ML19224B901
| ML19224B901 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 05/15/1979 |
| From: | Mattson R Office of Nuclear Reactor Regulation |
| To: | Jennifer Davis NRC OFFICE OF INSPECTION & ENFORCEMENT (IE) |
| References | |
| NUDOCS 7906270170 | |
| Download: ML19224B901 (3) | |
Text
,,,q C UN g
'hkW
- /
v&g g panc ff
[o, UNITED STATES h#
$c NUCLEAR REGULATORY COMMISSION N'
'd Y. )f j y i
,//
c e
wasmucros. o. c. 20sss AgL-/ :
/
(G MY 151973 MEMOMNDUM FOR Johr,G. Davis, Acting Director Office of Inspection and Enforcement FROM Roger J. Mattson, Director Division of Systems Safety SUBJECT TMI-2 INVESTIGATION The attached copy of an August 9, 1978 letter from B&W to the Davis-Besse licensee was handed to me by a reporter at last
~~
Friday's ACRS meeting. At page two it indicates that B&W advised at least one of its owners some months ago of the need to observe both pressurizer level and RCS pressure to assure that the RCS is filled. Since this is a topic of interest in evaluating the circumstances surrounding the THI-2 accident, I thought it might be of use to your people in their investigatory activities.
h P //
Roger J. Mattson, D. rector Division of Systems Safety
Enclosure:
As stated cc: P Collins, DPM
. Ross, DPM R. Tedesco, DSS T. Novak, DSS V. Stello, DDR E, Gs,4, A)IL YL 255 052 0
7906270/78 j
i.
s f
[
fth i
~
y fI L' '
Be'ocock &Wilcox ro-,c.,,,. a,c w m
.'C.
5 V
P.O. E r 1260. Lyts:..Nr;. va.
9 Te:esm:nt: (E04)3SL'11)
August 9,1978 SC:: rLc3 6 23-C :. -
A l12322 73.3.1 i,.
-0 S:7 /1k/293
- * ;i
- a. <
4-4 i-y,b M.
T. D. 26.rray, Station Superittendent
,2 Davis-3ess e Nu Ice.r Pover Static:
).v4 5501 No-th State Rot te #2 y,
0:.h Earbor, Chio 43kk9 a,
Subject:
SM'JD Rapid Ccoldovn Transient 6* i,.
<.1t A+. a, -
D:ar Ter:/:
s w,.
u
,r On March 20, 1978, Eencho Seco experienced a severe ther=.11 transient initi-ty the loss of electrica.1 pcVer to a substantial portice of the Ncn-N ::le e.::-
Instrunentation (IOC }. Tne loss of rever directly caused the 1 css of Con.r-t
- +%
,g g' Ecr. indicatien of many plant pr.re=e;ers, the less of input c.f these par -
y
=eters to the ple.nt ec=puter, e.nd err netus input s i gn a.1s (- 1 drange, :.er:.,
1 "fl or othervise i=:orrect) to the Integrated Centrol Syste: (ICS).
- ,1 j
[,j Tne ple.ct respcase vas c:t the usual trer.sient in that t :e ICS responded tt
.P the erroneous ineut
--- u mthe:r than actus.1 plant cenditio=s, and result.
si
.r.ls
[.h in a Reactor Protection Syste. (P2S) trip on high pressure.
Subsequent to fd/ j the Reactor Trip, the errcneous signals to the IC5 contributed to the rapit.
. f.',
~
c 01devn of the ECS, Plant __cperctors had extrene difficulty in dete:._ining i
the true. status of ecce of the pie.=t parc=eters and in ec:. trolling the plc.:
u./
b ecaus e of the errene.ous indicaticas in the Control 300=.
- e. i, n.:
y J{,
An investigatien of the events fol.leving th.is loss of pcVer points out a Oc j.j fcr a close lock at operator traini=g and e=er5ency operating procedu-es fc i-any 1 css of ICTI pc er (or portion the: eof).
7:e f 012cving recc-end. tic:s
.1h, are cade to assist your ste_"f in a review of tra.ining end procedures to ass-proper operator actica for events of this nature.
4r-f 1.
Operaters should la _tra_ined to ecocni ze n less 9_f_pev tm a?1 or a d
enjerity of their ?CTI (e.g. indicators f ail to nid-range, sutenstic or 11 canual transfer to alternate inst..:;ent strings brin 63 00 resp::se).
6
Tne 1 css of power is e=;he. sized here rather than the fcilure of any cc:
8.b inst: ::ent or centrol signe.1 which are adequately covered in current
~
s 7 7'gf sic'.:.latCT training Ccurs es.
.W
. c, j
. h. *,:4
.Da
~,e-e F
r b T
. s.<s ',
Jj v
..?*
- e.y
%,,.-. g
, u...7,c
.-.c w,, a m
-- r-r-y
,1 01 1 2"T -' mr-.
c ilgjhhh.uo.7,:4dL awhc,.mWwu D
='N el..e;.LnMG; 4;JJ,.4.4%
.m.
/
/
A
~
,,;c..h ty kox SOM #LO3 62'-C01L
/
Aucust 9, 1978 ya:e 2 I
2*
Given that the cparator ces deter =ine that electrical pcver hcs been lost to e.ll or part cf the T:iI, he s} cad k.=ov the location of tha pc.:er sup;.ly breakers, and have a precedure availabic to cuier.y r -
5510 pcVer.
t 3.
If the fault ca.not te cleared (i.e. the breakers to the pcVer supplies I -
recpen), the operator should have a' list of al*.ernate instru=entation
{
ave.ile.tle to hin, and he shculd be thercup.ly traine:. In7T.s us e.
a:ples are:
Lx-a a.
ISD.S panels b.
RFS panels i
c.
ICI (Issential C'entrols e:d Inst:.:=entatien) t SRCI (Safety Related Centrols and I stru=entatica) d.
c.
Beccte shutdevn pe=els f.
Local gages s.
Picat ec=puter a
h.
p.e cgni:ing that no precedure een cover all pessible ec=bita*icts of IC.I L
failures, the operator's respense should be keyed to certain variables.
If the cperater reali:.es that he has a= instru=entation prebie= (es op-
)
pcsed to a LOCA or stec= line bree.h, for c::e=ple), he can li=it the tre=sient by centrolling a few criticcl va-iables :
l 2.ressuri' el (.vi a EPI cr no:=al Md.eup Pu=ps )
c.
b.
RCS pressure (via Pressurizer hea*.ers, spray, E/M relief valves, e.c.)
Stea= Gene ator level (via feed flev, feedvater valves, etc. )
c.
d.
Stsc= Generator pressure (via turbine bypt.ss syste=)
'The pressurizer le el and ECS p: essure assure that the Reactor Cccle..
S ste= is filled; the Stet:
f dece,y heat re=cval.
0 Gescrator level end pressu e assure adequa*.e
}
Attach:ents 1 cad 2 are provided to give a brief descriptics of the events ic11cvi 6 this loss of M;I power at Rancho Sece As can be seen by this transient,'pr,0 pt precise operator action e-.d, e ability to recqcnize e
. css of ICi! pover are critical factors in 11=iting the severity of a trans-e t such as tb s.
~
l f,ycu have a.y questic:,s or ev ents, ple tse advis e.
I I
Yours truly,
H, 9b.m ~
Ivan D. Creen f,fy)6]l 50h$
Site Cpernticas Manager
- m.. a.,,
a o r, -
,e
- See attached sheet.
/-
's; p.
I r
r w
n ;,..
.m.3. o,, w. x..m...u..; w.g. x enm F.S ? h h O A n cs w IO W u W i M M M e h aM ;*" 6 4.. m.e. -
,.s.
.g.
.