ML20215N508
| ML20215N508 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 10/09/1986 |
| From: | Cloutier W, Klein A, Pepe D PUBLIC SERVICE CO. OF NEW HAMPSHIRE |
| To: | |
| Shared Package | |
| ML20215N505 | List: |
| References | |
| YAEC-1571, NUDOCS 8611060150 | |
| Download: ML20215N508 (9) | |
Text
i
'M 0
1:1 s
liAZARDS ANALYSES OF SEABROOK STATION
-CilARC0AL FILTER UNITS Seabrook -S tation Public Service Company of New Hampshire New Hampshire Yankee Division October 3, 1986 Prepared b , N W V
'b David M. Pepe Date
' Reviewed by - Mww lo. F. E Alethnder T. Klein Date-Approved by /6!G/hEl 441 data J. Cloutier [pfe Yankee Atomic E1cetric Company Nuclear Services Division 1671 Worcester Road Framinghan, Massachusetts 01701 0611060150 861009 PDR ADOCK 05000443 F PDR
t j
DISCLAIMER OF RESPONSIBILITY This document was prepared by Yankee Atomic Electric Company (" Yankee").
The use of information contained in this document by anyone other than Yankee, or the Organization for which the document was prepared, is not authorized and with respect to any unauthorized ese, neither Yankee nor it of ficers, directors, agents, or employees assume any obligation, responsibility, or liability or makes any warranty or representation of the material contained in the document.
i 1
i l
l t
l l
l
'f f I TABLE OF CONTENTS SUBJECT PAGE Introduction .......................................... 1 Background .......................................... 1 Discussion .......................................... 1-6 Conclusion .......................................... 6 Evaluation of Charcoal Filter Unit Fires Appendix I at Scabrook Station - PLC Iodine Adsorber Fire Test - Appendix II Nuclear Consulting Services, Inc.
l l
-111-i I
i
INTRODUCTION This report describes a llazards Analysis conducted on Seabrook Station's charcoal filter units. Table 1 identifies Seabrook's eight (8) air handling units and their location.
BACKGROUND Seabrook's approach to a fire within our charcoal filters is fire prevention and detection as outlined within the guidelines of Item II.B(3) of 10CFR50, Appendix R, which states, "specify measures for fire prevention, fire detection, fire suppression, and fire containment, and alternative shutdown capability as required for each fire area containing structures, systems, and components important to safety in accordance with NRC guidelines and regulations."
To address internal charcoal fires, an analysis was conducted on all Seabrook charcoal filters to determine the maximum temperatures of the charcoal adsorber sections due to decay heat from iodine and its daughter product decay without air flow. This analysis showed that the overall maximum temperature would be limited to 170 F. Additional analyses indicate that the maximum temperature for the HEPA filters (due to decay heat from the particulate iodines accumulated in these filters) will be limited to 187 F. These temperatures are well below the maximum limit of 300 F recommended in ANSI-N509-1980. Thus, there is no possibility of an internal charcoal fire due to decay heat.
Seabrook's charcoal adsorber filters are also protected from external fires since they are contained in a combination of heavy metal casing, wire debris screens, and fire retardant HEPA filters as recommended in Regulatory Guide 1.52, Design, Testing, and Maintenance Criteria for Post-Accident Engineered - Safety Feature Atmosphere Cleanup System Air Filtration and Absorption Units of. Light-Water-Cooled Nuclear Power Plants, Revision 2, March 1978.
Further, transient combustibles are limited administratively. Any welding or open flame sources will be controlled and limited and a fire watch will be maintained per plant administrative procedures during these activities. These precautions will prevent external sources from causing internal combustion to the filters.
However, a fire hazard analysis is developed in this report to address the effects of a postulated charcoal fire in the filters and its impact on equipment needed for safe shutdown. A realistic, however conservative approach was used to model the charcoal fires since charcoal is a slow burning medium.
DISCUSSION The following assumptions were used in this hazard analysis.
- 1. Fire will be detected by reliable and early warning system.
l
- 2. From detection, which is alarmed in Control Room, Operations per Operating l Procedures will shutdown air flou to the filters. Assume five minutes time from alarm conditions to shutdown of air flow. Charcoal is assumed to be ignited in this time frame.
]
- 3. The Fire Brigade will respond to the charcoal filter within 10 minutes from notification by the Control Room for all protected plant areas except containment. This notification is per Operating Procedures. For'a fire within containment, the Fire Brigade will repond within 15 minutes.
- 4. Ignition of the charcoal starts at the top of the charcoal panel. This is assumed conservative since a fire located lower on the panel would burn the retaining mesh and drop the charcoal from the air flow path precluding rapid fire propagation.
- 5. Since a fire cannot be started due to decay heat internally, the fire must
) be started f rom an external source. Assume an outside source is carried l into the filter unit. All the units have HEPA Filters on the inlet before l the charcoal bed. Each HEPA Filter is made up of a grouping HEPA filter i
elements 24" X 24" X 11-1/2". Each element is a throwaway, extended medium, dry-type filter, which are open face, rectangular, fire-resistance type design for radioactive service. Assume the source carried internal by air flow ignites one HEPA filter element, 2' X 2' totall This 2' X 2' filter element is assumed to ignite a charcoal panel 4 ft.y. .
- 6. Air flow through phe charcoal panels is assumed to be from the start of ignition. 4 ft. area of charcoal will burn under air flow condition for a period of 5 minutes time. At this point forced air flow has stopped and the resulting fire will be analyzed under natural draf t air flow.
- 7. Air flow velocity through the charcoal during forced ventilation is 40 feet per minute which is Seabrook's charcoal filters design velocity.
- 8. Further assumptions are used in Appendix I, " Evaluation of Charcoal Filter Unit Fires at Seabrook Station", 9-29-86 by Professional Loss Control, Inc.
and are noted in that Appendix.
The Hazard Analysis consist of 3 parts, (1) Determination of charcoal bed burning rates, (2) a heat transfer model of the charcoal filters and (3) effects of the heat transfer on safe shutdown equipment.
(1) Determination of Charcoal Bed Burning Rates A charcoal fire test was conducted by NUCON in their ASTM D3466 Test Rig. Data from this test was used by Professional Loss Control, Inc. (PLC) in their unsteady state heat transfer model of each of Seabrook's charcoal filters excluding CBA-F-8. Each Seabrook filter was revict ed seperately. NUCON's ASTM D3466 Test conducted for Seabrook used the same type of charcoal used in Seabrook's charcoal filter. The test normally is performed at 100 feet per minute air velocity, however 40 FPM velocity was used which is Seabrooks filter design velocity. The bed depth is normally 1.0 inch deep.
1 For Seabrook's test a 2.0 inches deep was used which is the limit of the ASTM D3466 appara tus . Seabrook's bed depth is 4.0 inches. Use of the test data by PLC is conservative since the test was conducted under forced air ilow over a one hour period. Seabrook's charcoal filter heat transfer model assumes five minutes time from charcoal ignition to shutdown of air flow where-as air flow will be shutdown five minutes after detection of a potential fire which most likely occurs before sufficient temperature is available to ignite charcoal.
A fire wind tunnel (FWT) test was conducted by NUCON on a 24 inch X 24 inch face area carbon adsorber specimen. The depth of the bed tested was 4.0 inches. Again, the charcoal used was the same type used at Seabrook, 2% KI and 2% TEDA inpregated carbon.
The charcoal was ignited by preheating inlet air to the charcoal specimem. Once the specimen started burning which was 8 minutes af ter C0 production levels of 50 ppm, the anticapated alarm setpoint, (normal ambient levels of 2 ppm C0), the specimen was tested for 5 minutes further under forced air flow of 40 fpm. Air flow was then stopped and inlet and outlet temperatures were monitored for one hour.
The purpose of the FWT test was to look at the actual test size modeled by PLC under fire conditions.
1 Air flow conditions under forced ventilation were the same for the FWT test versus Seabrook's charcoal filters. Once the ventilation was stopped and natural draf ting began, the FWT test was not similiar because of duct configuration differences.
Seabrook's charcoal filters have outlet dampers, long HVAC duct runs, and in some cases inlet dampers which are isolated once the filter fans are shutdown. Thus natural drafting through Seabrook's filters would be small. The FWT test with natural draf ting indicates the charcoal fire will contain itself to a limited fire with decreasing temperature af ter stopping forced ventilation.
Results of the FWT test show under conditions used in PLC model, carbon loss for a test duration of one hour was 4.53 lbs which is 10% of the test dry carbon weight.
Also that C0 levels increase well above normal environment levels will before a fire starts.
(2) Heat Transfer Model The PLC unsteady heat conduction analysis looked at each charcoal filter except CBA-F-8 the net heat transfer to the filter surface accounts resulting from the charcoal temperature data supplied by NUCON. Radiation and convection heat transfer was also considered in PLC's analysis.
Radiation Heat Transfer from the fire was considered taking into account the geometric of each of the charcoal filters. The burning charcoal surface area was conservatively assumed to be 26 inch square. Since HEPA Filters are 24" X 24" dimension, the external burning source could be 22" X 22" square. The larger burning surface accounts for any fire proporagation under the five minute forced ventilation period. The temperatures used in the analysis were measured within the charcoal bed on the outlet side. The highest of any of the temperatures measured was also used. Radiation Heat Loss from the steel housin~g to its surroundings was also considered.
i For convective heat transfer, forced convection within the filter housing was neglected. The forced air stream would be heated and moved outside the filter unit to the ductwork, thus cooling the housing. This assumption is conservative. Free convection heat transfer was considered on the outside of the filter housing.
Attachment II gives the detailed methodology and results of the analysis.
The following conclusions are drawn from a fire involving the charcoal cells in the air handling units.
- 1. The worst case maximum localized steel plate housing temperature was ~
calculated to be 704 F. This temperature is substantially below that required for structural failure of the steel housing.
- 2. S tructural failure of any steel beam or column in the vicinity of these filter units cannot be caused by heat transfer from the filter housing.
- 3. The maximum radiant heat emissjve flux from the housing at 704 F, calculated to be less than 10 KW/m , is less than half the critical radiant flux necessary to ignite the worst case cable jacket materials as determined by EPRI sponsored tests at Factory Mutual Research Corporation -
(EPRI NP-1200 part 1).
(3) Safe Shutdown Equipment Revieu From the conclusions of the heat transfer model there would be no structural steel failures in the vicinity of Seabrook's charcoal filters. Thus no safe shutdown equipment would be effected due to steel failures. Equipment further than three feet from the charcoal filters also would not be effected based on the maximum heat flux from the housing.
An evaluation of safe shutdown equipment was conducted looking at the equipment within and including three feet f rom each of the charcoal filter units.
CBA-F No charcoal fire modeling was done on this filter. It is assumed that a charcoal filter fire will cause loss of all equipment within its fire area (i.e.
CB-F-3B-A).
i Seabrook's present Appendix R Safe Shutdown Study shows this to be acceptable. Also '
there is no concern of damage to structural steel since all this steel in this fire area is fire proofed.
CAP-F There is no safe shutdown equipment used during a fire in this fire area, PAB-F-3A-Z, within and including three feet of CAP-F-40.
CAH-F There is no safe shutdovn equipment used during a fire in this Fire Area, C-F-3-Z, within and including three feet of CAH-F-40.
EAH-F-9,69 - There is no safe shutdown equipment used during a fire in this fire area, CE-F-1-Z, within and including three feet of EAH-F-9,69.
FAH-F-41, 74 - There is no safe shutdown equipment used during a fire in this fire area, FSB-F1-A.
. _ . . ,,..m_.__- _ . _ . . - _ . - . . - . - - - . - - . . _ _ . , _ _
t PAH-F There is no safe shutdown equipment used during a fire in this fire area PAH-F-4-Z, within and including three feet of PAH-F-16.
CONCLUSION:
The hazards posed by the heating of the steel housing from a charcoal bed filter for fires under the operational guidelines to shutdown forced ventilation of the filter in question, will not jeopardize the safe shutdown of Seabrook Station.
i TABLE 1 FILTER ID SAFETY /NON MEETS RG 1.52 AREA DETECTION FIRE AREA EAH-F-9 SAFETY YES YES CE-F-1-Z CONTAINMENT ENCLOSURE EL 21' 6" EAH-F-69 SAFETY YES YES FAH-F-41 SAFETY YES YES FSB-F1-A FUEL BUILDING EL 84' 0" FAH-F-74 SAFETY YES YES CAH-F-8 NON NO YES C-F-3-Z CONTAINMENT PAH-F-16 NON NO YES PAB-F-4-Z PRIMARY AUXILIARY BUILDING EL 81' 0" CAP-F-40 NON NO YES PAB-F-3A-Z PRIMARY AUXILIARY BUILDING EL 53' 0" CBA-F-8 NON NO YES CONTROL ROOM HVAC EQUIPMENT ROOM EL 75'
! CB-F-3B-A
-__ _ _ - _ _ _