ML20100M368
| ML20100M368 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 11/30/1984 |
| From: | Hukill H GENERAL PUBLIC UTILITIES CORP. |
| To: | Murley T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| References | |
| 5211-84-2286, IEB-84-03, IEB-84-3, NUDOCS 8412120293 | |
| Download: ML20100M368 (8) | |
Text
-
Nuclear L"su"s";ftre" "
Route 441 South Middletown, Pennsylvania 17057 0191 717 944 7621 TELEX 84 2386 Writer's Direct Dial Number:
November 30, 1984 5211-84-2286 Dr. Thomas E. Murley Region I, Regiona.1 Administrator U. S. Nuclear Regulatory Commission 631 Park Avenue King of Prussia, PA 19406
Dear Dr. Murley:
Three Mile' Island Nuclear Station Unit 1 (TMI-1)
Operating I,icense No. DPR-50 Docket No. 50-289 Response to IEB 84 Refueling Cavity Water Seal Failures The potential for and consequences of a refueling cavity seal gasket failure at TMI-l has been reviewed as requested by IEB 84-03. This review has resulted in the conclusion that no new action to further mitigate the potential or consequences of a seal failure need be taken and that adequate design features and administrative controls exist at TMI-1.
The physical configuration of the TMI-1 cavity seal and its lack of any active sealing component makes gross seal failure during refueling evolutions extremely unlikely. In addition, the arrangement of the fuel transfer canal, the cavity seal plate, and the spent fuel pool area precludes the possibility of draining the fuel pools to a level which would uncover the stored fuel.
In the event of significant seal leakage while handling fuel, the decreasing water level would be detected by visual monitoring and an automatic low-level alarm. TMI-l procedures initiated by these indications provide guidance on appropriate operator action.
h2120293841130 ADOCK 05000289 G
PDR GPU Nuclear Corporatloa is a subsidiary of the General Public Utilities Corporation
Dr. Thomas' E. Murley 5211-84-2286
-Additional details on the plant specific layout and administrative controls are included in Attachment I.
Sincerely, f
M
.bd 2 kill Director, TMI-l HDH/WSW/JGB/kds
Enclosure:
Attachment I cc: Document Desk Sworn and subscribed to before.me this n
. day of - -
a m )
1984.
a&C ma )
Lk =af Notary Public
- I DAkLA fiAN 8tnY, sqiANT PU8 tic MIDDLETOWN BORO. DAUPHIN COUNTY MY COMMISSION EXPlRES JUNE 17,1965 Member, Pennsylvania Assocation of Notanes
n.
~
5211-84-2286 Attachmant I Page 1
~
=M FUEL TRANSFER CANAL SEAL CONFIGURATION The TMI-1 seal' plate. assembly is a flat annular steel plate that fits around the upper reactor vessel flange and bridges the gap between the reactor vessel land the transfer canal floor. Prior to refueling, this plate is lowered to the transfer canal floor and sealed along the inside and outside edges:per Figure 1.
The outer seal is a '3/4" x 7/8" x 900" passive. rubber
~
- gasket located between the seal plate and canal floor,-upstream from the
.72 seal plate studs. The inner seal is a 3/4" x 1 3/8" x 720" passive rubber gasket located between the seal' plate and reactor vessel, upstream.
from the gasket lip. This lip, as well as the torque from the 72 studs,.
retains the inner gasket and would prevent it from being pushed through the 1/2-inch seal plate / reactor' vessel gap.
Additional pressure to retain the seals and to increase their effectiveness is provided by the head of water present during refueling operations. In the TMI-l design the refueling water actually decreases the likelihood of gross seal failure as opposed to the design described in the Bulletin.
The relationship between the core, transfer canal seal plate, spent. fuel.
+ pool, fuel transfer tubes, and fuel handling equipment is shown in
. Figure 2 and Figure.3.
. At' TMI-1, seal leakage is more likely to occur during initial fill or drain operations when the pressutes on the seal plate gasket are changing. During these operations no fuel would be in transit and ' the consequence would then be limited to the operational. concerns related to the spilled water.
'In consideration of the above TMI-1 seal plate design, massive seal gasket failure is extremely unlikely.
FUEL TRANSFER GASKET SEAL LEAKAGE EXPERIENCE TMI-l has had seal plate gasket leakage mostly limited to very minor weepage with the exception of one leak which occurred late in the draindown following the 1979 Refualing. In this instance, approximately 2-3 ' feet of water was lef t above 'the seal plate when a noticeabe increase in leakage occurred.
The remaining 2-3 fect in the canal was drained out in about 45 minutes by a combination of the leak and normal drain down path.
It has been estimated that the leakage past the gasket was around 500 gpm based on these observa-tions.
Leakage which occurs during draindown is believed to be caused by the gaskets taking a " set" while the transfer canal is full resulting in insufficient-springback as the water level is dropped. This can be exacerbated by gasket distortion caused by traffic on the seal plate following the last torquing pass sequence.
~~
5211-84-2286
~
Attachornt'l Page 2
-FUEL TRANSFER GASKET SEAL LEAKAGE EXPERIENCE (Cont'd.)
Following this ' leak, a procedural change recommended by the vendor was made to minimize this type of leakage by ensuring that the final: torquing pass sequence be redone'af ter major traffic'on the seal plate (such 'as caused by head detentioning) and after draindown before any. subsequent refill.
The experience at' TMI-1 and other B&W sites has been that. leakage occurs at
. initial fill or draindown and not when full level Is present (as is,the case
.during_ fuel handling).
MAXIMUM LEAKAGE ESTIMATION A conservative estimate of' leak rate for a major gasket failure has been made even though the TMI-1 design incorporates no active component and leakage
-initiated with the transfer canal full is very unlikely. The maximum credible' leakage is estimated to be approximately 4700 gpm.
This 'is almost an. order of magnitude greater than any previously experienced -
. leakage and is based on a loss of seal ~ along 50% of the inner gasket circum-
- ference. Leakage of this size would result in an initial Fuel Transfer
-Canal (FTC) water level decrease of two inches per. minute while the Spent Fuel Pool (SFP) is not isolated. With the SFP isolated the rate would be about 5.3 inches per minute.
An operator handling fuel would have more than 25 minutes from receipt of the -
low level alarm to position the fuel over an appropriate storage location and start lowering the assembly. This would be_ the case under normal fuel handling conditions since the SFP would not be isolated. At the end of 25 minutes more than 4 feet of water will still cover the fuel assembly. Four feet is con-
~
sidere'd' adequate shielding to allow continued operator presence.- Since the slowest speed that the fuel assembly can.be-lowered is 5 fpm sufficient additional time is available to complete the fuel move.
Once the fuel is safely positioned in a storage. location the SFP would be isolated to reduce the total leakage to the Reactor Building Sump. At this-time the remaining level would decrease at the more rapid rate.
AVAILABLE MAKEUP CAPACITY
- With -respect to makeup capacity, the only significant source of borated water-other 'than the SFP would be recire of the spilled water from the Reactor Building Sump. This could be done from the control room in a very_ short time-frame and would provide makeup in excess of.3000 gpm. Use of this flow path is considered to be a drastic measure and would only be initiated if abso-lutely necessary to prevent fuel uncovery. Based on the potential for massive leakage and the time available for other operator action to safely store fuel in' transit, the use of this makeup path is not likely.
m-
'J.
5211-84-2286
,1 Attachmsnt I Page 3' AVAILABLE TIME WITH'NO OPERATOR ACTION Assuming no operator action, which is not considered credible, the top of fuel in transit would become uncovered in about 50 minutes. Initial cladding damage would occur at a later time after clad tempera;ures had
' increased to 1100-1300 degrees F.'
The actual additional time would vary based upon. the assembly history and cooling time.
.No uncovery of fuel not in transit would occur since_the core and the fuel in storage are both below the fuel transfer seal plate elevation. With no operator action the level in the SFP would decrease over a two-hour period and stop with l'6" of water above the stored fuel.
EMERGENCY OPERATING PROCEDURES TMI-l has two approved procedures which address operator action following seal plate leakage. Alarm Procedure PLB-4-9, " Fuel Trans Canal Hi-Lo", would
' be initiated on receipt. of a low-level alarm when water level drops below the 344' 3" elevation.
Either this procedure or operator observation will initiate action in'accordance with Abnormal Procedure 1203-43, " Fuel Transfer Canal Seal Plate Failure", if refueling evolutions are in progress.
AP 1203-43 requires evacuation of all unnecessary personnel from the Reactor Building, directs the operator to lower any fuel in transit into the core or FIC deep end, and to isolate the SFP. After the fuel _is stored and SFP
' isolated the canal is drained to the BWST to minimize the leakage to the Reactor Building Sump.
m --
5211-04-2206 FI.surt i
Attachment I Page 4 TAT.-l SEAL PLATE A SSIABLV
?.~.,;;* "
m..
. se * =
g F.3 *
)
y
,t--
.yg
//, /, /// // / /, /,,,,.,.,.v...,
w:[l ",..
]((y,._,
1.E-'*'~
Le m. -. 'e~
m p....
g
.,e.
.w SKTDGFe Ct.s<sne seus e. ass s s e
f, wr'2"'..efe,1848'~.~"'-=f,.
/ [\\
C4 s ~ *
/
_l t
-+ / /r -
..i'
.,,. h. 4 -
/ra w _
N'
<e l/
/
4 x
N m
- s.,.
/
> .).'. *w.
f i.
g l.,
y l
.,o.
1 n
0
.,*3 i
[ ~ - :...'.,..' '.:,...,,
~
f l
i y
3)
-. y *,..
.......t. - 9.; P-c
.~
- i s..
i /w
.s ama s,e.w I
i i
\\
l
\\
s d
a
./
m
\\\\ -
. r
.v.
7,e w
\\ \\
S i'
44#
/
g, * \\ ',
g' s
\\ ',
j
\\',
i i
\\ [#[
p.
Y~ ~., #,
Y
< %g%..g..
u,,.:..
N.
p$mt W
D oft.s.m <rew V
48eamos'O
5211-84-2286 Attachment I Page 5 F t c. u. u 1 RX. BLDG. FUEL HANDLING ELEVATIONS
( meAcrea atos.
\\
sen
,c.*ti.....
.g j
c m.o
..say W
l D/
n f
see u o'
- vsos..
I / * * * *C ' "' GE)
E *~
H\\\\
/d &2:
er isi e
i I
-d I M. UP,
E I'I r '** # -
FN I I I I
.:.7 5,
L I
' goNIN AL W arvA r auet.
W; j
n.*A. 3 W -(." to
.l
,.Lo.teve t c
34s'-on Mac m ' yN.{- 3*oJ yToP of F. A. AT Fou. Up
[I EUN. '335' lofs" q ntAcTest i
8 l l a ~~'
I.'
I I
Top.f-i i
'e 4 i
gguear,,
i f
fgS[jy E*****
L' ""3'##
ait.' 6..
r-w
- sw _
-r-k b
f A
~
..aame r..<
.y k' ' mes.kea
^-
1
~
_nav. se'- 6%"t(cens)
Q n'
i."
4 1
C
\\\\
,& A
' ~l c,t7 ng, m.
l t
w u
7
,**9 s
1, N
7.
"A e
g.4 l
' n m.ss -
ne
%c,<se w t w:e*
m.
5211-84-2286 Attachment I PIGuRE 3 FTc/sp p DImENSIoM S Eum l En t-er bl SFP A gyp, g 34t.' - -
- 340' stav 194.s s
esav 5
tav at,.
i I
32i
/
hat,7 31 % 5 '
8 I stav TgfA,,, i i
- - s 4.s'
'"v l8 Cau.PLhtt 868V zutum
=, _.
i stav
. 3os' ELtv FRONT VIEW l
_L T
1,,.-]
5 -a 252" 9 fC SFP-R 1
7 csma 4
s 2.
e
_,,t l:
g,gg.
_, o3* y l:
7tel" Meel*-290" W il4~ W TOP VIEW F 2'*~H
_. 34 e's.tv 3%'ELEV g
oPth1M 6 9CTwtEu r
- m. a m,. s
,,e
-32.t' ELEV
- sa' u.tv SFP
- 304s.5' Ettt.
I - )os' tLt V H288~-*l kab8~-*l SIDE VIEW
-