p was out of service for aligrt:ent and gland Icak);
3A 6 3C reactor feed pt=ps (3B pump in standby); 3B fecevster regulating valve in service (3A feedwater regulating valve in standby): low ficu feedwater regulating valve open approximately 20L d) Dryvell - inerted and at norant pressure of approxir.ately 0.2.5 psig. .
ECE: A graph of signifiesnt para::eters during the initial transient is attached.
, Secuence of INents_
~
e'
- 1) 14:13:08 3C ceedensace-booster purry tripped.
If0TE: Cause of trip m'<nown. The pux:p can only be tripped on under-voltage cr overcurrent. Prior to the pu:o trip coc.nonwealth
' Edison 8*n--y sad General Electric Ccepany personnet were 3
present in the ccndensats pu=p rocca working on a circulacina water hydraulic unit. They noted a very loud noise eccing fras 3C pu=p, It was prestred the pv=p tripped on overturrent as a result of c:cchanical binding. Neither the overcurrect or under It has voltaae relays at the pu=p break.ir showed a trip target.
act been detereined if tiie candensate er the booster pu=p was the -
cause of the noise.
- 2) 14:13:09 3A aM 3C reactor feed pu:npa tripped on Icw stectica pressure.
t
- 3) 14:13:10 Standby reactor feed pump 3B started.
- 1
,. , g a
(,.- x / l
. u
\
- 4) 1/ct}y21 - 22 l a) Ecester wcter levc1 decreased rapidly.
, b) Esectes scram on low water level. I i
c) Raceter varer Icvel continued d:r.:nvard rapf.dly to a;;reximtel y - 20". (r.CCS Initiatien is -5'1".) .
d) to cperator observed the follouing indicariene:
- 1) De " Flav 0=" light on the benchboard, indicating, a runout condition, blinked oc ed off.
- 2) to "Feedwter Panp Max Capacity" ar_miator cem up. his is also an indicatiers of a runout condstion.
- 3) E e "Less of Air to F. W. Ecg. Valve A" and "Less of Air to F.W. Ke;;. Valve B" annunciatore. nis alarm occurs at 85 psig, Lockout occurs until 75 pefg. ,
i e)
Vessel level started to inercase sed the operstar took canual actica batuce:2
-20" and -12" in enticipar. ion of a rapid IcVel >
increzan, as follows: .
1)
Ecduced "=catcr feedwater centro 11er" set point, to c1csa the feedwater regulating vaiva.
2)
Reduced manual output control pctcutleters on 'tsnpl/
auto" controllers te zero.
- 3) Transferred frac: " auto" to "cenual" an 'banucl/cuto" stations,
- 4) Eacually closed low flow feedvater replating valvo. _.
f) At -12" vessel level hesitated and the operaror started opcuf.ng l
the Icu flow F.W. valve to increase icvel.
g) Eewtor watcr level bcgen to increase rapidly.
b)
Aa scen an etc operator verified that icvel use. Increasing, he closed the low ficu feedwater velve ag,ata. (racn=ated at '
approximately zero inches reactor veter Icvel)
,1) As vessel level caca through zero, the operator started cIcsing the feedwater regulating valvc =ctor cperated isolation valve.
again in anticipatica of a rapid level in=renne.
,~ ~ ...
O o !
I
\
the feedvater f1w was i j) As the F.V.1 solation6 valve elesed$lb/hr. At sman tie }
reduced frec 5.7 x 10 to 2.3 x 10 stalled due to high
(
durict, the closure of this valve. ItFlow levelled out at approxt=stely j differcngialpressure.
2.3 x 10 lb/hr. i
$) 14-14:13 Main Steau Line Presa. Iow Croup 1 Isolation en &in Stern Line. Low Press.
Main faa:a Isolatien Valves closed.
Hsin Ste =i Line Drains closed.
Eectre. Loop Sa::ple Line closed, Isolstico Cm,cdsnmer Vent to Main Ste=11nc c1cced.
- 6) 14:14:14 i Reactor pressure reached a low point of 795 psip, after Main Steam Lino isoistion.
- 7) 14:14:29 - 32
- a. Low reactor voter Icvel trip reset (+17")
b.,14vc1 continued up to 80" asd steam lins started filling.
ETE: It is surmised that when the standby reactor feel pu=p started, flow increased to the point where the feedwater regulating Tais isvalve went into a runout ccadition. (Flow centrol mode).
substantiated by the f act that the pu=p flow 6corresponded to a runout lb/hr., and the flow coetrol point of 4pprcx1=stely 5.7 x 10 conditicu.
observation of the annucciator which indicates a runout
- The operatcr stated that the "Flev Cn" light en the benchboard, which a ru=out condition, only blinked on and then a would also indicate enould sve been on steady for the rin:aut I
off (Ihn " Flow Oc" lightAt scc:a ti=e during the reactor Jevel transient "3E" condtica). in an open position. Subsequent feedwatcr regulating valve locked cutcondition usa probably caused testirq has shavn that the lockout by liu air oressure which resulted fran rapid movecent of the valve.
occurred when the runout condition l
It is bc7.ieved that the lockout this time, the valve tried to l was resst on increasing level, at !
close rapidly causing a Icekout on low pressure. l c.. Turbine tripped on high ren ter water level.
- d. Operator broke condensor vacuum manually. I
- a. On increasing reactor pressure, the reactor this time, the operator put was level the isolation above the con-denser in service. et ,
inclation condenser supply line and the operator noted very little {
effect oc presaura vich it in operation.
e o a ~
\
i ll
- 8) 14:18:13 !
Stes= line filled and pressure increased to 1020 psig, jj
- e) 4:
- b) safety valve "3F" lif ted. ,
c) High dryx11 press. (2 psig) una reached. ! .
d) Diesel generater 3 and 2/3 started, e) Core spray Ptzspo started. !
f) LPCI pt=ps started.
i g) Eszitcr Etcirculation Pucps tripped, h) EPCI received an initiattan signal, but tripped on high twel.
- 1) Containemt isolation. (Group II)
- 9) 14118!)9 i Dryvell press. - 5 p >ig. Drywell pressure eerutimied to increase and pezied at 20 psig. then began to decrease.
It is estimated that the safety valve resamked open for approximately li minutes, based oa dryvell pressure inforneciou and reactor -
vessel pressure data.
Following closure of the safety valve, water level continued to increase in the reactor vessel due to feafwater input at a race h of approx'.mtely 2.3. x 106 lb/hr.
1 sector pressure increased gradually due to decay heat input and ..
- increased vesect water inventory.
I 10) 14:26:13 l a) Reactor feedwater "3B" pta=p cripped amually by operator.
l b) Eascter water lev *t at 130".
- 11) 14:20 .
Suppression chamber cooling placed in service. c 14:35 r
12)
Ocyve11 pressure had decreased to 13.5 psig.
- 13) 15:17 t'
Dryvc11 pressure had decreased to 10.5 psig. r 8
(
\
- 14) 18:09*
Drywell pressure has decreased to 4.5 psig.
- 15) 18-C0*
Installed ju:sper to allow r;cning tbs reacter water e.2=ple valves. ,
NUTZ: All systets were maintained in a "bettled-up" condition until s*=pling of retcccr nter and contd.n::ent environ:ncet could be acca::plished.
- 16) 19:1p Raactcr water s:: ple collected. Analysis indicated the activity to be cereal (2.7 x10 7 p C/1.)
- 17) 20:00*_
Drywell pre.asure had decreased to 4.0 psig.
- 18) 20:00* , ,
Folleving reacter water analysis, the cleanup syste: was placci in service a:xi reactor bicwcovn was established for water Icvel contrci. At this point, the levei hed reached approximately
,145" due to control rod drive cooliu6 water input.
l December 10, 1971
- 19) 04:00* ,
" - ~ -
~-- A drywell at=osphere s.t=ple vas obtained in preparation for venting . . _
and entry. The results were: 1 131 1.3 x 10-9 uc/cc
, Beta-c. - 6.3 x 10-11 ue/cc
' Alph.s 1.1 x 10+14 uc/cc
'20) 07:15, Started purging the dryucIl to ths Ennetor Building Ventilation '
Star's la accordence with station operating procedures. Reactor
- water te=perature was 186 F.
- 21) 1_0:45 Initial entry into the drywell was made to obtain atmosphere samples.
Oxygsa ceccentration was satisfactory. Analysis of airborne I activity: 1 131 8.5 x 10-10 ue/cc Beta + , .= 2.3 x 10'11 ue/cc i
Alpha 1.8 x 10-12 oc/cc -
L.
o . .
. . - =
~
i . g
~
, .f ; 4 belegram ,
,l _ Qg
'~'-? .? , .a
= DEB 072 KC054_
gggg
_. . .. f n vo :
K VEB192 AS PDF=TDVE MORRIS ((CC 8 920P CST =__. q g
COMM[(S[Il0N UNITED STATE ATOM [f,C = N 799 R00SE9ECT RD GCEN ECCYN [f.CC hNk t 2DPRE55 DRESDEN NUCl.'E AR PUdEN STAT [f0N UNil 3e THIS WJCG 4;t R.u.a ;..c y
M/f.Ny;.
CONFfRM A CONVERSATTON AT 5945 HOURS TH[1]S DATE W1TH dR E JORDAN REGARDING AN APPE ARAi4CE AT i453 NOUR ON DECEMBER i . NI.
8 I975 WHICH RESUCTED IN A PRIMARY. dONTNfNMENT PRESSUsE .
OF 20 PSTGe THE UNTT WAS OPERATING AT 792 MWE ST@DY, STATE fQ,.u;;
u . ..
1%
WiiEN A CONDENSATE--BOOSTER PUMP TRJ.PPEIb THIS CAUSED COSS .
A4 ._e D J. .:
OF THE FEEDWATER PUMPS ON COW SUCTJON PRESSURE AND A .
'q
.Y % -
RESUCg4I COW RE ACTOR WATER CEVEC SCRAMn APPROX ((M ATEC'( 50 i;gj p' 9. -
y.m .m 2
x y;
'2 E Ed .
- Telegram. - 3M;$b
,;fw, 4 1 - : j. -
SECONDS AFTER Tl1E SCRAM IHE MAff!N STE AM [fjSOC ATION VAC9ES f .' k *,l, . .
?".t.
CQOSED ON IHE COW STE AM C((NE PRESSURE. WATER CE9EC 50SE f ..
RAPIDCY, [f:N THE SEA _ggR VESSEC DURING THIS T((ME UNDER J f.
CONTROC OF THE REED-WATER dONTR00 SYSTEM AND REACNED INE LEVEL 0F THE STE Ad lif.NE. AT ABOUT 1418 HOURS PR[fM ARY TM W X
... TWO ~
, ,, C*w :b'b dONTAINMENT P.RESSURE RE ACHED IO PSIG AND ACC REQUIRED ggl1;2(
ecd 3 SYSTEMS STARIEDe CONTAINMENT PRESSUsE CONT [fNUED TO RISE, PE AKED AI 50 PSIG, AND BEGAN TD DROPe TORUS SPRAY,S @k,'f.$ .
WERE PCAdED [IlN SER9JCE AI THIS i'OlNT. TO ASSTSI IN THE DMM NE AT. REdOVAC Fs0M THE dONTAINdENT. AT 5000 H6URS CONTA[ljNMENT PRESSURE WAS 3o8 PS.'ijG AND DECsE ASING. VENffNG i 0F T.HE PRIMAR,Y dONT,A[IlNMENT HAS NOI BEEN EdPLlotED. SER9A1CANdE WU 1201 (R 5 69)
)