ML20058B044
| ML20058B044 | |
| Person / Time | |
|---|---|
| Site: | 05200001 |
| Issue date: | 11/15/1993 |
| From: | Taylor J NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| To: | The Chairman NRC COMMISSION (OCM) |
| Shared Package | |
| ML20058B047 | List: |
| References | |
| NUDOCS 9312010373 | |
| Download: ML20058B044 (8) | |
Text
.x.<
.p***c',
v 8~
i
~
UNITED STATES
'I I
NUCLEAR REGULATORf COMMISSION
/
wasmorow o c. mssam
- =**
November 15. 1993 MEMORANDUM FOR: The Chairman Comissioner Rogers Comissioner Remick Comissioner de Planque FROM:
James M. Taylor Executive Director for Operations
SUBJECT:
ADVANCE COPY OF THE DRAFT COMISSION PAPER, " DIVERSITY IN THE METHOD OF MEASURING REACTOR PRESSURE VESSEL LEVEL'IN THE ADVANCED BOILING WATER REACTOR AND SIMPLIFIED BOILING WATER REACTOR" To keep the Comission informed about key policy and technical issues bearing on future light-water reactor designs, I m enclosing a draft copy-of a Comission paper proposed on the above s'ubject. The staff believes that it is appropriate to have public and industry consnents on the issue discussed in this paper before requesting final Comission approval of the staff's posi-tion.
.The staff intends to issue the final safety evaluation report. (FSER) for the advanced boiling water reactor (ABWR) at the end of this year. Therefore, it may not be possible to issue this draft paper as a SECY paper before forward-ing the ABWR FSER to the Comission.
In that event, the staff will request the Comission's guidance on this matter through the FSER.
CONTACTS:
George Thomas, NRR 504-1814 Chet Poslusny, NRR 504-1132
[MO/O Enclosure
/
The Commissioners '
I The staff intends to place the enclosed draft paper into the public document room after three (3) working days from the date of this memorandum. The staff will provide the Advisory Committee on Reactor Safeguards (ACRS) and GE Nuclear Energy with a copy of the enclosed paper in preparation for discussion at the December ACRS meeting.
i Origir.sl signed by Janes M. Taylet James M. Taylor Executive Director for Operations
Enclosure:
As stated j
i I
.=
- g j
v:L a x i
ffE:
The Commissioners i
[BD3:
James M. Taylor Executive Director for Operations i
SUBJECT:
DIVERSITY IN THE METHOD OF MEASURING REACTOR PRESSURE VESSEL LEVEL IN THE ADVANCED BOILING WATER REACTOR AND SIMPLIFIED BOILING WATER REACTOR PURPOSE:
To request Comission approval of the staff recomendation that diversity of reactor pressure vessel (RPV) water level measurement be required in the advanced boiling water reactor (ABWR) and simplified boiling water reactor (SBWR).
BACKGROUND:
During the TMI-2 accident, low water level in the reactor vessel and inadequate core cooling (ICC) were not recognized for a considerable time.
Insufficient instrumentation was one of the causes of this problem.
TMI-2 Action Item II F.2, " Instrumentation for Detection of Inadequate Core Cooling," required licensees to consider additional instrumentation in order to provide an unambiguous, direct method to interpret indication of ICC.
The BWR Dwners' Group submitted the 3. Levy reports, SLI-8211 and SLI-8218, in response to the action item.
In these reports, the BWR Ovners' Group identified and considered methods for measuring RPV water level other than the differential pressure (dp) measurement currently and exclusively used in operating BWRs and proposed for the ABWR anti 5BWR. The evaluation was performed based on backfit considerations for the operating plants.
It was determined, by both staff and industry, that backfitting diverse level CONTACTS:
George Thomas, SRXB/DSSA/NRR 504-1814 Amy Cubbage, SRXB/DSSA/NRR 504-2875
w.
The Comissioners instrumentation on operating BWRs would not provide sufficient incremental safety benefit to justify the cost. The ABWR and SBWR were not considered in the study since their design had not yet been developed.
It should be noted that the discussion which follows is based primarily upon the ABWR design.
This is because the staff has not completed a detailed review of the SBWR design. However, the basic logic regarding level instrumentation importance, comon cause failure potential, and the desirability of diverse level instrumentation applies to the SBWR as well as the ABWR.
DISCUSSION:
The reactor vessel water level instrumentation in the ABWR will be used for both non-safety-related normal functions (controlling feedwater) and safety-elated functions (scram, containment isolation, and emergency core cooling systems (ECCS) actuation).
It also provides operator's with information necessary for act ons to ensure adequate core cooling in accordance with the emergency operating procedures. Many of the actions in the emergency procedures are keyed to reactor water level and many safety-related operator actions are taken to ensure adequate reactor water level.
As in operating BWRs, the level instruments for the ABWR RPV are all dp instruments.
Each instrument uses a reference leg, which is maintained full by a condensing chamber connected directly to the steam space in the RPV, and uses a variable leg which is connected to the RPV water space. All the dp level instruments operate on the same physical principle. Therefore, comon-cause failures caused by a design deficiency or maintenance error could result in inaccurate indication of reactor vessel water level.
The staf f's concern is based on experience with potential comon-mode failure mechanisms in the reactor water level instruments.
For example, during the past two years, anomalies have been observed in reactor vessel water level indication at several BWRs (Millstone-1, Pilgrim, LaSalle, and WNP-2) during controlled depressurization to comence plant outages. These anomalies consisted of " spiking" or " notching" of level indication, and in one instance, a sustained error in level indication., The effect of non-condensible gas in the condensate chamber and reference leg of the cold-leg type of water level instrun>ents has been determined to be the root cause of these level indication anomalies.
Testing has shown that under depressurization conditions, non-condensible gases can cause significant errors in the level indication.
To ensure high functional reliability of the instrumentation, the staff issued Generic Letter 92-04, ' Resolution of the Issues Related to Reactor Vessel Water Level Instrumentation in BWRs Pursuant to 10 CFR 50.54(f)," and Bulletin 93-03, " Resolution of Issues Related to Reactor Vessel Water Level Instrumentation in BWRs," requesting hardware modifications for operating reactors.
In response to these generic comunications, all BWR licensees have comitted to implement hardware modifications to their level instrumentation systems.
Similar cold-leg instruments are used in the ABWR design.
' b[2 '
~
The Commissioners Before the noncondensible gas level inaccuracies, there were other problems with dp level instrumentation used in BWRs.
In 1984, the staff issued GL 84-23, " Reactor Yessel Vater level Instrumentation in BWRs," to address concerns related to high c mtainment temperature during a depressurization event.
High containment temperature combined with reactor depressurization can lead to false water level readings as a result of flashing or boiling in the reference leg within the containment.
These known comon-mode deficiencies in BWR level instrumentation systems have been addressed at operating BWRs and by GE in the ABWR design.
It should also be noted that these particular deficiencies would not have compromised the automatic protective functions of the level instrumentation for accident scenarios initiated while at power, and that no previous incidents at BWRs of inaccurate level indication have been misinterpreted by plant operators so as to lead to unsafe actions. The staff concludes that the ABWR level instrumentation system without the proposed level diversity meets the minimum requirements of all applicable General Design Criteria (GDC).
However, in view of the importance of level instrumentation for safety in BWRs, and the experience discussed above where the potential existed to fail redundant level instruments due to a comon cause, the staff believes that the addition of level instrumentation which operates on a diverse physical principle is desirable and prudent for the purpose of guiding operator emergency actions.
Since the ABWR and SBWR are currently in the design stage, considerable flexibility exists in addressing this issue. Additionally, since these are advanced plants, a more robust solution to potential vessel level ina: curacies is appropriate.
Other evolutionary designs, such as ABB-Combuttion Engineering (CE) System 80+, provide diverse methods of RPV level measurement.
The ir. adequate core cooling instrumentation package in CE System 80+ includes reactor vessel level monitoring system probes employing both dp sensors and the heated.) unction thermocouple concept.
The staff is aware of a diverse i
method of level monitoring that is currently in use in at least one nuclear power plant in Germany employing ultrasonic measurement techniques.
In addition, a diverse level measurement system which uses heated junction thermocouples has been in use for the past five years at a Swedish BWR, and another Swedish BWR uses float switches for diverse level indication and automatic systems actuation. Other Swedish BWRs have decided in principle to install diverse level measurement systems.
GE does not agree with the staff recomendation ar.d has presented its position in a letter dated October 26, 1993. As part of the letter, GE presented the following sumary:
"ABWR water level instrumentation is rugged, simple and highly redundant for failure tolerance. All known operating problems have been addressed in this design and it is incredible to postulate simultaneous comon-mode failures which would yield identical errors in all the dp instrumentation.
Alternate technologies are unqualified for this application; further, there is no need to add this complexity, since the plant operating staff has ample additional indications of an impending 4
i
The Comissioners problem without relying solely on water level.
The EPGs direct the I
operator to use all information available to him and make. conservative i
(safe) decisions.'
In the attachment to the letter, GE also provided a list of indications of inadequate RP7 water level which are independent of the dp RPV water level instrumentation.
The staff recognizes that other parameters could aid the i
operator in assessing the adequacy of core cooling under accident conditions.
These include instrumentation for indication of reactor power, core neutron flux, the recirculation flow control system response, and feedwater flow and steam flow mismatch.
However, the staff believes that these indications could be easily misinterpreted or could be insufficient because they are only indirect methods of inferring reactor water level or core cooling.
The diverse method of level measurement is recomended for indication in the control roem only (there is diverse instrumentation, namely high drywell pressure, in both the operating BWRs and the ABWR design which provides diverse signals for automatic safety systems actuation for many event scenarios). This would provide a direct and back-up means for the operator to identify inadequate core cooling and to take appropriate manual actions to initiate and control safety systems as identified in the plant emergency operating procedures. The staff recomends that the diverse level measurement device be reliable, redundant, and capable of being powered by on-site power sources.
i As noted above, GE does not agree with the staff that diverse level indication should be required. However, during a meeting between the staff and GE on October 14, 1993, design requirements for diverse level indication, should it be required, were briefly discussed. Table 1, enclosed, provides a preliminary list of design requirements developed from this meeting.
This list requires significant development, and the staff is pursuing this further with GE.
However, the design requirements do illustrate the overall approach the staff believes to be appropriate for this diverse instrumentation.
j 1
SUMMARY
For advanced reactors like ABWR and SBWR, two independent and diverse methods of measuring the RPV level should be required because of the importance of RPV level instrumentation to BWRs and because operating experience has shown the potential for failure of redundant level instruments due to comon cause.
C00RDINATIQ3:
The Office of the General Counsel (OGC) has reviewed this paper and has no legal objection. OGC notes that Comission approval would be tentative, subject to further review in design certification rulemakings, and that comunications with GE regarding this Comission position should state this fact.
j j '.[ \\
. g :,, ; - a The Comissioners ;
RECOMMENDATION:
f i
The staff recomends that the Commission t
, t (1) ligig that this. draft paper will be placed into the public document room.
(2) fBL11 that the staff plans to discuss this position with the ACRS.
(3) fBtit that final Cosnission approval of the staff's position may be requested through the ABWR FSER review process.
James M. Taylor Executive Director for Operations
Enclosure:
As stated i
1 l
'l
l 1
i go TABLE 1 PRELIMINARY ABWR DIVERSE RPV LEVEL INSTRUMENTATION DESIGN REQUIREMENTS 1)
The system shall be highly reliable, Dut is not required to be safety grade or seismically qualified.
2)
The instrumentation will provide operator infomation only, including cor. trol room indication and alarus.
The control room indication and i
alarms could be provided on the operator console visual display units.
No automatic actuation functfons will be providec', and indication would not be provided on the r m te shutdown panel.
3)
The system shall be dua; ref ind2nt including separation of power supplies. The instrurer.id.on shall be capable of being powered by on-site power upon loss of offsite power, but it is not required that the i
power source be Class IE.
4)
The instrumentation shall be cepa'le of remaining functional during and u
following Design Basis Accident (LOCA) environments, but the requirements of the EQ rule (10 CFR 50.49) would not apply. Generic Letter 85-06, " Quality Assurance Guidance for ATWS Equipment that is not Safety-Related," should be used for guidance on survivability and quality assurance of the instrumentation.
5)
The level instrumentation shall be capable of measuring RPY water level from the top of active fuel to the emergency operating procedures control band limit, and could be either full range continuous indication or discrete position indication.
If the design involves discrete position indication, 3 or more level indications shall be provided with locations ranging from top of active fuel to the emergency operating procedures control band.
6)
Accuracy on the order of +/- 10% or +/- I ft and response time on the order of minutes may be acceptable depending upon the location and type of level measuring device used.
7)
The instrumentation is not required to be hardwired; the use of the plant multiplexer equipment would be acceptable.
To minimize additional vessel penetrations, existing instrumentation taps could be used where possible.
Measurement of water level outside the core shroud would be acceptable.
B)
The system would not be required to be included in the Technical Spect'ications, and the operability of the system would be controlled by the operational reliability assurance program (ORAP).
ENCLOSURE