ML20035A690
| ML20035A690 | |
| Person / Time | |
|---|---|
| Site: | 05200001 |
| Issue date: | 03/12/1993 |
| From: | Kelly G Office of Nuclear Reactor Regulation |
| To: | Duncan J GENERAL ELECTRIC CO. |
| Shared Package | |
| ML20035A682 | List: |
| References | |
| NUDOCS 9303290164 | |
| Download: ML20035A690 (3) | |
Text
'
if? "*%,o,i
.. p U"
UNITED STATES
(.k).,..m [/
3
- c. I NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20555 Docket: 52-001 March 12, 1993 NOTE T0:
$c Duncan, GE f FROM:
G nn Q y
[NRR j
SUBJECT:
DOCUMENTAT ON OF CONF RENCE CALL ON MARCH II, 1993 BETWEEN GE AND SPSB ON ECCS PUMP QUALIFICATIONS AND CLASS 11 SEQUENCES This note serves to document the conference call held yesterday among you, Bob Palla, and myself. The clarifications requested in this conference call were generated as a result of the March ;,1993 fax you sent me that detailed the qualification temperatures for the ECCS pumps. As we discussed, NRC found these qualification terrperatures to be much lower than expected and lower than were assumed in the ARWR PRA. These changes could significantly alter the results of the Class II sequences and have the potential to effect other accident classes as well.
If you have any questions on the enclosed documentation of our clarification requests, please contact Bob Palla or me.
Enclosure:
as stated 9303290164 930312 PDR ADOCK 05200001 A
4 i
l ISSUES RAISED DURING MARCH 11, 1993 CALL WITH GE 4
1.
Early venting appears to be an initial response to dealing with Class 2 sequences. A more detailed assessment of Class -2 sequences is needed.
This should include:
- identification of which vent lines would be used; the size, length, and loss coefficients of each line; where the lines draw from (wetwell or drywell); and whether they pass through the SGTS or go directly to the stack.
- a transient analysis showing that the vent line(s) would_not be i
isolated on drywell pressure (1.7 psi setpoint), or high offsite radioactivity release rate (in excess of LCO limit),
- a transient analysis of: containment pressure and temperature, suppression pool temperature and level, mass additir;n from the CST or other external sources, and
- an accurate representation of operator responses (per EPGs) in the transient analysis. This should include, as applicable:
opening / closing vent lines, transferring suction between CST and suppression pool, actuation of ADS, securing RCIC or HPCF on high pool temperature, initiating ACIWA, and securing injection from external scurces upon containment pressure or high suppression pool level.
2.
The Class 2 CET must be modified to reflect the impact of the ABWR EPGs on sequence progression and pump availability.
In particular, the CET should 'eflect: (1) early containment venting (as suggested in March 8, -
1993 letter from J. Duncan), (2) actuation of ADS (and loss of RCIC) as required by E0Ps in response to high suppression pool temperatures, (3) the fact that only the RHR pumps are capable of sustained operation at temperatures in excess of 212 F, and that these pumps are unavailable by i
definition for events in which the rupture disk is challenged, and (4) the need to secure injection from external sources in the long term, irrespective of whether adequate core cocling is assured, due to high wetwell pressure or water level. A detailed justification for the quantification of each branch of the revised CET is also needed, preferably based on fault trees.
3.
The frequency of Class 2 sequences from the May 26, 1992 updated level 1 event tree analysis submittal (1.lE-10) is inconsistent with the frequency of core melt obtained in the updated CET analysis (1.1E-12 in Figure 9 of CEB-92-39). The level I and level 2 interface for Class 2 -
sequences needs to be described in detail, and the inconsistencies rectified.
i 4.
Due to its importance to ABWR risk, the following additional information on the final design of the ACIWA system is needed:
- figures or P& ids showing firewater pump connections to RHR, and flow restrictino orifices,
- pump head curves, and flow rates assumed in MAAP analyses, and
- clarification on whether flow from ACIWA cac be routed to: (a) wetwell spray, and (b) wetwell spray and drywell spray simultaneously.
j c
~
'9 i
t i
5.
Due to apparent changes to the wetwell spray system design, the following information on the final design _ of the wetwell spray system is needed:
- mode of actuation in final design, j
- conditions _for initiating and teminating, and related E0P steps,
- sources of water, and
- assumptions that were'used in the MAAP analyses, and suppression -
j pool bypass assessment regarding wetwell spray operation.
]
6.
There appears to be major inconsistencies between the PRA analyses and d
assumptions and the E0Ps in their present form (as of Amend 24). Some examples include:
~~
- the E0Ps call for emergency depressurization on high suppression pool temperatures, but these actions are not-reflected in the CETs for Class-2 events,
- one portion of the E0Ps calls for use of the firewater addition mode of RHR under certain suppression pool pressure / level conditions (PC/P-6, Amend 24), yet another section of the E0Ps calls for securing this system under the identical conditions (SP/L-3.3, Amend.
i 1
22),
- the rupture disk pressure setpoint values from the earlier ABWR 1
design (80 psi) is still reflected in the E0Ps '(PC/P-5, Amend 22),
- water levels in several figures in the E0Ps (27.2m) represent a-i nearly water solid containment (PC/P-6); the analyses in support of the PRA are based on an understanding that the operators would secure injection from external sources (e.g., firewater) once-the containment water level reaches the elevation of the lower head of_
the RPV,
- the value of " bottom of suppression pool-to-lower-drywell vent (ll.7m)" in the E0Ps does-not correspond with either.the pool overflow elevation or the elevation of the RPV lower head reported i
in level 2 PRA submittals; it is unclear why this value is used in the E0Ps rather than the PRA values, and
- the E0Ps reflect a containment-over core philosophy even prior to i
actuation of COPS (PC/P-6, Amend 24). Such a philosophy is inconsistent with the results of GE's Level 2 PRA analysis (e.g.,
for Class 2 events) which indicates that'the probability of core j
damage as a result of containment failure is very low.
-i A deliberate, and systematic QA check on the consistency of the level l' and level 2 portions of the PRA with the final E0Ps for the ABWR is needed to assure that the E0Ps are adequately reflected within the PRA.
5 The scope of this review should include the level 1 event trees,.the containment event trees, and the supporting MAAP analyses (i.e.,
sequences presented in SSAR Section 19E.2.2).
i I
I i
[
i
.