05000302/LER-1988-001, :on 880107 & 9,automatic Actuation of Emergency Feedwater Occurred Due to Control Channels Detecting Low Level in Steam Generator B.Caused by Sluggish Operation of Startup Feedwater Control Valve
| ML20149E848 | |
| Person / Time | |
|---|---|
| Site: | Crystal River  |
| Issue date: | 02/08/1988 |
| From: | Moffatt L, Eric Simpson FLORIDA POWER CORP. |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| References | |
| 3F0288-07, 3F288-7, LER-88-001, LER-88-1, NUDOCS 8802110302 | |
| Download: ML20149E848 (5) | |
| Event date: | |
|---|---|
| Report date: | |
| 3021988001R00 - NRC Website | |
text
.
g. me w s wetaAm as tutaroav c=es.o
an.aoveo o.e =o sino.4io.
LICENSEE EVENT REPORT (LER)
..ciury He oocarr = amen m
... =
CRYSTAL PlVER UNIT 3 o i s I o I o I o i 31012 1 lod 014
" ' ' ' ' " Improper Positioning of Turbine Bypass Valves and Sluggish Response of a Startup Feedwater Control Valve Results in a Low Level in "B" St<>am Generator and Subsequent EP Actuation Eveer? OATS im LES musstem te eePo#7 DAT1171 oft.e A pecauties esprotves see woJctw 047 vtaa vtaa 88L* $ 'h Q@
e.cas7w oav vsAm F a'cauv v 'eames occatt a.unseeaiss N/A oistotago, ; ;
0l1 0l7 8 '8 8l8 0l0l1 0l0 0( 2 0l8 8l 8 N/A o,g,o,o,o, r=
apone sue rno suasuant to two a.aviawe e, ie e,= t <c,
, nu
"*M
- 3
..se.,
n, y
.n,,,,,,,,
mn.,
y m.=<.unie seassem es ni.x H nnw n.
0l 0l 0 m aeanum seanwm
.nwmum gay _,.
3e asse.wiwee es nweno esnwww nas
- ma4, m egetHg HW1 se 73enH3HSi et 731s4W*msHet 30 aseteH1H,6 go.F3esugHae se F3eefeMaa tac 8sals0 coertACT poa ft.as tga et31 n6p o e u..
L. W. MOFFA'IT, NUCLEAR SAFETY SUPERVISOR 9 IO I4 7 19 l 5 l 1614 (8 [6 conspttT1 okt Laa post eacie -- _ Mary pa, Lune Descasses em fiese sepost? f te "O
'[d'N8
Cause
tvttau Mmess?
CAV80 systtu Conspositief
'82]#'O Pit La ' ;p I l l l l l l
) l _j _
l l,
t I
I ! I f 1 I I
I f i 1 l I suppteasterfA4 me. IIIPectte He tso=Th Cav vtan q,es no.
onecreo s<,ourne, ca res 7o sue
..o=
v amer **cv w i.m
. a
- : n -- :
n.
6 l
On January 7,1988, Crystal River Unit 3 was in the Hot Stantef corrlition (Mode 3).
An autcmatic actuation of Emergency Feedwater occurred due to 2 out of 4 EFIC channels detecting a low level in the "B" sicam generator. A similar actuation also eu,wred on January 9,1988. In both cases the Emergency Feedvater Systan responded as designcd to the conditions in the "B" steam generator.
'Ihere were several factors contril:utirq to the event of January 7, includirq a slight steam generator pressure increase and sluggish control response of the j
"B" startup fecdwater control valve. 'Ihe January 9 ovent iras due to an l
imprtper integral setting in a module in the "B" startup control valve control l
circuit.
j I
In both cases the EF actuations were reset once nonnal level control was established with a main feedvater purp. As a result of both actuations, work has been parforrod on the "B" startup footeater control valve and its control circuit. In ackiition, new procedures are being written for preventive mintenance of the Integrated Control Systcen, arri operators will be remirdcd of the increased monitoring required as a result of having nultiple cx:ntrol stations in manual.
u.
eewnowo* %0 dl PDR ADOCK 050 c
5 1
1
a C,
NAC P.rm 364A U s NUCLEAlt KEQuLAfoAY CouurssioN LICENSEE EVENT REPORT (LER) TEXT CONTINUATION Aa*=ovto ove %o v$ce E mptats s'3195 F AC8Lif v NAME di DOCA ET NUMelm (37 LE A NUMSER 141 PAGE (31 vgat S E QVI%T' Ab
- E W 4'O %
CRYSTAL RIVER UNIT 3 0 (51010 l0 l 3l 0l 2 8l 8 0l0l1 0l0 0l2 0F 0 14 rw ru
.m.. m :. :anc r.,. nw,wn EVmP MSGTPPIW:
On JanJary 7, 1988, Crystal River Unit 3 (G-3) was in the Hot Standby condition (Mode 3).
We cperating main feedwater (W) punp [SJ,P) was beity cx>ntrolled in anual with a fixed discharge pressure of approximately 1100 PSIG. %e turbine bypass valves (TBV's) [JI,V] were also in marual, due to the performarre of a rod drop timiry surveillan
'Ihe steam generators (SG's)
[AB,SG] were beirq controlled at the low level limit by using the startup (S/U) feedwater control valve [JB,ITN] in autmatic.
In addition, a surveillance precedure for checking the main steam safety valve (MSSV) [SB,RV) relief setpoints was in progress, which required the operators to mintain steam header pressure at a constant value.
At approximately 0400, an autcznatic actuation of the Emergency Feedwater System (EF) [BA] occurred when 2 out of the 4 Emergency Feedwater Initiation ard Control (EFIC) [BA,G A) channels detected a l w level condition in the "B" steam generator.
Due to the logic configuration of the EFIC system, only the steam-driven EF ptmp [BA,P) started ard delivered flow to both steam generators.
Only 2 of 4 EFIC channels detected a low level due to slight calibration differences in the level transmitters [LT]. Also, with an EF ptmp ard a W ptmp both deliverity flow to the steam generators, the level was very rapidly restored to the normal level for the required conditions.
On the main Feedwater System was verified to be controllirg level, the EF ptmp was secured.
On January 9,
1988, G-3 was in the Startup Mode (Made 2).
Moderator tenperature coefficient testing had been cca:pleted at approximately 1000 as part of zero power physics testiry. Both the primry and secondary systems had been stabilized followirg the test.
'Ibe cperating main W punp was beirg controlled in manual, and the TBV's were beirg controlled in autcmatic.
At approximately 1110, an EFIC actuation occurred due to a low level condition in the "B" steam generator.
Once again only 2 of 4 EFIC channels detected the low level and the steam-driven EF pmp started as designed.
%e "B" S/U feedwater control valve did not respord to its demarded control signal.
A cx>ntrol rocm cperator was in the process of opening the low load feedwater control valve to try to regain level when the EFIC actuation occurred.
CAUSE
'Ibere were several factors khich contributed to the January 7 event.
'Ihe "B" SG level was oscillating nearly 15 inches prior to and followiry the EF l
actuation.
'Ihis was apparently a result of sluggish operation of the "B" S/U l
feedwater control valve, which was thought at the time to be due to sticking of the valve stem.
We dcurreard level oscillations were only approximately 4 inctes above the level for EF actuation. m is margin is extremely small, khich l
increased the chance of actuating Emergency Fecdwater due to a slight SG level l
I i
).g,3.m.
i hmC Pere Je4A U S huCLE A2 EEQut ATOJiv COMMIS$10*s LICENSEE EVENT REPORT (LER) TEXT CONTINUATION a** movie eve o mo_w f arings gi3t gg P ACILIT V N A.sl (U DOC K E T NUMG E R 421 LtR NUMSE R tel P AGE 131 l
" W.' '
'T3 "aa CRYSTAL RIVER UNIT 3 o ls jo lo lo l3l0l2 8l 8 0l0l1 0 l0 0l3 OF 0 l4 rm _. =. :_ : ec r anw.nm A slight level transient did occur as the result of an increase in SG pressure, possibly due to the TBV beirg positioned too far closed.
'Ihis SG pressure increase would not normally have resulted in a transient, because the m punp would have increased its speed to maintain a constant differential pressure across the S/U feedwater control valve.
Hwever, the m pu::p was in manual, therefore the pmp speed remained constant.
A slight urderfeed occurred, ard the level oscillation tes low enough for 2 of the 4 EFIC channels to detect a low SG level in "B" steam generator.
The event of January 9 was similar in nature. The "B" SG level was fluctuating as nuch as 20 inches prior to ard followiry the EF actuation. One of the level oscillations downward was low enough for 2 of the 4 EFIC channels to detect a low SG level condition aid initiate emergency feedwater.
The cause of the SG level oscillations '1re determined to be due to an inproper integral setting in an Integrated Control System (ICS) module in the "B" S/U N control valve circuit.
EVFNP ANMHSIS:
In both cases, the Emergency Feedwater System responded as designed to the conditions in the steam perators.
In both cases, only "C" and "D" SFIC channels detected the low W c1 condition for longer than the required 2 seco:d time delay.
The reason that channels "A" and "B" of EFIC did ret detect the low level condition is most likely due to slight calibration differences in the level transmitters for these channe]s.
Once an emergency fecdwater puap was started, the SG level was rapidly increased, thus precludirq the "A" ard "B" channels frca detecting a low SG level.
Tae effect of the Emergency Feedwater actuations on primary system parameters was minimal.
A decrease in Reactor Cbolant System (RCS) [AB) pressure, cold leg temperature (TC) ard pressurizer (PZR] level was noted, yet no paramater decreased belcu any applicable limits for the plant corditions.
These decreases are eJqrcted with an emergency feedwater actuation, due to injection of colder water at higher flow rates into the steam generators.
Based on the above, the Safety significance of these events is minimal, wto<wnVE ACPIN:
Following the EF actuation of January 7, the EF system was reset once main fecdwater was verified to be operating ard pIrperly controlling.
Operators halted the testirq of the MSSV's so that there wtuld be less interference in ccrpletirg the rod drop timing test. Once the red drcp test was satisfactorily ccepleted, testirq of the MSSV's was then resumed, and subsequently ccmpletcd.
Even thotgh the operators on duty at the time of this event knew enough to place the turbine bypass valves in manual for the rod drop test, the proccdure did rot mention this. The surveillance procedure will be revised to add this requirement.
In addition, this charge will also ackl a caution to closely ronitor steam header pressure with these valves in manual, to prevent an g.o.,
]
g MC pere 3esA U S NUCLEAR RigyLATORY COMwitS404 LICENSEE EVENT REPORT (LER) TEXT CONTINUATION amovio cw. No mo_eio.
Supiagg gai gg P ACILif v NAt#t Hi DOCILif NUna0E R E2p LER NUMSIR 441 pAgg (33 "t!M.
"'rJ:
"*=
CRYSTAL RIVER UNIT 3 o ls lo lo lo'l3 l 0l 2 8l8 0j 0l1 Oj 0 0l 4 or 0 l4 i
rnn a -
. e.,: - :Nncr man <m inadvertent SG pressure increase and 94w=_=1t underfeed of the steam generator. In addition, an Operations Study Book entry will be made whicts will remind Operators of the increased monitoring required as a result of i
having nultiple ocntrol stations in manual.
1 Also following the January 7 event, the valve sta of the "B" startup control valve was lubricated to prevent sticking.
Even though the SG level oscillations still existed, no further trmbleshooting was performed because the S/U control valve itself was satisfactorily stroked.
In Ausgat, further evaluation of this proble should have been required, since the work performed did not correct the proble with SG level oscillations.
I Following the January 9 event, the pilot valve for the "B"
startup control valve was cleaned and adjusted.
However, this did not change the actual valve i
position versus the demanded irpzt. The valve was controlling properly, as it had before, but level oscillaticms were still as much as 20 inches.
The integral rate for a module in the valve control cirmit was then adjusted fran 5 repeats per minute (RRf) to a value of 1 RIM.
This integral rate change danpened the level oscillation to approximately 4 inches, whickt prevented the SG 3evel fran oscillating close to or below the level for Emergency Feadwater initiation.
As a result of those events, it is clear that additional actions are necessary.
(
Probles snocuntered with the ICS are often difficult to work on because it usually requires tuning of the systan while it is controlling the plant. There have been several new prrminnes generated, however, which would identify and correct many ICS problans prior to their d1scovery during normal system cperations. These kvciidures are under the category of Preventive Maintenance Prnwhwes, an:1 are currently in the review Dro ss.
FREVIOLE SIMUN1_]M2Eh The initiation of energency feedwater en low SG level has occurred many times, most notably during startup fran the Refuel V outage, which was the first one with the newly installed EFIC systen. These events were reported in IER 85-12.
i 8 70ew 3644
['
O 0, 000. )OO*
60 e
z.s Florida Power C 0R P O R A TIO N February 8,1988 3F0288-07 U. S. Nuclear Regulatory h i m ion Attention: Document Control Desk Washington, D. C. 20555 Subject: Crystal River Unit 3 Docket No. 50-302 Operating License Ib. DIR-72 Li nsee Event Report Ib. 88-01-00
Dear Sir:
Enclosed is Licensee Event Report (IER) 88-01-00 kilich is stinitted in accordance with 10 CER 50.73.
Should there be any questions, please contact this office.
Sincerely, 6% bd 4 g E. C. Sirpson, Director Nuclear Operations Site Support WIR: mag Enclosure xc:
Dr. J. Nelson Grace Regional Administrator, Region II Mr. T. F. Stetka Senior Resident Inspector
$6%'
f I
A Florida Progress Company