ML19347D879
| ML19347D879 | |
| Person / Time | |
|---|---|
| Site: | Browns Ferry  |
| Issue date: | 04/09/1981 |
| From: | Mills L TENNESSEE VALLEY AUTHORITY |
| To: | Harold Denton Office of Nuclear Reactor Regulation |
| References | |
| RTR-NUREG-0737, TASK-2.F.1, TASK-2.K.3.16, TASK-2.K.3.18, TASK-2.K.3.27, TASK-TM NUDOCS 8104140381 | |
| Download: ML19347D879 (7) | |
Text
..
TENNESSEE VALLEY AUTHORITY CH AT TANOOG A, TENN ESSEE 37401 400 Chontnut Street 'Dtur II April 9, 1981 N%
[gf
~
s D*
[6-Mr. IMrold R. Denten, Director TJT 1 ggBl* [3 Office d Muclear Peactor Faqulatten p on U.S. Nucient "nJulatory Cm mission
\\f. o.s.fjdse p
washington, DC 20555
Dear Mr. Denton:
[7%>
In tM Matter of tM Apolication of
)
Ikcket Noc. 50-259 Tennessee Valley Authority 50-260 50-296 In re= pone to D. G. Eisenhut'n CetcNr 31, 1980 letter to All L!censees of Opart. ting Plants rmarding post-NT re.piremnts, 'IVA provided an initial sutnittal to the NTC for the Browns Ferr'f Nuclear Plant by letter frcm L. 31. Mills to H. R. Ihnten detM Dec*r 23, 1930. Dv:1ccuro 1 centnins revisM rmrnnem for P. MEG-0737, itcw, II.F.1.1 ('hble Effit'ent ?"oniter) and II.K.3.27 (Cm Pcference IcVel for Vernel IcVel Tn9truxntation). Our response to these itses has becn discussed with your staff.
Also, as requested by D. G. Eisenkut's letter, previded in Drlccure 2 are the results of cor fessibility studien for NUrdTr-0737, items II.K.3.16 (P v3tetien of Challenges and Failures of Relief Valven) an1 II.K.3.18 Styliffcation of ADG I.ogic).
Veey truly yours, 71* Jim '3IIRI NITEPI*I 1Y k
L. Fi. Mills, enMer Nuclear Pwaletion arvl Safety Ssrrn ts $ :uh cr bed. fere re nn this dA day of 1981 ibt.m Public
~
My Cctrnissicn Expires I-Enclosures P00R ORIGINAL 8 1 0.4 1 4 0 0 FI
^"'* " """* '""'""'
ENCLOSURE 1 e
REVISED RESPONSE FOR NUREG-0737 ITEMS II.F.1.1 AND II.K.3.27 BROWNS FERRY NUCLEAR PLANT NUREG-0737, Item II.F.1.1_
- Noble Gas Effluent Monitor Revised TVA Response The design for this modification has been completed. However, the final design drawings for this modification will not be issued until August 1, 1981. We expect to have the modifications completed by January 1, 1982, although some "nonclass IE" effluent elements and monitors will be utilized for an interim period. We are making every effort to upgrade those "nonclass IE" effluent elements and monitor to Class IE but this is dependent on availability of the equipment.
NUREG-0737, Item II.K.3.27 - Coumon Reference Level TVA Response TVA has reconsidered its earlier position on item II.K.3.27.
We agree to provide a common reference level point for level instrumentation. We anticipate that either the bottom of the vessel or the top of the active fuel will be.used as a reference point. We will utilize the same reference point for all three units. We expect to complete this modification by July 1, 1981.
However, this modification will consist of installation of visual indication to level instrumentation (for example, tape or markers).
We do not plan, nor do we see the necessity for, procedural changes (operation, test, or maintenance procedures) or technical specification changes. We also believe that our operators are sufficiently trained to incorporate this minor change without additional formalized training. This modification change was discussed with D. Vare111 of your staff in a conference call on April 9, 1981.
It is our understanding that Mr. Varelli found this approach to be acceptable. We also plan to continue to utilize a vessel level instrument range chart which is posted on the control pannel and allows the operator to immediately interrelate any level measurement. Reference to our use of this chart was made in our i
June 23, 1980, response on this item.
It is our understanding that this approach satisfies the intent of this item for NUREG-0737.
v.
,-7.
~,,
I ENCLOSURE 2 i
i NUREG-0737 REQUIREMENTS BROWNS FERRY NUCLEAR PLANT 1
Item II.K.3.16 - Reduction of Challenges and Failures of Relief Valves -
Feasibility Study and System Modification TVA Response - TVA has participated in the BWR Owners' Group study of this item and agrees with the general conclusions of the report. While acknowledging that the consequences of stuck-open relief valve (SORV) transients are not a safety concern and reactor shutdown is uncomplicated as proven by numerous field occurrences, TVA is sensitive to the need for avoiding SORV's.
In fact, we have expended considerable effort as discussed below to improve the performance of S/RV's at Browns Ferry t
Nuclear Plant. The most significant change was switching from the Three Stage Target Rock valve to the Two Stage version. Since the changeover, valve performance regarding SORV events has improved considerably.
TVA comments on the NRC suggested methods to reduce SORV's as applied to Browns Ferry are as follows:
1.
Additional anticipatory scram on loss of feedwater.
a Since the vessel level scram is only 20 inches below normal operating level, we do not believe this proposed change is worthwhile.
Furthermore, contrary to NUREG-0626, this modification does not reduce pressurization rates.
2.
Revised relief valve actuation setpoints.
We agree that raising the S/RV setpoints is of benefit. At Browns Ferry, the lowest S/RV setpoint has been raised to 1,105 psig. The principal advantage is that the " simmering" problems which have contributed to observed valve failures are alleviated. This j
methodology has eventual limitations in that peak vessel pressure is also increased as the relief valve setpoint increases. A related modification was also performed which increased the bore size of I
the valves, thus allowing the lower setpoint valves to accommodate more relief capacity.
j 3.
Increased emergency core cooling systems flow.
The high pressure injection system capacity is fully adequate.
Problems occur when the systems fail to operate.
Effort is being expended to increase the reliability of HPCI and RCIC.
4.
Lower operating pressure.
The less in thermodynamic ef ficiency of the plant is too great to consider lowering primary system pressure.
L' l
. 5.
Early initiation of ECC systems.
Relief valve forced actuation is caused by only two events.
A.
Turbine stop/ control valve closure.
B.
Vessel isolation (MSIV closure).
Relief valve actuations caused by turbine events are too fast to be assisted by early initiation of high pressure systems. The same is true for spurious vessel isolations. For events such as loss of feedwater when it is evident that the vessel will ultimately isolate, the operator will manually start the high pressure systems prior to their automatic initiation setpoint; therefore, we do not see much benefit in this proposal.
6.
Heat removal through emergency condensers.
This modification is physically impractical for Browns Ferry.
7.
Offset valve setpoints to open fewer valves per challenge.
We have aircady accomplished this modification at Browns Ferry. We have three groups of valves at 1,105 psig, 1,115 psig, and 1,125 psig.
8.
Install block valves.
The BWR Owners' Group position addresses this item well. We are in agreement with that position.
9.
Increasing high steam line flow setpoint.
In practice, spurious isolations due to exceeding the high steam line flow setpoint have not been a problem.
10.
Lowering pressure setpoir.; for MSIV closure.
The main steam line low pressure isolation setpoint has been previously lowered to 825 psig. This change has eliminated spurious isolations due to low pressure.
No benefit is anticipated by further reductions.
11.
Reducing testing frequency for MSIV's We have reviewed the challenges to S/RV's as a result of personnel error or equipment problems during testing. Three types of testing are implicated:
A.
MSIV operability tests.
B.
Tests associated with MSIV isolation logic.
C.
Turbine and associated trip logic testing.
_3_
Testing appears to be a relatively large contributor to total S/RV challenges, mainly because of the large number of tests associated with the three areas above. We believe that the testing frequency could be reduced without impacting overall system reliability. TVA will evaluate the present testing frequency and propose changes.
Technical specification revisions will be necessary in some cases.
12 and 13.
More stringent valve leakage criteria and early removal of leaking valves.
Valve leakage was a major contributor to spurious openings of the Three Stage Target Rock valve.
The Two Stage valve is much less susceptible to this phenomena. We do, however, believe that a vigorous maintenance program contributes to the overall performance of the S/RV's.
Presently at Browns Ferry, each main steam safety relief valve is removed and steam bench tested following each fuel cycle. This program goes beyond the present requirements in the technical specifications and Section XI of the ASME code. The testing pro-cedures now include defined criteria for set pressure and seat leakage to determine valve acceptability; however, the test pro-cedure-also require the monitoring and recording of valve stroke time, disc 1;it, and blowdown reseat pressure. These parameters are also evaluated to determine proper valve operation.
Steam bench testing is also required following any activity that will affect valve operability or set pressure prior to installing the valve for service. During unit operation, discharge tailpipe temperatures are monitored on a weekly basis and evaluated to determine if the valves are operating satisfactorily. The results of this evaluation may recommend that a valve be removed from service and replaced at the earliest convenient opportunity to preclude any malfunction. This has been done several times in the past when we observed problems with the Three Stage valve.
Other modifications which TVA has implemented with some benefit in reducing S/RV challenges are a= follows:
1.
Reduce vessel level setpoint for MSIV isolation.
The vessel level setpoint for MSIV isolation may be reduced, which reduces vessel isolations. The philosophy for the change is that following reactor transients and scram, a certain amount of decay I
heat steam must be vented to limit vessel pressure. This steam will be routed to the condenser via the bypass system if the vessel is not isolated.
If the condenser is not available (vessel isolated),
then the steam must be relieved through the S/RV's. The total steam flow rate either to the condenser or S/RV's is about equivalent, so vessel water inventory depletion rate is approximately the same in each case.
TVA lowered the vessel level setpoint by 20 inches, which signifi-cantly reduced vessel isolations, especially those caused by turbo-l generator trips. This change also allows the operator more time to i
i i
manually start high pressure injection pumps or regain lost feedpumps prior.to reaching the MS1V isolation setpoint. TVA will investigate if further reduction of the level setpoint is beneficial.
2.
Equivalent manual action to " Low-Low Set" feature.
The operating instructions for isolated vessel events call for manual operation of relief valves to prevent second and subsequent lifts of multiple relief valves.
3.
Recirculation pump trip.
This modification, which trips the recirculation pumps during turbo-generator trips, reduces peak vessel pressure and also assists in water level control.
In summary, we believe that we have already implemented the features which have major impact on reducing S/RV challenges and SORV events. Our experience to date with the Two Stage Target Rock valve has been good; Long-tcrm efforts should be aimed at improving the reliability of ths auxiliary systems which generally are,the initiating failures for transient events.
b
. ;~
Item II.K.3.18 Modification of Automatic Depressurization Logic TVA Response - TVA has participated in the BWR Owners' Group study of the item and agrees with general conclusions of the report. We do not, however, believe that the safety benefits of the proposed modifications necessarily offset the decreased reliability of the ADS circuitry due to increased complexity and the risks incurred with higher probability of inadvertent actuation of the ADS.
The proposed modifications also conflict with our general operating philosophy which reserves automatic functions for those situations where the operator cannot be expected to act reliably due to time or complexity constraints.
bote that all events which require manual ADS are slow, uncomplicated, and well understood, for which the operator is well trained and familiar with both the equipment and systems response. Because of these considerations, operator reliability is very high.
In addition, the current design allows operator flexibility in the event of unforeseen circumstances or plant conditions.
The new symptom-oriented guidelines clearly emphasize the situations when manual ADS is required, yet also have contingencies for unanticipated occurrences. Also, for future reference, note that the proposed modifications do not appear consistent with potential ATRS mitigating systems.
Based on the above discussion and the BWR Owners' Group report, TVA prefers to retain the current ADS logic system.
f
/
l
/