ML19282D128
| ML19282D128 | |
| Person / Time | |
|---|---|
| Issue date: | 04/06/1979 |
| From: | Ornstein H NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| To: | Eisenhut D Office of Nuclear Reactor Regulation |
| References | |
| FOIA-79-98 NUDOCS 7905140252 | |
| Download: ML19282D128 (19) | |
Text
s
'h g
8 A
UNITED STATES
.; p 1
NUCLEAR REGULATORY COMMISSION
- .3 x )., $
WASHINGTON, D. C. 20555 s., *..../
((
MEMORANDUM T0:
D. Eisenhut, Deputy Director, Division of Operating Reactors FROM:
H. Ornstein, Technical Specialist, TA EDO
SUBJECT:
FEEDWATER EVENTS AT B&W PLANTS Enclosed is a typewritten copy of the material I provided you on April 4, 1979.
A1_so included are copies of relevant B&W and NRC memoranda.
fWWYh 3
H. L. Ornstein Technical Specialist, TA EDO
Enclosures:
As stated above cc:
P. Check J. Watt j,7,,
OR Q h FolA Rot 5 HAM 774 3 79051409 5?-
9 w
h e-e e e N***e-
- +
51 a
THI-2 3/29/78 (Zero Power Physics Testing 3 RC Pumps Operating)
RCS depressurization Vital bus 2-IV de-energized - caused opening of an electromagnetic relief val ve.
Reactortripped(duetoindicationthatAloopRCPumpswereoff-whi/IE B loop RC pumps were on - actually 1 "A" loop pump was not in service at beginning of transient and failure of the vital bus 2-IV caused the RPS to sense that the remaining A loop pump was lost while both B loop RC pumps were operating). The re' actor depressurized, the operators closed the RCS letdown isolation valve. Temperature compensated pressurizer level indication was lost, as was RCS pressure indication powered from the Vital Bus 2-IV.
The operators did not know why the depressurization was occurring - as they had no position indication of the electro-magnetic relief valve.
At 1 min 53 see SFAS actuation for safety injection initiated.HPI pumps took suction from the BWST and the NA0H tank - at 2 min 23 see The Safety Infection Signal Bypassed.At 4 min 13 see the vital bus was reenergized through its alternate source, thereby closing the electromagnetic relief valve and returning all instrumentation to service. The depressurization ended at a pressure of 1173 psig.
f i
e
~5 e
5.t 1.--
tt TMI-2 4/23/78 (30" power - 3 RC Pumps Operating)
Excessive RCS cooldown and depressurization Reactor tripped due to a " Noise spike on NI8 power range detector - The reactor tripped because RPS Channel C was already in the tripped state...
due to the inoperability of NI 7."
There was a turbine trip causing pressure increases in the OTSG's. The main steam relief valves on the SG's lifted and did not immediately reseat properly (finally reseated after 2-4 minutes with SG pressures at 550 and 600 psig).
I The operator took the proper immediate action in manually cutting back feedwater demand, shutting the letdown isolation valve, starting a second makeup pump, and openin j operating makeup pumps.g the high pressure injection valves on the side of the The operator failed to initially recognize that the J feed pump was in manual and did not run the feed pump speed back until approximately 1 minute and 20 seconds had elapsed.
The Integrated Control of the feedwater valves had not yet been initially tuned at the time of the event.
Integral vice proportional control was the dominating signal of the feedwater valves and although the valves responded in the proper direction, they responded much slower than the traditionally g.#
expected response.
Thus, the feedwater valves slowly going shut, rapidly 9g decreasing steam generator pressure and a constant feed pump speed, too much water was fed into the steam generators.
a The safety valves failing to reseat at the proper pressure coupled with over-
~~'D f the reactor coolant system. feeding the steam generators caused a rapid depr The reactor coolant temperature varied from A
5830F to 4640F in 3 minutes.
The RCS shrinkage from the cooldown caused the pressurizer volume to drop below the minimum indicated level range approximately o
one minute after the reactor trip.
g Due to the rapid depressurization of the y k By present design this injected NaOH into the reactor cool gg the high pressure injection lines.
} Pressurizer level was restored two minutes into the event as a result of safety injection, the Turbine Bypass valve going shut and some of the B side Main Steam Relief Valves going shut.
Feedwater latch occurred 2 minutes into the event and tenninated feedwater flow to the steam generators.
Feedwater latch was the key event in terminating the transient.
.W.
e g
~.
i s
~
s
~;L...
TMI-2 11-7-78 (92% power)
LER 78-65-99X Reactor Trip and Safety Injection During a power runback due to loss of 1 feedwater pump.
On November 7, 1978, TMI-2 experienced a reactor trip during a power runback from 92% rated thermal power.
Prior to the reactor trip, testing per TP 800/05, Reactivity Coefficients at Power, was in progress. All operatfng garameters were normal except for RC Tave which had been elevated to 588 (6 F above normal) for temperature coefficient measurement.
At 0523:37, a heater drain tank low level alarm was received. This automatically tripped operating Heater Drain Pumps HDPlA and 1B which g
normally supply approximately 30% of the total feedwater flow to the suction
(
of the feedwater pumps.
The feedwater pumps tried to meet the increased feedwater demand, however, Condensate Booster Pump COP 2C tripped on low h
suction pressure. This autcmatically tripped Feedwater Pump FWP13.
qs..-
The Integrated Control System (ICS) began a power runback to 55% rated thermal power based on the loss of one feedwater pump. However, due to the elevated Reactor Coolant System (RCS) temperature required by the testing in progress, k
the reactor tripped at 64% power. This trip occurred prior te completion of the power runback, as all four Reactor Protection System (RPS) channels y
received a variable temperature pressure trip signal. At this point, the operator secured the letdown flow (closed Makeup Valve MUV376). A second makeup pump was then started prior to the safety injection.
g RCS pressure continued to decrease and safety injection was automatically initiated at 1640 psig thus limiting the pressure decreasa to 1550 psig at 25 seconds after the reactor trip. The decreased RCS volume due to the cooldown and depressurization caused the pressurizer volume to decrease below :ero indicated level for approximately 30 seconds. However, calculations show that the pressurizer was not emptied during the transient. Approximately 2 g minutes after the reactor trip, RCS pressure increased above 1600 psig.
ee*=
a &
s 4
TMI-2 LER 78-69-99X 12/2/78 (22% power)
Reactor Trip with Safety Injection due to pinned open main Feedwater regulating valve.
On December 2, 1978, TMI-2 experienced a reactor trip frem 22%
rated thermal power, while switching from the startup to the Main Feedwater Regulating Valves.
Prior to the reactgr trip, all operating parameters were normal except A
for RC Tave of 584 F.
Tave was higher than normal due to Feedwater Heaters Q
being placed in service.
hg Due to the changing FW flow, the startup feedwater valves (FW-V25A/B) reached 80% open. As the main feedwater block valves (FW-V10A/B) opened, FW valve d/p decreased to zero, prompting the operator to increase feedwater pump k
speed.
It was later determined that the Main Feedwater Regulating Valves L
(FW-V30A/B) were full open by manual hand wheel with Instrument Air isolated.
The increased feedwater flow resulted in a rapid RCS cooldown. At this point the operator secured letdown flow (closed MU-V376), started a second makeup oump, reduced feedpump speed, and closed the main feedwater block valves.
The overfeeding of the OTSG's resulted in the reactor trip on low RCS pressure, followed by Safety Injection 215 minutes later.
Due to the above operator action RCS pressure recovered to above the Safety Injection setpoint within 17 seconds.
e m.,
e e
a emei
9 Rancho Seco 78-1 3/20/78 Excessive cooldown transient 70% power (From I&E report 50-312/78-03) lost non-nuclear instruments - including SG and pressurizer levels - and all RCS temperatureyJpss of RCS hot leg temp input to the ICS caused termination of feedwater flow.
Reduced heat removal in steam generators causedRCStemperatureandpressuretoincrease. The reactor tripped on high RCS pressure followed by a turbine trip.
The secondary sides of both steam generators emptied due to operation of condenser bypass valves, atmospheric dump valves and auxiliary steam loads.
Pressurizer level was maintained (using computer indication) by manual operation of a high pressure injection pump.
"A" ste m generator level control (actually lost at time zero - but the channel drifted slowly downward - while "B" SG channel drifted slowly upward) initiated emergency feedwater injection (the turbine driven auxiliary feedwater pump had started on loss of feed-water flow)
RCS cooldown started as a result of emergency feedwater low to A steam generator and possibly also due to main feedwater pum smanuallyoperated).
Decreasing RCS pressure (1600 psig) actuated all safeguards pumps and the motor driven auxiliary feedwater pump.
Full auxiliary feedwater was initiated
=e-to both steam generators.* RCS hit a minimum of 1425"psig which was then increased and maintained at 2000 psig by manual control of an HPI pump.
Restoration of the NNI restored all lost indications and controls - operating personnel secured the auxiliary feedwater pumps and started RCS pressure reduction via pressurizer spray.
MM
- maximum flow on both aux feedwater pumpg automatically provided to both steam generators.At Davis Besse 1 and Crystal River 3 safety features Actuator signal does not initiate maximum AFW flow rate to both steam generators.
~
'A Rancho Seco LER 79-01 1/5/79 (100% power) Loss of ICS - Excessive cooldown re Short on ICS - loss of logic power sm3the feedwater valves back to the 50% position - caused RCS pressure to increase resulting jn a high pressure trip./69 red /?CS depressurization to 1600 psig actuatd HPI and auxiliary feedwater.
ICS was restored after 5 minutes, and feedwater flow increased.
The operator then terminated most of the feedwater flow - 2 minutes later the main feedwater pumps were tripped - thereby allowing aux feedwater to supply the OTSG's.
DuringthetransientthegbTSGwasfilledtothetopoftheoperatingrange, and it stayed at that level for 10-15 minutes. SMUD believes that the excessive feedwater to the "B" OTSG from the AFS was "the single most significant cause of the excessive cooldown rate."
No info on pressurizer level - SMUD will review (by 4/30/79) an evaluation of the necessity of aux feed on safety features actuation.
B&W is reviewing the transient and will forward recommendations to SMUD for corrective action.
~ _**JT:
G
'L'~ - ;_
... _ _.._._____.____.__'"_..___ZZ"
~ " ~
s NP 32-77-16 Davis-Besse 1 9/2f/77 9% power Loss of RCS pressure duet to failure of pressurizer power operated relief valve 9% power -bypass system operational. spurious signal resulted in closure of feedwater control valves to 1 steam generator.Upon a low-low steam generator level. d.]' main steam and feedwater isolation valves closed and 2 aux feedwater pumps started.
A problem (don't know what) developed in 1 AFW pump, and it was shutdown by the operator. Tne plant operated at 9% power on 1 SG with 1 aux FW pump.
Feedback to the RCS resulted in increasing pressurizer level and pressure. The operator indicated a manual scram. (T=1 min. 47 sec)
RCS pressure increased to the setpoint of the pressurizer relief valve.
It cycled 9 times and then stuck open. When pressure dropped to 1600 psig ECCS was initiated (2 min 51 sec).
Full high pressure injection flow was established, and started to raise pressurizer level. AT T= 6 min 14 sec.
the operator stopped the high pressure injection pumps (The operators had bekn
" heavily involved t,efore this time in regaining seal injection flow to the reactor coolant pumps which had been stopped by the SFAS actuation." By T=5 min 20 see the appropriate SFAS signals had been overridden and the normal flows restored to the seals of the pumps).
RCS pressure continued to decrease until saturation pressure was reached, steam formed in the RCS (T=8 min) - causing an upsurge of water into the pressurizer and the pressurizer went off scale.
During the level increase the operator saw the RCS avg temperature and pressurizer level increase - he then stopped one reactor coolant pump on each loop (T=9 min) to reduce the heat input to the RCS.
At approximately T = 21 min., it was determined that the power relief valve was remaining open and the block valve was closed, isolating the power relief valve on the pressurizer and stopping the venting of the reactor coolant system to the quench tank. At T = 31 min., pressurizer level came back on scale. At T = 41 min. the operator started a second makeup pump to try and stop the pressurizer level decrease. This additional cold water started the 4
reactor coolant system on a slow decreasing temperature transient. At k
T = 43 min., pressurizer level reached the low level interlock and cut off the pressurizer heaters. At T = 49 min. the operator started a high Q.
pressure injection pump to try and stop the decreasing pressurizer level.
h.
With pressurizer level well on its way to recovering, the operator stopped b
the high pressure injection pump (T = 53 min. 24 sec.). At T = 57 min.
4 he restored reactor coolant makeup flow to normal.
This stopped the slow k
decreasing reacter coolant temperature transient which started at T = 41 min.
g All plant parameters were now fully under centrol and the plant was brought to a steady state condition, and a normal plant cooldown started.
,.u.:.
.eop
t.
s r;.
y T
1 0 Cbh<X>
'11/29/78 - Unit 3 One reactor building pressure sensor channel "
vas intentionally tripped, spurious trip on another channel 0 4 psi, caused E.S. actuation, operators took manual centrol and secured E.S. cquipoent.
12/14/78 - Unit 1 Tecdvs.cr pu=tps tripped, steam generators went dry,.EP Inj ection actuated, PORV's _ operated satisf actory.
C 7/6/73 - Unit 1 E.S. actuation due to operator error, EPI inj ection r* * *:ed PZR l'avel *Mve level instrument range, but premmure remained-less than 2219 esie.;;perator action properly avertec W ing 'at preuzurizer re11ef valves.
11/22/74 - Unit 2 ' During, cooldovn subsequent to a resctor trip resulting from a RCP meal leak the core flood tanks were depressuriza by bleeding, nitrogen to the quench tank.
This caused the quench tank rupture disk' to burst, and the resultant jet severed the i= pulse, line on the prexsurir.er level instru=cntation and as:sged pressurizer insulstion. Nor=sily, CT tank bleed would be to the vaste nas filter or vent header.
However, the valves required to operated to accomplish this were inaccessible due to the sesi leak.
AO-270/74-2.
5/30/74 - Unit 2 Operator error resulted in EPI injection into RCS when vs1vcs were creencously opened (no ES actuation signal).
Reactor trio resuirad a,
eressurizer'hi'gh level.
NOTE:
They early found problems with the PORV block valves and by October 1976 all three units had replaced the valve with an Impr.oved design.
They h$ve installed a deflector pist.e bet.veen the quench tank and the pressurizer to keep stean/ vater mixture fron hitting the pr essurizer.
Ccmpleted as follova Unit 1-5/76, Unit 2 - 6/77 and Unit 3 - 10/.76 t
e Y
e I
'O e
y 100% power Full Load "urbine Trip Test + Failure of 1 trEn of Aux Feedwater Turbine Trip, main feedwater pump trip, automatic actuation of emergency feedwater pump. A MOVrequired for feeding Sa "A" steam generator through the AF4 nozzles failed to open. Feedwater t steam generator continued satisfactorily.
Subsequent operator action restored feedwater to both steam generators.
=
ee.T--
..ua
.y.....
Oconee 3 6/13/75 A0 287/75-7 Excessive Cooldown Rate Routine maintenance shutdown 100% -1>l5%.h# nile at 15 % power there was a mismatch between power gen 0 ration (ll5 MW)and unit load demand.
(65 FW).
ICS attemptElto match load with power. The main steam bypass valves opened, and then closed when the main steam pressure decreased, feedwater flow and steam generator level oscillated, as did RCS temperature and pressure.
The power operated relief valve opened at 2255 psi but failed to close when the pressure dropped.
Decreasing RCS pressure caused a reactor trip and HPI actuation. The operator closed the block valve after the reactor trip to terminate depressurization. The block valve was later reopened because of rising pressurizer level, and was again closed when the pressure dropped to 800 psi - terminating the transient.
9 4
m m y
-m e.m w~
w
12/14/78 Oconee I-R0-269/78-77 98% power A short ca 2 sed ICS T ave recorder error ICS withdrew control erroneously-reactor trip on high temp /prc_is.
Both Normal FW pumps tripped on high discharge pressure.
Emergency FW pump started then stopped when normal FW pumps were reset and started.
2 hrs, later OTSG Tevels dropped to 6 and 0 inches (VS 110 inches nonnal)
SCAlevelwasrestoredwithin3hrswhereasittookabout8hoursuntilSb wds filled through the emergency FW header.
Apparently malfunctioning ofI valves in the normal and emergency feedwater paths led to the long fill time for SCB.
The HPI was actuated on low RCS pressure during the event - however from the existing information it is not clear when this occurred.
C
4 Crystal River 3 3/2/77 40% power Loss of At bus-excessive cooldown rate "B" vital AC bus was lost due to a failed output diode in the inverter.
Power was lost to ICS causing reactor crip, turbine trip and opening of the steam dump valves (50t). Main FW pumps tripped (due to loss of vacuum) and SG feed continued from the emergency FW pumps.
RCS temperature dropped 1640F within 15 minutes.
Plant normalization was begun imediately, with vacuum restored within 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />.
-g -
Ciir-T. ".
"d' -
4
.s..
Babcock &Wilcox a... s,..m c.: =
_=
P.O. 5:n 12C. Lyr:.:.;rg. va. 2:[
Te:,: : :: ;D ) 3!:.5111
- +
g:g:.2.
Augus t 9, 19_e g e....-4 c. g..s
,.,y e.
q j_3-:t._e
- e... _~
c-
=
1 S ? n.k/295 k+
a v:
ra.
- r..
r.
1.8 r.'. 3.
1.i.. _.e.., c. a. 4,...
c.g e J. g e g e...
- 7. -
. ~.
Esvis-lesse ;uel e.: ?c.er Static:
r.
f*
ec o-.; e '.h S *. a*.
.:.-...- I2
,+
Cai Et:::r, Ohio L3LL9 i
..p
.e... -
e..,?..c.:. _ a. g c c n 1 c..
- n. _ g.s e..
- i..
3 L
i-
'e z - Te - / :
[
3 C-W - '- 20, 1973, p.a.n:he Seco e::perienced a severe the =d transi'e:- 1: itis. e!
hy.he 1 css of electrical s.c er to a substa:-ial pcrtic: ef -he :~ct-:: :le e -
E,:
Ins t:.:=e:.atien (2.I ). Le less of power directly cau.s ed,the ?. css of Centr:1 i,
?. e= i= dies.tien of =z.y pier.t pare =eters, the les s cf input of -he s e para-L
=ete s to the plant ec=puter and err:necus input sig=als ( idrz.ge, :ero, I
cthervise incorrect) to the Integrated C :trel.Syste_ (ICS);
r.
,_ c u
3;.
i ne plast respense vas =ct the u.sual tra.sie:t i= that the ICS resp::ied to g:
the e.re= ecus i:;ut sig a.ls rather than ac.ual plan. cenditic:s, a=d : esulted t
_# - a %.. a '..-.*- '. e c.# c. S3 s *. e= (.2*5 ) *.. # p c _ 1 '...'.. -. s s" e.
" u, s. s~.. -...
'p'
.. a c, _ ~..., _n
.... e _. e.,:3 s. e,.. a.' s *. o "...
T. C.. -.... '. -.. * "..... - ' '
4 r.
3 c.~ _' d.v.
o.' '.5.
V. _S.
.'_ a.. c. e a.c. s b a e... _ - di."..# -"_'.v.
.'..#.e_'.'....,
.s.
. s '..-.u.s o.' s
. c.'.k.. e e1 as*.,. a z=... s a. d. #.. -.. _ ' ' _ '., ^..
[
tecau.se of the erre: ecus indicatic=s is the Centrol 3===.
I' r
_4... s '. #., a'.i c.. c ' '.'.. e v...*. s
.'e.' ' c.-#.., ***s.'.ess o.'
.c '...s c".'.'a
..*.r-
'cr a cicse Icek at operater training and e=ergene/ operating preced: es for i-any less of.liI po er (er pcrtic: thereof).
Le fellevi:s rec _endttic:s L
- s. e cade to assist you.r'ste.fr in a revitie of trai ing r..d precedures to e.sst. q proper c; err.ter actics for eve:ts of this =z.ture.
1.
C;eraters ishou.1d be tre.ined to recc., ize a less of 7:ver:to all er a a3c : y o.,
w.,
3m-.- -(e.e.,.
4,cic_..c.s
.e.43_ _. o __ d z.-. e,., a...- _.=.. #.. c-
+.
u...
. _., _.. _.,... a. s.... o a '
- a a. e _# _ s...,. _. s.. '
=-s.-. '.... s _... s,.. =. ).
E
". _-. _' s s o.'. c..
.* s e _,.*.1s ' e d." e.-.. a*........ a.. '....' _= ' ' " -. c.' _<.. ; c -.
.=.
s 3
=
v
ar. a..- e"_.
_'v c.....' #-.--c-....
4.s.
c_.
- c.._._, s<e..,,
s
.,a.
..._4_,s.,c u..ses.
e
. e
-s::
7 Mem h e OM N m
e ge -
- Me M.
=
w Nd'
'. ~
'acco. 4L P.HCo%
620-00;*.
Augn:. 5, 197.,8 7 ace 2 r+
2 *.
Give: the.t the ope. rater ere dete =ine tha.t ele:trical pr.er has bee:
icst to a or pa t of the NNI, he should h ev the ic:stics of the pc.er supply breakers, and have e. procedure ave.ilsle to q@*
- y re-g e.i: pover.
3.
If the fcult can=ct be cleared (i.e. the tres.ers to the pr.er supplies recpen), the operator should have a' list of a.ite=r.te i=stre=entatien available to hi=, e.nd he should be thercughly tre.ined i 1.s use.
Ix-r.=ples e. e:
a.
ESTAS pe.nels b.
7.75 pe.nel's ICI (Esse tial Centrels e.:d Instrc=e:tatics) c..
d.
53CI (Safety Related Centrols and I:strs==: stic ')
e.
3e=:te shutde : pe.nels f.
Loca.1 gages g.
?last ec=puter h.
Zecc7.1:ing that' no pre:ed: e ca: cover a.11 pessible ec=bi::.-i=:s cf ICTI
- /
'=4 ' - s, the operator's respcase should be heyed to certain va.riab' es,
t
.If the opers.ter realizes that he has as instru=entatie: probl.e= (as ep-1
- sed te a LOCA or stez= li=e bred., fer ext =ple), he ca li=it the tra.sie:t by ce= trolling a fev critical va-iables1
\\~
Presstri:er level (da E?I er n==d Ms.keup Pu=ps 1 a.
b.
RCS pressee (via Presseizer her.te s, spray, I/M relief valves, etc. )
c.
Stem C-enerater level (via feed flev, fe e dva.t e r va.1ve s, et c. )
d.
Stes: Generator pressure (via tubine byna.es syste=}.
T e pressu-izer level and 3CS pressee. assee that the Zet:.cr 'Cocir.:t Syste= is filled; the Stez= G.ecerater level and pressure assure ade-::.:e dect.y heat re=cval.
Attach =ents 1 end 2 are provided to give a brief descriptic: of the events fellevi=g this less of NNI pc.er at Easche Seco.
As c r.: be see by this tre.nsient, "prc=pt precise operater a: tics e=d the ability to reccg ize a 1 css of ICi! pcrer t.re critical factors in li=iti g the severity of a tri.=s-1e:: sue.1 as this.
If.you have any que.,nier,s or e _e ts, ple ase sivise.
Yours trely, Iva: D. Gree:
=
g-Site Opera. tic:s Mr.ager
--...._3, e::'.
See atta:hed sheet.
. ~.,
- * ~ ~ '
V-
~
, 3~76' s. !
3 3 =cosevco moio
+w../,1
.. m.... _,,
..,33
, =.
=
January 8, 1979 t=
Docke: No. 50-500/501 50-329/330 MEMORANDUM TOR:
J. F. Streeter, Chief, Nuclear Supper: Section 1 TROM-J. S. Cresvell, Reactor Inspec:c SU3 JECT:
CONVEYING NEW INFORMATION TO LICE:: SINS 3 CARDS -
DAVIS-3 ESSE UNITS 2 6 3 AND MIDLASD UNITS 1 & 2 During the course of =y inspections at Davis-Eesse, car:ain issues have co=e
- c ny a::en:icn which I a= submitting f or consideratien f:: fervarding to the A:ctic Safety and Licensing 3 card which has proceedings pencing fer the aforenen:1ened facili:ies.
This sub=ittal is made pursuan: :o Regienal
?re:edure 1530A (Nov enb er 16, 1978), step 3 and infer =a:i:n sa; plied to
=,e per step 1.
The issues for considera:icn are:
1.
During a recent inspection a: Davis-3 esse Uni: 1 infer:2 tion has been a::ained which indicates that at certain condi:icns of reactor
-E.;..y coolan: visecsi:y (as a function of tenpera:ure) cere lif:ing may occur.
The licensee infor=ed the inspector tha: this issue involves other L&,W facilities. The Davis-Besse FSAR sta:es in Sectien 4.4.2.7 :
The hydraulic force en the fuel asse=bly receiving :he =es:
flow is shown as a function of syste flow in 71sure.4-39.
Additional forces acting en :he fuel assembly are :he asse=bly weight and a hold devn spring force, which resul:ed in a net devnvard force at all ti=es during ncr=al sta: ion opera: ion.
The licensee s:ates that there.is a 500 F interleck f er the s:arting of the fourth reactor cociant pu=p.
Ecuever, no Technical Specifi-ca:icn requires that the pu=p be started a: or above this te=pera-ture.
A concern regarding this matter would be if asse=blics '=oved upward into a posi:1ca such that control red cove =ent would be nindered.
2.
Inspection Report 50-346/78-06, paragraph 4, reported reactivity -
pcver escillations in 'the Davis-3 esse core.
These escilla:1cns have also occurred at Oconee and are attributed :o staa: genera:er level escillaticas.
3&W report 3AW-10027 s:stes in A9.2:
- Sua
{h U.f
()L{ 2 $ $ b A A
c.; -
s 4
..J. 'F. Streeter 2
January 5, 1979
+t The OTSG laboratory codel test results indicated tha: periodic oscillations in steam pressure, steam flow, and stea: genera:rr pri=ary outlet temperatures could occur under cer:ain c.,di:f ons.
It was shown that the oscillations were of the type associa:ed with the relationships between f eedvater heating chamber pres-sure drop and tube nest pressure drop, which are eli=inated er reduced to. levels of no consequence (no f eedback to reseter system) by adjustment of the tube nest inle: resis:ance.
As a result of.the tests, an adjustable orifice has been installed 1n the downco=er section of the steam genera: ors to provide for adjustmen: of the tube nest inle: resis:ance and to provide
- the means for elimination of oscillations if they should develop during the operating lif etime of the genere: ors.
The 1.itial orifice setting is chosen conservatively to minimize the' need f or f urther adjus:=en: during the star:up tes: pregra.
We also note tha: the' effec: on the inecre det:c:or sys:e= f er monitoring core paraceters during the oscillations is ne: clea r.
3.
Inspection and Inforce=ent Report 50-346/75-06 docu=ented tha: pres-
.surizer level had gone of f scale for approxima:ely five =inutes dur-ing the Nove=ber 20, 1977 loss of offsi:e power event.
There,are some indications that other 3&W plants =ay have proble=s maintaining pressurizer level indications during transien:s. 'In addition, under mm certain condi: ions such as less of feedwa:er.a: 100~ power with :he reactor coolant pu=ps running the pressurizer ma'y void cc=pletely.
A special analysis has been performed concerning this event.
This analysis is attached as Inclosure 1.
Because of pressuri:er level maintenance problems the sizing of the pressuriser =ay require f ur ther revi, v.
e Also noted during the event was the fact that Teold' vent of f scale C
(less than 520 F).
In addition, it was noted that the makeup flow monitoring is li=ited to less than 160 gps and that makeup flev
=ay be substantially graater than this value.
This infor=ation should be examined in light of the require =ents of CDC 13.
4.
A =emo' from 3&W re~garding control rod drive system trip breaker caintenance is attached as Enclosure 2.
This ce=o should be evaluated in ter=s of shutdown margin maintenance and ATVS censiderations par-ticularl? in light of large positive moderator coefficien:s 211ovable with B&W facilities.
0 54; O
e 0
..-t
. n..r.. n. m m.
v
'. /
J! 'F. Streeter 3
January 8
!E79 9.:._.; 5.
Inspection and Enforcement Report 50-346/78-17, para.;raph 6 refers to inspection fi5 dings regarding the capabili:y of the incere dete:-
ter syste= to deter =ine vorst case ther=al condi tions.
The rasetor can be operated.per the Technical Specifications with the center incore s tring out of service..
If the peak power locations is in the center of the core (this has been the case at Davis-Besse),
factors are not applied to conservatively moniter values such as Fq and T delta R.
- 6. describes an event that occurred at a 3&W facility which resulted in a severe ther=al transient and extre=e dif ficulty in centrolling th'e plant.
The af ore =entioned f acilities should be revieved in light of this infor:ation for possihle saf ety i= plica-tions.
.)
N*
db-
,)
. & ~. "
J. S. Cresuell Reactor Inspecter Enclosur~es:
As stated cc v/o enclosures:
G. Fiorelli
""1: C. Knop T. N. Tz=bling S
I e
- e o
e
. gg-q:y
C. Q.
Describe the fuel damage that may have occurred during this accident.
A.
Based on a preliminary evaluation of RCS pressure and temperature during the accident, fuel assembly outlet thermocouple readings, signalsfromthefixedincoreself-poweredneutrondetegtors acting as thermionic elements at temperatures above 700 f, and estimates of hydrogen evolution, it is estimated that at least the upper five feet of the core was uncovered during three periods for a total time in the range of two to four hours.
A preliminary estimate is that 15% to 30% of total Zircaloy inventory is oxidized, but that little or no fuel pellet melting occurred.
Fuel assembly structural' components, such c: the control rod guide tubes, and the control rods remain intact. Additional details are provided in the attached memorandum.
O