ML20205C484
| ML20205C484 | |
| Person / Time | |
|---|---|
| Issue date: | 01/10/1987 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-2482, NUDOCS 8703300192 | |
| Download: ML20205C484 (11) | |
Text
1
~Y f f'
).
fw gup7 p ;
b!
I, d
DATE ISSUED: 1/10/87 PROPOSED MEETING
SUMMARY
FOR THE INSTRUMENTATION AND CONTROL SYSTEMS SUBCOMMITTEE MEETING ON DECEMBER 18, 1986 WASHINGTON, DC PURPOSE:
The Subcommittee on Instrumentation and Control Systems met on December 18, 1986 in Washington, DC, to discuss the effect of adverse conditions such as high temperature on solid-state components in nuclear power plants.
In addition, the NRC Staff briefed the Subcommittee on the reliability of reactor vessel level indication system (RVLIS) in the presence of the upper head injection (UHI) internals after the UHI accumulators have been discon-nected.
ATTENDEES:
ACRS NRC STAFF J. Ebersole, Chairman R. Hernan W. Kerr, Member D. Hood C. Michelson, Member J. Wilson D. Moeller, Member R. Karsch C. Wylie, Member M. Chiramal W. Lipinski, Consultant M. El-Zeftawy, ACRS Staff WESTINGHOUSE C. Sun, ACRS Fellow D. Adomaitis OTHERS P. Morris W. McCaughey (BG&E)
M. Patterson (BG&E)
R. Reeves (TVA)
A. Wyche (Bechtel)
J. Trotter (NUS)
MEETING HIGHLIGHTS, AGREEMENTS, AND RE0 VESTS:
1.
Mr. Ebersole, Subcommittee Chairman, introduced the members of the Subcommittee and its consultants and stated the purpose of the meeting.
II. During its 312th meeting, April 10-12, 1986, the ACRS reviewed the proposal by the Duke Power Company to remove the Upper Head Injection 33 2 870110 DESIGNATED ORIGINAL 2482 PDR
, N/2E Certified By
I&CS Meeting Minutes December 18, 1986 System (VHIS) portions of the emergency core cooling system at the McGuire Nuclear Station Units 1 and 2.
The UHIS was originally added to improve core cooling during large break LOCAs. The UHIS includes pressurized accumulators which provide a high volume flow of water directly into the upper head if the reactor coolant pressure decreases below approximately 1000 psig. During the discussion of this issue, Mr.
Michelson expressed some concern regarding the reactor vessel level indication system (RVLIS) that might be adversely affected during blowdown because of the presence of the UHIS internals after the UHIS accumulators have been disconnected.
III. Mr. Rudy Karsch, Reactor Systems Branch /PWR Licensing Div-NRR, described the RVLIS as a two-train differential pressure measurement system (see attached figure). On each train there are three AP transmitters. These AP transmitters are of differing ranges to cover differing flow behav-iors with and without pump operation as follows:
(AP ) Upper range - This transmitter is connected between the hot 3
legs and the top of the vessel and it is not to be used except during head venting. When the vessel is full and the RC pumps are off, the instrument indicates 100%. This unit indicates off scale in the 0% direction with vessel full and pumps on.
(AP ) Narrow range - This transmitter is connected between the b
vessel bottom and the hot legs and it is independent from the UHIS flow. This transmitter is set up to measure the collapsed level in the vessel (0% to 100%) with the pumps off.
(AP ) Wide range - This transmitter is connected between the vessel c
bottom and the hot legs and provides an indication of reactor core and internals pressure drop for any combination of operating RCPs.
This transmitter is scaled to read 0% vessel empty and pumps off,
I&CS Meeting Minutes December 18, 1986 and 100% with vessel full and all pumps on. With pumps off and vessel full, it reads 15%.
I The NRR Staff stated that the RVLIS is completely independent of the UHI internals and could be accurately used with either a UHI plant or a non-UHI plant. The RVLIS should perform the following functions accu-rately with the UHI accumulators being disconnected:
1)assistin detecting the presence of a gas bubble or void in the vessel, 2) assist indetectingtheapproachtoinadequatecorecooling(ICC),and3) indicate the formation of a void in the RCS during forced flow con-ditions. The NRC Staff indicated that the reactor operator is not instructed to put too much faith in the RVLIS information as a part of the recovery process.
IV. Mr. Matthew Chiramal, Office of Analysis and Evaluation of Operational Data (AE0D)/NRC, presented the case study report regarding the effects of ambient temperature on electronic equipment in safety-related in-l strumentation and control systems. He stated that the study was initi-ated by the event which occurred at the McGuire station. On June 4, 1984, while both units at the McGuire station were operating at 100%
power, a total loss of the control area ventilation system occurred for over 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />. As ne control room temperature increased, numerous j
alarms on Unit I high reactor coolant loop C T were received, as well av as alarms on Unit 1 pressurizer level. The operators had opened the
~
doors between the control room and the computer room, which still had cooling available. Operators also opened the doors of the W PCS 7300 cabinets, which contain the solid-state circuit cards generating the alarms. The licensee then used protable fans with ducting to provide cooling from the computer room to the PCS 7300 cabinets. Neither unit tripped during the event.
On November 19, 1985, an IE Information Notice No. 85-89 was issued to alert recipients of a potentially significant problem involving the loss
I&CS Meeting Minutes December 18, 1986 of solid-state instrumentation following the failure of control room cooling.
Three additional events were considered in the AE0D report:
- 1) Davis-Besse(September 21, 1982) -- failure of a power supply in an SFRCS cabinet caused a full trip of one channel, 2) Summer 1 (December 16, 1982) -- failure of an instrument loop power supply caused the indicator for reactor building pressure to fail low, and 3) Palo Verde 1 (December 16, 1985) -- failure of a fan in an ESF cabinet caused the overheating of the ESF load sequence module.
AE0D report indicated the safety significance of these events as:
- 1) Elevated temperatures in safety-related instrument cabinets cause decrease in the reliability of solid-state devices due to increased failure rate,
- 2) Loss or inadequate cooling can result in common cause failure of redundant instrumentation channels, and
- 3) Generic aspects of this problem has not been fully recognized by personnel at operating nuclear plants.
Operational experience shows that for instrument cabinets located in control room and auxiliary buildings, opening cabinet doors to allow heat dissipation may not maintain equipment in an acceptable temper-ature environment for the 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> and 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> duration of a pos-tulated station blackout.
The AE0D report recommended the following:
I&CS Meeting Minutes December 18, 1986 and for loss of forced cooling to instrument cabinets should be provided.
- 2) Training in the use of the procedures should be provided.
- 3) Supplemental cooling equipment should be made available.
- 4) All plants should periodically measure or continuously monitor the environmental condition inside the instrument cabinets.
- 5) The room ambient temperature specified in the technical spect-fications should reflect the actual measured temperature in the instrument cabinets located in the control room.
6)
In the resolution of USI A-44 (Station Blackout), evaluation of the design adequacy of I&C system equipment needed to function during recovery from a station blackout should be considered.
Y.
- 1) Mr. Ron Hernan, NRC/NRR, Technical and Support Branch, briefed the Subcommittee regarding the reactor vessel level indication for B&W plants. He indicated that for such plants the (ICC) instrumentation includes core exit thermocouples, hot-leg-level measure system, and subcooled margin monitors. B&W has a topical report on this issue and Mr. Hernan has agreed to send a copy of this report to the ACRS.
The NRC Staff has agreed to accept the B&W argument that the exist-ing instrumentation was satisfactory for detecting ICC in 1980 and 1981 time frame. However, after the TMI-2 event, the NRC has rejected to endorse the original submittal by the applicant for license to operate the Midland plant Units 1 and 2 (B&W plant) in response to item II.F.2 of NUREG-0737, " Clarification of TMI Action Plan Requirements." The NRC Staff requested more submittals to
ISCS Meeting Minutes December 18, 1986 include additional differential pressure instrumentation between the vessel head and the lower level of the hot leg.
The ACRS, in its letter of June 8, 1982, has also recommended that the applicant for the Midland plant to review the potential for providing indication of water content or level within the reactor vessel. Nevertheless, the operation of the Midland plant has been cancelled. Other plants like Oconee 1 has installed additional 4.P instrumentation between the vessel head and the lower level of the hot leg with no penetration from the botton of the vessel. Oconee 2 and 3 have been committed to do the same.
2)
Mr. Hernan stated that the AE0D report regarding the " effects of ambient temperature on electronic equipment in safety-related instrumentation and control systems" was issued on December 5, 1986 and has just been submitted to NRR. Mr. Denton will delegate the appropriate branch to review the report. NRR has recognized the issue and probably will take 4 to 6 months through the priori-tization process to reach a technical resolution of this problem.
The NRC Staff is also looking for industry input and may be EPRI or INP0 proposals on this issue.
VI. Mr. Dennis Adomaitis, Manager I&C Applications Engineering, E, itemized the solid-state systems that are in use:
- 1) 7300 process protection and control systems, 2) solid-state protection system, 3) nuclear instrumen-tation system, 4) digital rod position indication, and 5) rod control.
In addition to Mr. Adomaitis, Mr. Pete Morris Q), described the evolu-tion of E digital equipment. The development started in mid-70s. Late 70s, E obtained preliminary design approval for the integrated pro-tectionsystem(RESAR414)with8-bittechnology.
In 1981 E expanded the program to 16-bit technology, and in 1982 y initiated a program to apply programmable digital systems to operating plants. Ealsomen-tioned that the programmable digital (silicon base technology) systems
I&CS Meeting Minutes December 18, 1986 are receiving wide acceptance in non-nuclear applications due to the fact that they: a) improve accuracy, b) reduce heat load, c) do not require scaling, and d) reduce test and maintenance time through auto-matic testing (the testing includes seismic, environmental, noise and faulttests). One of g solid-state circuit cards testing includes a 12-hour period at 120"F, then an 8-hour period at 70 F, and then back at 120*F for another 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> duration. The operating temperature is designed to be 70 F 2 10 F.
Mr. Adomaitis indicated that following the McGuire event, the licensee took temperature measurements inside the PCS 7300 cabinets and determined that with an ambient temperature of about 72 F, the cabinets had internal temperatures of up to 125*F on the top rack. The McGuire operator's estimate of the ambient temperature, during the event, was 90 F before alternate cooling was provided.
Mr. Adomaitis stated that the normal range for circuit card failure rate is approximately 3 to 6% and E considers up to 10% failure rate to be acceptable, y has also recognized that some utilities added heat sinks withoutE'spriorknowledgeorrequesttodoso.
VII. As a result of the Subcommittee's discussion, the Subcomittee members raised some concerns regarding the following:
In regard to the RVLIS for UHI-plants, Mr. Ebersole requested from the NRC Staff to supply him with the emergency operating procedures (E0Ps) which include E guidelines to ensure that the utilities do not alter the procedure and include the RVLIS information as part of the recovery process.
Mr. Michelson expressed concern regarding the adequacy of B&W reactor vessel level indication especially in the existence of loop-to-loop oscillation, and the NRC Staff's decision of not to pursue this issue any further.
4
IACS Meeting Minutes December 18, 1986 Mr. Wylie has made the remark that Oconee 1 did connect the level instrumentation from the bottom of the vessel taping contrary to Mr.
Hernan's statement.
As a result of the McGuire event, though the licensee's remedial action to provide better normal cooling appear to have increased the reliability of the solid-state cabinets under design operating conditions, the safety concern following loss of all control room HVAC units still remains.
The Information Notice issued by the I&E office does not have any specific action nor does it require a written response from the licensees.
Mr. Wylie indicated that consideration should be given to the need to establish equipment qualification requirements for critical instruments even though they are located in mild environments.
Mr. Ebersole is concerned regarding the Palo Verde 1 event that a loss of instrument cabinet cooling has the pctential for both initiating a loss of all AC power to safety-related electrical buses and also preventing recovery from available AC power sources. This presents basic design flaws in the auxiliary control room.
Mr. Michelson expressed some concern regarding the fire protection instrumentation which may be affected by the rise in temperatures or cannot be restored after cooling down has been established and if this situation would constitute a licensing issue.
Mr. Ebersole is concerned regarding the lack of knowledge of the magnitude and margin of the solid-state components to malfunction.
He also indicated that a total loss of AC power does not appreciably diminish the heat loads in the control room.
I&CS Meeting Minutes December 18, 1986 Mr. Ebersole would like a PRA study to be performed which accounts for the effects of high temperatures on solid-state components.
Mr. Michelson would like a NPRDS data base search to be established to determine if more electronic component failures in IAC systems have occurred.
Mr. Wylie expressed some concern since the technical specifications limit for the control area is 120 F at McGuire and the maximum ambient temperature in the control room during the event was 90*F.
Thus, significant de0radation of the system occurred even though the ambient temperature during the event was well below the design and technical specification limits.
Dr. Moeller requested an approximate number of how many circuit cards at McGuire and how many of them have failed due to the temper-ature rise. W's response was approximately 500 cards and 70 cards have failed. Dr. Moeller indicated that this constitute more than 10% failure rate by which W has claimed to be the acceptable limit.
Mr. Michelson indicated that he would like to have some assurance if the control room would burn down to a cinder, the plant could still be shut down.
Outcome The Subcommittee Chairman has decided to give a short briefing to the full Committee concerning Subcommittee activities at the January 1987(321st)ACRS meeting.
NOTE:
Additional meeting details can be obtained from a transcript of this
o 18CS Meeting Minutes December 18, 1986 meeting available in the NRC Public Document Room,1717 H Street, NW, Washington, DC, or can be purchased from ACE-Federal Reporters, 444 North Capitol Street, Washington, DC 20001, (202) 347-3700.
- \\s b
ORNL-DWG 82-10182 WARE HEAD PENETRATION l
8
-A L-@
h
rs gg 9
9 9
9 e
b a
b ggg CDRE NARROW RANGE I
TRAIN A TRAIN 3
.i Reactor vessel level instrument system for UHI plant.
1 L
. - -. _ -_