ML20059F744
| ML20059F744 | |
| Person / Time | |
|---|---|
| Site: | Oyster Creek |
| Issue date: | 08/31/1990 |
| From: | Devine J GENERAL PUBLIC UTILITIES CORP. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| 5000-90-1970, C320-90-751, NUDOCS 9009110336 | |
| Download: ML20059F744 (6) | |
Text
b
]D; 9
+
GPU Nuclear Corporation u.
y-One Upper Pond Road
-l a
Parsippany, New Jersey 07054 E
201 316-7000-s' TELEX 136 482 O
Writers Direct Dial Number:
w u
August 31, 1990 y
5000-90-1970 C320-90-751 i
U. S. Nuclear Regulatory Commission Attentions-Document control Deek Washington, DC 20555 l
.r:
- \\
Dear Sir I
1
Subject:
Oyster Creek NuclearTGenerating Station Docket No. 50-219 License No. DPR-16 Hardened Vent j
By letter dated. June 15, 1990, the NRC staff forwarded its backfit analysis l
regarding installation of a hardened wetwell vent at the Oyster Creek Nuclear.
f Generating Station, and requested that GPU Nuclear (GPUN) commit to-install
{
such.a vent system by January, 1093.
We wet with the staff on Jaly 24, 1990, g
along.witn other BWR utilities, to discuss the technical questions about the 4
value af a hardened vent in view of Oyster. Creek. specific design features sr.J the technical value of completing the IPE.before deciding. By letter dreed
.Aug'.st 20, 1990, the staff reconfirmed its belief that a hardened vent should g
be' installed at Oyster Creek.
IWe believe that a decision to proceed with a complex.and expensive safety system modification such as this one should be based on the results of a l
methodical, plant specific evaluation such as the Oyster Creek plant specific
' individual plant examination (IPE) now underway. Our letter dated -
- October-30, 1989,. committed to install a hardened vent at Oyster Creek if the Oyster Creek IPE demonstrates that such an installation is warranted. The staff's backfit analysis addresses some, but not all', of the technical issues
)
bearing on our decision. Our technical evaluation and comments on the staff's jo' analysis are provided in the enclosure.
I r
00106 w
- 9009110336 90K)B31 PDR AIX)CK 054)D0219
. p PNV g JU
C li f
\\
, 4: 2.. -
Ha'rdened V:nt C300-90-751 Page 2 i
However, we have decided to proceed with the installation of a harcened vent at Oyeter Creek on the assumption that the staff's conclusion is valid. We will commence engineering shortly, and we will meet with the staff to agree upon the design concept and requirements.within six months. We plan to complete that installation prior to restart from the cycle 14R. refueling outage, which is currently scheduled to commence in October 1992.
Sincerely, b
J i
.+
f-t,q)oc 1 J. C. DeVine, Jr.
i>
Vice President and Director Technical Functions JCD/PFC/crb 1 -
(C320751) cc: Administrator Region I U. S. Nuclear Regulatory Commission 475 Allendale Road King of Prussia, PA' 19406 NRC Resident Inspector Oyster Creek Nuclear Generating Station Forked River, NJ 08731 Mr. Alex Dromerick, Jr.
U. S.. Nuclear Regulatory Commission Hail Station P1-137 Washington, DC 20555 L
f C32051
j e
.e ENCLOSURE
SUMMARY
OP CPUN REVIEW QE NRC STAFF BACKFIT ANALYSIS n
A HARDENED WETWELL VENT hI OYSTER CREEK l
INTRODUCTIOE We have reviewed the NRC staff's backfit analysis for the subject modification t
forwarded with their letter of June 15, 1990, including the underlying analysis 7
contained in the october 19, 1989 internal NRC memorandum from 8. Sheron to A. Thadani. We have also had the benefit of an exchange of information with l
the staff in a meeting on July 24, 1990 with members of the BWR owners Group.
i DISCUSSION There are a number of points in the staf f's analysis that appear to have been '
treated in a less than rigorous manner.
We outline below only those aspects that have major effects on the conclusion.
1.
The analysis is not specific to t!* Ovster Creek desion A.
In its October 19, 1989 memorandum the staff points out that it was not practic;A for them to perform a plant-specific-analysis and that PRAs of other plants were used to draw conclusions about oyster Croek. The staff analysis acknowledges that the reference plant (Millstone 1) has only one isolation condenser whereas Oyster Creek has two redundant isolation condensers to mitigate TW l
sequences. Long term makeup to the shell-side of these isolation condensors is provided by one of two condansate transfer pumps or the fire water system which is supplied by an electric driven pump or one of two diesel driven pumps.
Two other electric driven pond pumps could also be used.
These sources, which provide significant reliability, were not explicitly considered by the staff.
Other means of decay heat removal available at Oyster Creek (see attachment) were also not explicitly considered.
i B.
The staff's analysis "... assumes core melt will occur on loss of containment cooling and containment failure," but readily acknowledges that "... this is not necessarily the case."
It 4
depends principally on the failure mode of containment under the overpressure conditions that would be expected to develop over an i
extended period (many hours, even up to a day or two) of rejecting
?
decay heat to the torus without containment cooling.
No plant specific analysis of likely containment failure modes under these C32051
- t 4
ll l
[,
l conditions has bsen performed by the staff. This will be performed c:
by CPUN as part of the Level i PRA for oyster Creek.
In addition, it is unlikely that the containment would fail at all under these conditions because.of the use of the existing torus vent.
Use of i
this vent to assure that the primary containment pressure limit is not exceeded is currently contained in the emergency operating 3
procedures, and operators have been trained to follow these procedures.
In addition, given the long time available and the fact tnat senior management will be present, will help assure that venting is done when appropriate.
{
C.
Other benefits of venting cited by the.backfit analysis such as i
preventing the failure of emergency core cooling system pumps due to inadequate net positive suction head (NPSH) and the reclosure of j
the valves in the Automatic Depressurization System (ADS) are not' i
applicable to oyster Creek. Oyster Creek utilizes electromatic
+
relief valves for ADS that are not sensitive to containment pressure like the air-operated valves considered in the staff's analysis.
Furthermore, NPSH for core spray pumps would be lost j
under near saturated conditions (within 50'F of saturation) in the
[
torus and venting would not be expected to have a favorable effect on these conditions at oyster Creek.
In fact, venting of any overpressure would reduce the NPSH available and could result in premature failure of core spray.
i 2.
The analveis does not account for the existino torus vent In our meeting with the staff on July 24,'1990, they indicated that
[
they did not " allow credit" for the existing vent because they were:
unable to determine if adequate makeup sources to the taactor would
{
remain after the postulated failure of the existing " soft".ductwork.
The. staff's analysis assumes that almost all of the intermediate L.ed-late core melts could be eliminated by venting provided alternate injection sources to the reactor vessel are availablo.- At oyster Creek these sources are (assuming core spray is disabled). main feedwater and i
condensate, or the control rod drive system which can provide enough flow to remove decay heat after the one hour decay period following.
shutdown.
Both feedwater and condensate are outside the reactor 3
building and (subject to verification in the PRA) are not expected.to.
be adversely affected by the environmental factors associated with failure of the existing " soft" ductwork vent within the reactor
[
building.
Consequently, for oyster Creek, the efficacy of the existing l
vent in preventing core damage is comparable to the hardened system suggestod by the staff.
l
.t i
l 02051 I
3 ',..
, 3.
The analysis neelects to examine viable alternatives Alternatives for recovering the containment cooling function that may be preferable to hardening the existing vent ductwork are not considered.
For example, cross-connecting fire water or some other.
source to the containment spray or emergency service water system is not considered. This latter alternative appears particularly attractive for Oyster Creek because it would preclude loss of NPSH for the core spray pumps and provide a means of preventing containment failure (diverse from the existing vent system) that would preclude the release.of radioactive fission products to the environments and it could be provided at significantly less cost.
4.
The analysis does not show a favorable cost-benefit ratio Even using the inappropriate and conservative analysis outlined above and conservative assumptions about averted costs, the staff estimates that only 796 man-rem would be averted per million dollars expended for oyster Creek.. This does not meet the benefit-cost threshold for backfits of 1000 man-rem per million dollars established by the commission. We believe a plant specific. analysis would give results similar to those provided to NRC by Northeast Utilities for Millstone
-(i.e.,.a cost benefit of much less than 1000 man-rem per million dollars.
EQNCLUSION
[
Based on the above points we conclude that the staff's analysis does not support their conclusion that hardening the existing vent is a cost-beneficial modifi9ation for oyster Creek.
l-a l
?
l l
l l
s e
s C320M1 t
'I e
^~
M 4
- =
ATTACHMERI EXISTING METHODS FOh REMOVING DECAY HEAT AT OYSTER CREEK
- 1. - Reject heat to the main condenser with makeup provided by main feedwater or condensate (normal mode).
2.
If the main condenser is not available, rejcet heat to one of two isolation
't condensers. These in turn reject heat directly to atmosphere.
Long term f
makeup to the shell-side of the isolation condensers is provided by one of two condensate transfer pumps or the fire water system which is supplied by an electric driven pump or one of two diesel driven pumps.
Two other electric driven pond pumps could also be used.
- 3.
If isolation condensers are unavailable, heat is rejected through EMRVs to
(
the torus (this would also be the heat removal mode for LOCAs and stuck open EHRV sequences). Makeup to the reactor vessel can be provided by the core spray system, main feedwater or condensate, or control rod drive pumps which have sufficient capacity to remove decay heat after the one hour i
decay period =following shutdown. Utilize the containment spray system to l
remove heat from the torus and reject it to the emergency service water
.f system.
4.
If containment spray or emergency service water is unavailable and the i
torus temperature approaches NPSH limits for the core spray pumps, core spray suction can be switched to the condensate storage tank.
If this-fails, fire water can be valved to inject through the core spray system.
Either of these options would significantly increase the time available (many hours).to recover containment spray /ESW.
Another alternative that will be examined in the PRA would be to hookup fire water to containment spray or ESW.
5.
In the event all of the above methods fail and reactor temperature and pressure can be sufficiently. reduced, the separate shutdown cooling system with three pumps and three heat exchangers could be used for long term decay heat removal.
6.
Other methods that utilize the reactor water cleanup system or feed and c
~
bleed'of the torus could also be employed.
7.' In the unlikely event that all of the above methodo fail, the existing torus vent system could be used to remove heat from the torus with makeup provided by main feedwater or condensate.
If the existing vent system
-c ductwork were to fail, the resultant environmancal conditions in the reactor building would not be expected to adversely affect either of these sources because they are located in the turbine building.
t t
h l
C32051 l
1 l
-