ML20045D074
| ML20045D074 | |
| Person / Time | |
|---|---|
| Issue date: | 12/10/1992 |
| From: | Michelson C Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| REF-GTECI-106, REF-GTECI-NI, TASK-106, TASK-OR ACRS-2833, NUDOCS 9306250302 | |
| Download: ML20045D074 (11) | |
Text
{{#Wiki_filter:. 4 BMS-4f33 ,, y fW blAll93 e na -CERTIFIED BY: I l[ 'C..Michelson 12/10/92 DATE ISSUED: 12/10/92
SUMMARY
/ MINUTES OF THE ACRS SUBCOMMITTEE MEETING ON AUXILIARY AND SECONDARY SYSTEMS JULY 8, 1992 BETHESDA, MARYLAND e
5 INTRODUCTION The Auxiliary and Secondary Systems Subcommittee met on Wednesday, July 8,
- 1992, in Room P-110, 7920 Norfolk Avenue,
- Bethesda, Maryland.
The purpose of the meeting was to discuss the proposed resolution of Generic Issue -106, Piping and the Use of. Highly Combustible Gases in Vital' Areas. The entire meeting was open to public attendance. Mr. Herman - Alderman was the Cognizant ACRS Staff Engineer for this meeting. A copy of the presentation schedule for the meeting is included in Attachment A, and a list of documents submitted to the Subcommittee is included in Attachment B. The meeting was convened at 8:30 a.m.'and adjourned at 3:40 p.m. Attendees: Principal meeting attendees' included: ACRS C. Michelson,' Subcommittee Chairman I. Catton, Member C. Wylie, Member J. Carroll, Member P. Shewmon, Member H. Alderman, Cognizant ACRS Staff Engineer i Princinal Sceakers Idaho National Encineerina Laboratory. (INEL) J. Bickel G. Simion W. W. Madsen RPK Structural MechaDics Consultinc R. P. Kennedy U.S.-Nuclear Reculatory Commission C. Graves G. Mazetis K. Kniel r DCIGNAIED ORIGINAL - 930$250 2 21210 CecMfied B7 -PDR ACRS 2833 PDR i
, Aux.'& Second. Sys. Subcte. 2 July 8, 1992 OPENING REMARKS Mr. Michelson convened the meeting at 8:30 a.m., reviewed briefly the schedule for the meeting and stated that the subcommittee had received neither written comments nor requests for time to make oral statements from members of the public.- He remarked that it was a fine example of some good work on the combustion and explosion potential of hydrogen gas that was done in conjunction with identifying what the problems might be with such events for a nuclear power plant. However, he noted that he had some problems with the proposed resolution of Generic Issue 106. One is why the generic letter relates only to pressurized water reactors since boiling water reactors have a large amount of hydrogen present. The other problem is why there j no guidance on how this issue is resolved for evolutionary plants. Mr. Carroll noted that he would like to understand how the INEL report relates to Licensing Issue 136 (hydrogen water chemistry) and Generic Issue 106. He also said that he thought the discussion on the resolution of Generic Issue 106 should include information on how to fight hydrogen fires. PRESENTATION BY THE NRC STAFF - Mr. C. Graves. NRC Introduction Mr. Graves said that this issue first came up concerning hydrogen use for the volume control tank (VCT). There was also a concern about the use of hydrogen detectors. The Standard Review Plan 9.5.1, on fire protection, was revised in 1981 to include guidance on the use of hydrogen in nuclear plants. This was the first time that anything was said about limiting the flow of hydrogen into the auxiliary building. In 1986, the issue of hydrogen water chemistry installations at BWRs came up. During the staff review of this issue, Electric Power Research Institute (EPRI) produced a topical ~ report (EPRI NP-SR-A) providing guidance for the hydrogen storage facility and the distribution system. The separation distance guidelines from this report were used as an initial screening mechanism during plant surveys made in the study to resolve GI-106. In the acceptance letter related to the EPRI topical report, the staff suggested that EPRI modify the guidelines to extend them'to the volume control tank application and the generator application. This was not done. Scope The scope of GI-106 was expanded in 1989 to include both BWRs and PWRs. The expanded scope included risk from: The storage and distribution of hydrogen for the VCT in PWRs and main generators in PWRs and BWRs.
s Aux,'& Second. Sys. Subcte. 3 July 8, 1992 other sources of hydrogen such as battery rooms, the waste gas e system in PWRs, off-gas systems in BWRs. Small quanti. ties of hydrogen and other combustible gases such as prc, pane and acetylene that would be used in maintenance, testing and calibration. Mr. Graves said that results of the scoping study for GI-106 revealed that there was a very small risk from the battery rooms, the PWR gaseous waste systems, the BWR off-gas systems,-and the portable bottles. Mr. Carroll said he agreed with_everything on the list except the BWR off-gas system. Mr. Graves said that the technical specifications have strict requirements to keep the hydrogen and oxygen concentrations well below allowable limits. Mr. Carroll cited an example of the Big Rock Point Plant where they had an explosion in the off-gas system. Mr. Michelson pointed out that the Subcommittee was trying to determine whether or not the BWRs have problems. Mr. Michelson said that some of the battery rooms vented to the atmosphere while others vented into free flowing air-conditioning systems. He asked if there was any problems in either of these cases. Mr. Graves said that INEL representatives will address this issue during their presentation. Ma-ior Concerns Mr. Graves said that one of the major concerns was a.large hydrogen release and a resulting large hydrogen deflagration or a detonation. _This would be a large release in a single event. Another scenario would be' hydrogen build up resulting from small leaks to an unacceptable level. Mr. Carroll said that he would add a third item, the potential for detonation resulting from improper fire-fighting of a hydrogen leak. Mr. Graves said that the only safe way to stop a hydrogen fire is to stop the flow of hydrogen. Mr. Graves said the main problem is that a non-seismic or seismically induced failure of a hydrogen rystem Nuld causes a hydrogen release which, in turn, could cause a detonation or a deflagration resulting in loss of safety-related equipment and leading to a potential core damage. Mr. Michelson asked how many hydrogen systems are seismically qualified. Mr. Bickel, INEL, said that some portions of the systems were seismically qualified and other portions were just field run pipes. Mr. Michelson asked what was the staff position concerning the failure of small-bore non-seismically qualified piping during an earthquake. Mr. Graves replied that Regulatory Guide 129 states that if a failure of a non-safety related system could cause damage
1 Aux '& Second. Sys. Subcte. 4 July 8, 1992 i i to a safety-related system, then that non-safety related system should be designed so its failure would not cause the damage during a seismic event. Mr. Bickel, INEL, said that the scope of their work was not to assess how sturdy this piping was. The scope was i to assess the effects of hydrogen once it gets out. 1 Mr. Simion, INEL, said that when they integrated the fragility of the pipe with the seismic hazard curves, the initiating event frequency was about ten to the minus five. For this accident scenario, the calculated core-damage frequency was lower than the core-damage frequencies calculated to result from non-seismic hydrogen event interactions. Resolution Mr. Graves discussed several ways to minimize the potential damage from hydrogen detonation and deflagration. One possible approach is to limit the hydrogen flow rate from the storage facility to the i postulated break so that the average hydrogen concentration near the break will be below the lower flammability limits. This can be accomplished with a restricting orifice, excess flow check valve, or a valve controlled by a flow element. The amount of hydrogen released in a single event can be limited by limiting the amount of hydrogen in the system. Another method of control is to operate with the system normally isolated and manually add hydrogen as needed. Mr. Michelson asked if the generic letter requires testing of the flow control valves. Mr. Graves said testing is included. Mr. Michelson expressed his concern that the resolution of GI-106 concentrates on PWRs. Mr. Graves said that results of the scoping study showed no safety-related equipment in the turbine building of BWRs, and consequently there is a negligible risk for BWRs. Both Mr. Michelson and Mr. Carroll said that some BWRs have safety-related equipment in the turbine buildings. Hydrocen Detonation Study: Mr. W. Madsen. INEL Mr. Madsen said that he was involved in blast wave work for the Department of Defense. He was asked to review the INEL study involving detonation. The first thing he noticed in the study was that they were using TNT equivalences, which he said you cannot use for any kind of a gas. He pointed out that they used a computer j code called COMBWAT. The computer code computes data for the combustion process. The code then calculates an incident wave and attenuates the wave with distance. The program computes time histories, pressures, temperatures, and velocities for that wave. A second wave follows the first wave, and the program calculates this also. The code computes the reflected wave and the time
Aux.'& Second. Sys. Subcte. 5 July 8, 1992 history, pressures, temperatures, and velocities associated with this. Mr. Madsen said most of his calculations were based on open spaces. Dr. Catton asked about a detonation inside a building. Mr. Bickel, INEL, said for a detonation inside a building, they assumed it would be catastrophic. Any safety-related equipment in the building would not be available. Mr. Madsen said that in a confined space, if the concentration was above 12 percent mole fraction, a deflagration would turn into a detonation. Mr. Michelson asked about the detonation pressure range. Mr. Madsen said that they could be as high as 400 psi. Mr. Bickel, INEL, said that if you had a large release in the turbine hall that detonated, there would be massive damage to the building. Most of the PWRs have their decay heat removal systems, such as auxiliary feedwater, in separate buildings. Some have other systems, such as feed and bleed. These systems will allow shutdown under these conditions. Of major concern are the plants that don't have feed and bleed as an alternative. decay heat removal. scheme. Some of the plants have their entire auxiliary feed water system located at the very lowest level of the turbine building. These plants are vulnerable and something should be done about them. Mr. Carroll asked whether the scoping study report had been provided to EPRI. Mr. Graves said that it was not included as part of the resolution of GI-106. He said the report was to have a peer review prior to release. Mr. Michelson asked what the utility should do when it receives the generic letter. Mr. Graves said, in his opinion, they should look at the location of the hydrogen lines and determine where equipment is located that could result in core damage upon loss of the equipment. They should limit the flow of hydrogen to a break. It would be easier to prevent a hydrogen detonation. He said that they have limited the resolution to the distribution lines to the auxiliary and turbine buildings. In response to several Subcommittee comments regarding the resolution of the concerns of GI-106 for evolutionary plants, Mr. Graves read a statement that was recommended for evolutionary. plants: " Safety-related equipment should not be located in the turbine building because of the hazards associated with hydrogen fires and explosions and large oil
- fires, and the large uncertainties in estimating the consequences."
Mr. Michelson expressed his concerns, stating that if you postulate detonations in the turbine building that can destroy the barrier between the turbine building and the reactor building or the turbine building and the auxiliary building, you have to go back and fix the problem. He said he didn't think you can concede the
Aux. & Second. Sys. Subete. 6 July 8, 1992 loss of the barrier between the safety-related and non-safety related areas. He said that this wasn't addressed in the generic letter. He also pointed out that the concerns also pertain to BWRs and the generic letter should cover BWRs. Mr. Carroll asked if there ever was a hydrogen explosion in a power plant. Mr.
- Simion, INEL, said there hasn't been a hydrogen explosion of the magnitude they assumed in their study.
Bui_ldina Response Calculations: Mr. R. P. Kennedy RPK Structural Mechanics Consultina, Yorba Linda, California Mr. Kennedy discussed the results of a study that was performed on ) the required separation distance to prevent unacceptable damage to concrete walls from hydrogen detonation from external gaseous hydrogen storage. He said that the study was performed assuming an external detonation with no intervening barriers between the detonation point and the wall. Intervening barriers would reduce the impact on the wall and reduce the damage. The study deals with the waves hitting the wall normal which is the worst possible angle that they could hit the walls. l l He discussed the differences between an external unconfined detonation (free field) and a confined detonation. The experience has been that in a free-field detonation about 5 to 28 percent of the hydrogen involved in the detonation, with most cases being closer to 5 percent than to the 28 percent. For a confined detonation, about 30 to 90 percent of the total hydrogen involved in the detonation. The confinement will result in higher pressures. The walls will cause reflections, and the reflected waves will create impulses that are larger than those resulting from a free-field detonation. For the hypothetical detonation inside a commercial nuclear plant, the internal walls have less capability of withstanding the effects of detonations than do the external walls. He said that due to the issues he just discussed, the separation distances that he will present for free-field detonations on exterior walls are likely to be significantly unconservative in most cases for interior detonations. Mr. Kennedy mentioned two definitions used in his study. The first is separation distance, which is the distance from the wall to the point of detonation. The second is detonation volume which is the volume of hydrogen times the fraction that involved in the detonation. He presented some of the results of the study. For an 18 inch thick wall, with reinforcing steel of about one half of one percent, for a 300 cubic feet detonation volume, the separation distance is 15 feet. Assuming the same parameters, with a detonation volume of 740 cubic feet, the separation distance is 30 feet. With detonation volumes of 1500 cubic feet and 4500 cubic feet, the separation distances are 63 and 172 feet, respectively.
Aux, & Second. Sys. Subcte. 7 ' July 8, 1992 Mr. Kennedy pointed out the influence of wall thickness. Generally, the results vary directly with wall thickness. If the wall thickness is doubled, the separation distance drops by one half. The separation distance is cut appreciably by the amount of reinforcing steel in the wall. He noted that comparison with the. EPRI study on gaseous hydrogen storage reveals that the EPRI study is unconservative. Scopina Analysis for Generic Issue 106: Mr. J. Bickel Idaho National Enaineerina Laboratorv Mr. Bickel said that the objectives of their study was to identify the sources of highly combustible gases in PWRs and BWRs and the types of risks associated with these combustible gases. To accomplish the objectives, they did a very detailed analysis on a Westinghouse plant that was licensed prior to the adoption of SRP 9.5.1. Af ter completing the detailed plant study, they studied the effects of GI-106 on other plant types including BWRs. He listed the concerns for PWRs: Some plants have the auxiliary feedwater systems located in the turbine building where they are vulnerable to hydrogen 4 explosions. The CE system 80 plant doesn't have feed and bleed capability. e The older CE plants have very marginal feed and bleed capability. j Some plants have component cooling water systems located in e the turbine building where they are vulnerable to hydrogen i explosions. Some plants have diesels located in the turbine building. Mr. Bickel said that in the course of their study, they found a number of plants that did not meet the EPRI criteria on the i location of the hydrogen tank farms with respect to distance from adjacent buildings. As a result of further' analysis, some non-conservatisms were identified and this area should be pursued. Mr. Bickel said they had narrowed their areas of concern to transients with the additional loss of decay heat removal systems and transients which induce LOCAs. A postulated scenario'might be a hydrogen release due to a passive failure, such as a line or seal failure and hydrogen release. Subsequently, a plant transient damages other equipment. Alternately, a seismically induced event shakes and fails hydrogen piping or components, releasing hydrogen and causing a transient. Core damage will not occur unless. decay heat removal is lost.
Aux. & Second. Sys. Subcte. 8 July 8, 1992 With a pressurized water reactor, if the decay heat removal system is lost, the auxiliary feedwater system can be used. In the unlikely loss of auxiliary feedwater system, a plant can resort to feed and bleed operations. Some of the older plants have the auxiliary feedwater systems located in the turbine building. The newer CE System 80 plants do not have feed and bleed capability and some of the older CE plants have marginal feed and bleed capability. The older CE plants have small power-operated relief valves (PORVs) and positive displacement pumps. To utilize feed and bleed in such a plant, the operators would have to act quickly if there was a failure of the auxiliary feedwater system. Mr. Bickel pointed out that for BWRs, the concern about loss of decay heat isn't as significant as for PWRs due to the large number of decay heat removal options in BWRs: water can be added to the reactors in a number of ways. Mr. Michelson said that he didn't think the BWRs were a non-problem and he thought the BWRs should be mentioned in the generic letter. Analysis Methods: Mr. Simion, INEL Mr. Simion discussed some of the aspects of the analysis that was performed by INEL in support of the resolution of GI-106 on a representative Westinghouse PWR and on a BWR. Considerations included the vintage of the plant and the availability of a probabilistic risk assessment, a fire hazards analysis, and an external events analysis. They identified pertinent fire zones containing combustible gases, especial'y hydrogen, and identified the types of safety equipment in those zones and in adjacent zones. The next step was to identify the initiating event frequencies, based on a licensee event report review and nuclear power plant experience review of data on hydrogen events. For the period of 1969 to 1989, there were a total of 83 hydrogen events, 45 at BWRs and 38 at PWRs. The events at BWRs were primarily related to the of f-gas systems. The number of these events decreased dramatically following the 1980 implementation of the radioactive effluent technical specification. Mr. Simion presented the results of a hypothetical scenario involving a hydrogen fire or explosion in a turbine building with subsequent loss of decay heat renoval. It is assumed that the auxiliary feedwater system, located in the turbine building, is lost. He reported the core damage frequency for plants with feed and bleed capability as 6.9 x 10-7 events per reactor year (RY). For plants without bleed and feed capability, he said the core
Aux, & Second. Sys. Subcte. 9 July 8, 1992 damage. frequency (CDF) is 9.5 x 10-6/RY. He indicated that it would be better to prevent the hydrogen accumulation and detonation than to prevent the subsequent damage from one of these events. Mr. Simion said they considered nine alternatives to resolve these problems and calculated the costs associated with these alter-natives. The cost-benefit analysis resulted in two suggested alternatives that were cost effective. The first was installation of excess flow check valves. The second was manual operation of the hydrogen supply systems. t
== Conclusions:== Mr. Graves, RES Mr. Graves said that based on the INEL studies, the BWR risks were low. Based on the low risks, they concluded that the NRC didn't have any basis to require BWRs to respond to the generic letter. He said that in the case of PWRs, there were fixes that were cost effective and simple to apply. He noted that as a result of the detonation calculations, the tank farm issue will have to be revisited. He said that they are aware of the increased significance of hydrogen detonations within buildings. Mr. Michelson said that they hadn't looked at the barrier wall protecting the reactor building against explosions in the turbine building. Mr. Graves said there wasn't a detailed evaluation of the impact of a detonation on that wall. Mr. Michelson asked if i the staff would be satisfied if a utility said that they would j install the check valves and not do the analysis. Mr. Kniel said that would satisfy them. ACTIONS, AGREEMENTS, REOUESTS The Subcommittee decided to submit a proposed letter on this topic to the full Committee for consideration, and to have the Committee briefed on this subject during the July ACRS meeting. NOTE: Owing to lack of time, the letter on GI-106 was postponed from the July ACRS meeting to the August ACRS meeting. NOTE: Additional meeting details can be obtained from a transcript of this meeting available-in the NRC Public Document Room, 2120 L Street, NW, Washington, DC 20006, (202) 634-3273, or can be purchased from Ann Riley and Associates, Ltd., 1612 K
- Street, NW, Suite
-300, Washington, DC 20006, (202) 293-3950.
c / ~ ATTACINDTI A [ ADVISORY' COMMITTEE ON REACTOR SAFEGUARDS' i SECONDARY-AND AUXILIARY. SYSTEMS' SUBCOMMITTEE 7920 NORP7LK AVENUE, ROOM P-110- .OTLT 8, 1992. BETHESDA,-MARYLAND - TENTATIVE SCREDULE - i Introductory Remarks by Chairman 10-Min 8:30 a.m. - C. Michelson, ACRS: t Introduction and Resolution of G.I.-106 60 Min 8:40 a.m. - C. Graves + Detonation Calculations 30 Min .9:40 a.m.. - W. Madsen
- * *
- BREAK * * * *
-15 Min 10:10 a.m. Building Response Calculations 40 Min-- 10:25 a.m. .R. Kennedy Cost-Benefit Analysis Methods 40 Min 11:05 a.m. - J.-Bickel' Closing Remarks and Discussion 15 Min 11:45 'a.m. - INEL and C. Graves' l .. Adjournment '12:00 p.m.. 'l q 4
_ _ _ _ _ _ = _ - _ _ _ _ -t ^. ~ ATTACHMENT B - Documents Provided to the Subcommittee 1. EGG-SSRE-9747, Improved Estimates of Separation Distances to Prevent Unacceptable Damage to Nuclear Power Plant Structures From Hydrogen Detonation for Gaseous Hydrogen Storage 2. NUREG-1364, Regulatory Analysis for the Resolution of Generic Issue 106: Piping and the Use of Highly Combustible Gases in Vital Areas 3. NUREG/Cr-5759, Risk Analysis of Highly Combustible Gas Storage, Supply, and Distribution Systems in Pressurized water l-Reactor Plants 4. Proposed Resolution of Generic Issue 106, " Piping and the Use of Highly Combustible Gases in Vital Areas"--Staff Presentation 5. Generic
- Letter, Request for Information. Related to the Resolution of Generic Safety Issue 106" Piping and the Use of Highly Combustible Gases In Vital areas," Pursuant To 10 CFR 50.54(f)-Generic Letter 92-xx 6.
Computer Code : COMBWAT 7. Separation Distances to Prevent Unacceptable Damage to Concrete Walls From Hydrogen Detonation for External Gaseous Hydrogen Storage-Robert P. Kennedy 8. Highly Combustible Gases in Vital Areas-INELNUREG-1364, Regulatory Analysis for the Resolution of Generic Issue 106: Piping and the Use of Highly Combustible Gases in Vital Areas -{ -( _ - _ - - - _ - _ _ _ _ _ _ _ _ _ _ _ -. _ _ _ _ _ - _ _ _}}