ML20211N354

Forwards Summary of 990506 Meeting with NEI Re Exchange of Info on Reactor Oversight Program Fire Protection Issues. Meeting Agenda,List of Attendees & Written Info Exchanged at Meeting Also Encl

ML20211N354
Person / Time
Issue date:	09/02/1999
From:	Spector A NRC (Affiliation Not Assigned)
To:	NRC (Affiliation Not Assigned)
References
NUDOCS 9909100168
Download: ML20211N354 (76)

Text

--

September 2,1999 j i

MEMORANDUM TO: File FROM: August K. Spector, Communication Task Leader l Inspection Program Branch (Original signed by:)

Diwsion of Inspection Program Management Office of Nuclear Reactor Regulation

SUBJECT:

PUBLIC MEETING REACTOR OVERSIGHT PROGRAM FIRE PROTECTION ISSUES MAY 6,1999 l

On May 6,1999, a public meeting was held between the NRC and the NEl to continue I exchanging information on the reactor oversight program fire protection issues. The meeting agenda, a meeting summary, a list of attendees and a copy of written information exchanged at the meeting are attached.

Attachments: As stated l

Contact:

August K. Spector 301-415-2140 l

l

/

co3 p1 Distribution:

Central Files PUBLIC llPB R/F

%l} F$ -f b~ Q 40 g 4 g-6 Hee ( 4 DOCUMENT NAME: A:\MAY6MTG

• See previous concurrence.

To receive a copy of this document, indicate in the box: "C" = Copy without enclosures "E" = Copy with enclosures "N" = No copy OFFICE IIPB:DIPM f7 IIPht> LPM g l l l NAME AKSpector P WMA DATE 09/ 6, /99 s ofi /99 T OFFICIAL RECORD COPY Nbd 1 9909100168 990902 C PDR REVGP ER

y- -

l l

f* U:o gf a UNITED STATES y

j NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20665 4001

% l

• • • • • September 2, 1999 1 MEMORANDUM TO: File FROM: August K. Spector, Communication Task Leaderf Inspection Program Branch Division of Inspection Program Management Office of Nuclear Reactor Regulation

SUBJECT:

PUBLIC MEETING REACTOR OVERSIGHT PROGRAM FIRE i PROTECTION ISSUES MAY 6,1999 I l

On May 6,1999, a public meeting was held between the NRC and the NEl to continue exchanging information on the reactor oversight program fire protection issues. The meeting agenda, a meeting summary, a list of attendees and a copy of written information exchanged at the meeting are attached.

Attachments: As stated 4 I

Contact:

August K. Spector j 301-415-2140 l 1

l l

I k.._. . - _ . .

fi i

I.

I MEETING

SUMMARY

MAY 6,1999 '

NRC presented a draft copy of Fire Protection Baseline Protection Procedure and discussed.

~ NRC provided Draft copy of the Fire Protection Risk Significance Screening Methodology and discussed. Agreed to meet on May 24,1999, to continue discussions.

I i

l

)

i l

1 l

k Attachment 1

r AGENDA MAY 6,1999 Discussion of Fire Protection inspectable Areas i

i Attachment 2 C____._______ -

.. I ATTENDEES' Public Meeting May 6,1999 Fred Emerson Dave Modeen Steve Floyd '

-Tom Houghton -

, r NUCLEAR REGiJLATORY COMMISSION Alan Madison August Spector Morris Branch-John Hannon Steve West Steven Stein ,

l Patrick Madden J.S. Hyslop

- OTHERS Kim Green, NUS Info Services l l

l l ,

1 l.

i Attachment 3 l

l

~

[>

L Clarifications on NRC doc. comments

! Inspection basis (last sentence) is inconsistent with the objective:

Yhe last sentence of bases "This inspectable area verifies aspects of the initiating events and Mitigating Systems comerstone for which there are no performance indicators to measure performance" conflicts with the final phrase of the inspection objective " ensures that procedures, equipment, fire barriers, and systems exist so that the capability to safely shut down the plant is ensured" l The inspection bases indicates that the " conventional" fire protection aspects are to be inspected while the objective indicates that both the " conventional" and the

" App. R FSSD" aspects are to be inspected.

l 02.02.b appears to be demanding that we have an unannounced fire drill in a l high risk area at a time of their choosing:

l l Ultimately the intent of the comment was to have the inspection module make it clear that a fire drill ja a required element of the inspection for the licensee (as

- indicated by the NRc response) and that requiring it to be unannounced in a specific location will have a resource impact on the licensee. An unannounced fire drill is one for which the time and location are totally unknown to the drill participants and is a requirement of the program, as each fire brigade / member is required to participate in a minimum number of drills at a specific frequency l

including unannounced drills. To alter the scheduling to have an unannounced drill in the specific week can upset the normal drill scheduling. This is not to say

, the sites will not do this but to indicate that it is not a no impact request. As to l high risk area, most plants will have only a few (5-6) areas that would be

considered high risk in a PSA sense. Other areas such as lube oil storage areas would have a significant conventional fire risk or hazard. Normally all plant power block areas are covered in the FSSD analysis, so all fire areas are " plant

' areas that could require the conduct of a post-fire safe shutdown".

L Triennial general guidance, should say pick a different high risk area than the previous inspection, to be most comprehensive should also include a sampling check of lower risk areas. Otherwise the utilities will do a " teach the test" and rigorously keep the high risk areas at 100% and let other medium and low areas degrade:

The comment was intended to aid in the eventual review of the full breadth of fire areas of a plant, reducing the potential to have that inspection team re-review an are previously inspected and found acceptable. Unless there is an explicit statement to vary the areas inspected from the previous inspection, the same areas could indeed be chosen as the selection criteria would be the same. The only difference driving choice of a different area would be industry or site specific events triggering a need to review a specific area different from the previous I

L

p inspection (such as the JAFitzpatrick hydrogen fire triggering inclusion of a site's hydrogen storage area in the scope of the inspection).

r 03.01/2 include the referenced information from the FPFI module to allow this to be self sufficient:

The commenter's impression was that the routine and triennial inspections in this draft module would, under the new assessment process replace the FPFl. If this

were true having a inspection module that would be used as only an information I source for other modules and not as a stand alone inspection implementing l document did not make sense, if the FPFI is not bein;; auperceded by this l module, a reference to its use and place in the hierarc..y r would be appropriate in the beginning of the module. Ex. The FPFI module XXXXX is used in the evaluation of y comerstone degradation evaluation, not this module. This would l

clarify the relationship of the modules like the relationship of the plant status

_ module to this module referred to in the specific guidance 03.01.

03.02A should also request supporting calculations:

The calculations referred to in the comment are not the supporting calculations for the IPEEE, but the supporting calculations for the FSSD analysis such as structural steel calculations, hydraulic calculations for sprinkler systems etc. and

are intended to be included in 3.02a, inspection preparation not 3.02b, fire risk l report.

03.02b the inspection results and non plant specific fire event information have i

little to do with ranking fire areas 'according to risk and so should not be the responsibility of the SRA to acquire /ad;ess:

This information, while possibly useful to the SRA is not information that requires his/her specialized training to acquire , and therefore should be provided by a

! more generalized person to the SRA, i.e. the inspection results for the site as related to fire issues could be extracted by the site resident inspector. The intent of the comment was to conserve the limited SRA resource.

I l

Leon Whitr.Ay - Baseline wpd rffff l P79e 1 j i

t 5/3/H INSPECTABLE AREA: Fire Protection l

CORNERSTONES: Initiating Events (10). Mitigating Systems (90)

INSPECTION BASES: Fire is a significant contributor to risk. In many cases, the risk due to j fires is comparable or exceeds the risk from internal events. Fire protection defense-in-depth is l accomplished through control of combustibles and ignition sources, mitigation of fires that do occur through fire detection and automatic and manual suppression capability, and a well analyzed and implemented post-fire safe shutdown capability. Safe shutdown capability includes the existence of adequate fire barriers to establish the fire area or fire zone configuration and to ensure the shutdown equipment functionality assumed in !! e post-fire safe '

shutdown analysis. If defense-in-depth is not maintained through a well funstning licensee fire protection program, post-fire safe shutdown of the plant may be challenged. This inspectable area verifies aspects of the Initiating Events and Mitigating Systems cornerstone for which there are no performance indicators to measure performance.

LEVEL OF EFFORT: On a monthly basis, the resident inspector will tour high fire risk plant areas to assess: control of transient combustibles and ignition sources, operability and availability of fire detection, manual and automatic suppression capabilities, and barriers to fire propagation (e.g., fire doors, fire dampers).

In addition, every 3 years, an inspection team consisting of a fire protection engineer, a mechanical engineer, and an electrical engineer will conduct a one week, risk focused, onsite inspection of all three components of defense-in-depth, with emphasis on post-fire safe shutdown capability and configuration management, but including review of fire protection administration and fire protection systems and features.

01 INSPECTION OBJECTIVE The inspection objective is to assess whether the licensee has implemented a fire protection program which adequately controls combustibles and ignition sources within the plant, provides adequate fire detection and suppression capability, and ensures that procedures, equipment, fire barriers, and systems exist so that the capability to safely shutdown the plant is ensured.

02 INSPECTION REQUIREMENTS 02.01 Routine Inspection. The resident inspector's assessment of the licensee's control of transient combustibles and ignition sources is addressed on a more frequent basis in the Plant Status inspection procedure. Select high fire risk areas based on the plant specific risk information matrix or the generic RIM 2 document for the subject reactor plant. [It is planned that risk information will be contained in an attachment to Manual Chapter 2515 of the new NRC reactor oversight and inspection program.)

a. Monthly the resident inspector will tour high fire risk plant areas to assess the operability and availability (but not necessarily the design) of fire detection and manual and automatic suppression capabilities and equipment and barriers to fire propagation, and fire protection related compensatory measures.

1

) II. son Whtoey Basehne.wod P ge 2

!k T,- ,

I~~

b.- Annually the resident inspector will observe a fire brigade drill (preferrably in a high risk fire area), or actual response of a plant fire brigade in any plant area.

- 02.02 Triennial Insoection. Conduct a one-week triennial team inspec'Jon of the licensee's fire protection program emphasizing post-fire safe shutdown capability and configuration management, but including review of fire protection administration and fire protection systems and features.

d

a. Inspection Preparation: The inspection team leader will manage and coordinate a 2-3 day laformation gathering site visit accompanied by the team members and the senior reactor analyst (SRA) designated to support the team. The SRA will

. provide a report to the team leader containing plant rpecific fire risk results. The team leader will use the fire risk results report and input from the other team members to develop an inspection plan.

b. Inspection Conduct: The inspection team leader will manage and coordinate the conduct of a one week triennial fire protection inspection emphasizing post fire safe shutdown capability and configura*lon management. The inspection should be either plant area-based or system-based, depending on the structure of the {

licensee analysis. The triennial inspection may also include observation of a simulated post-fire safe shutdown from '.;tside the control room (e.g. from a remote shutdown panel and/or remote control stations, or possibly at a remote

- shutdown panel associated with the plant simulator), and observation of a fire brigade (and possibly offsite fire department) drill for a simulated fire in a high risk area.

{

INSPECTION GUIDANCE:

I General Guidance 1

Triennial Inspection. The purpose of this triennial team inspection procedure is to selectively )

review the fire protection program at the subject reactor plant. In one week the inspection team  !

can not possibly confirm the post-fire safe shutdown capability of the equipment necessary to 1

- provide all safe shutdown functions in each plant fire area. Rather, as described below, the team should use all available plant-specific risk, event, and technical information (including the results of licensee self-assessments) to confirm safe shutdown capability (i.e. post-fire existence of all safe shutdown functions) for fire scenarios in from one to five selected and prioritized plant areas, rooms or zones, in addition, the team will assess the design adequacy of fire protection systems and features and fire protection program administration (including administrative controls such as compensatory measures and fire brigade response and capabilities).

, General topical areas for team member review during preparation are: the power plant's design, layout, and equipment configuration; the current licensing basis (i.e., fire protection regulatory l framework); and the licensee's strategy / methodology for accomplishing post-fire safe

( shutdown. Additionally, prior to the site visit (and using any modification packages obtained I during the information gathering visit), the team members should become knowledgeable

( regarding plant design changes or modifications implemented since the plant's post-fire safe shutdown capability was last reviewed by the NRC staff.

L

' [ Leon Whitr,ey - Baseline.wpd

~

Page _3 l Q -

Specific Guidance 03.01 Routino Insoection. Fire prevention is the most important attribute of a fire protection program. Consequently the prevention attribute is addressed in the more frequently performed Plant Status procedure.

a. Monthly: This inspection should not attempt to address all high fire risk areas each month, it should focus on one or possibly two plant areas and concentrate on the operability of the detection and suppression systems and fire barriers for the areas. The monthly is not intended to perform a design review for that area.

The Specific Guidance Section and Appendices A and B of the referenced draft FPFI inspection procedure provide guidance for how to review the fire detection and manual and automatic suppression capabilities, and barriers to fire propagation.

Compensatory measures are put in place to compensate for degraded detection, suppression, or fire barrier features. The compensatory measures should be adequate to compensate for the degraded function. Guidance for the review of I

compensatory measures is in IN 97-4P,

• Inadequate or Inappropriate Interim Fire Protection Compensatory Measures."

b. Annual: Fire brigade response is the major component of manual fire suppression at commercial reactor plants. The attributes that are important to consider for evaluating fire brigade performance are contained in the Specific Guidance section and Appendix A of the referenced draft FPFIinspection procedure.

03.02 Triennial inspection.

The triennial inspection is intended to provide a risk focused look at the three components of defense-in-depth (prevention of fires, detection and suppression of fires, 3 and the ability to safely shut down after a fire) with emphasis on post-fire safe shutdown j capability and configuration management. The triennialinspection is the only portion of the baseline inspection program that focuses on the design of the fire protection systems and features that assure post-fire safe shutdown. The inspection will adopt, as appropriate, techniques developed in the referenced draft FPFI inspection procedure.

The inspection team should consist of a fire protection engineer, a mechanical engineer, and an electrical engineer. The risk focus will be provided by a senior reactor analyst who will provide extensive input into the planning process.

a. Inspection Preparation:The team members should develop an awareness of the plant's current post-fire safe shutdown licensing basis through review of 10 CFR 50.48,10 CFR 50 Appendix R (if applicable), NRC safety evaluation reports (SER) on fire protection, the plant's operating license, Updated Final Safety Analysis Report (USAR), and approved exemptions or deviations. l The team members should become familiar with the licansee's methodology for i accomplishing post fire safe shutdown conditions. Sources of information include: calculations and analyses, the Updated Final Safety Analysis Report (USAR), the latest version of the Fire Hazard Analysis (FHA), the latest version l

p ILeon WNtney - Baseline.wpd

~~

Pega 4 j

?

i i.

of the Post-fire Safe Shutdown Analysis (SSA), fire protection / post-fire safe shutdown related 10 CFR 50.59 documentation, plant drawings, and emergency / abnormal operating procedures. The intent of this effort is to make the inspectors familiar with:

e The licensee's methodology for achieving safe shutdown conditions in the event of fire in any area of the plant, o The systems credited for each fire area, room or zone by the licensee as surviving the fire for accomplishing required shutdown functions (e.g.

reactivity control, reactor coolant make-up, reactor heat removal, process monitoring, supporting functions) necessary to comply with the safe shutdown requirements of 10 CFR 50.48 (a) and plant specific licensing requirements.

• The support system requirements of each shutdown system, and j e The licensee's approach for identifying and resolving associated circuits of concern. The electrical review should include the assumptions and boundary conditions used in the performa.:ce of the analyses.

e The procedures in place for accomplishing post-fire safe shutdown.

. The team members should become familiar with the historical record of plant specific fire protection issues through review of plant specific documents including: previous NRC inspection results, intemal audits performed by the reactor licensee (e.g., self-assessments and Quality Assurance audits), Event Notifications submitted in accordance with 10 CFR 50.72 and Licensee Event Reports (LERs) submitted in accordance with 10 CFR 50.73.

a. Fire Risk Report: The senior reactor analyst (SRA) will develop and present to the inspection team leader a fire risk results report using the process of Appendix l of the referenced draft FPFI inspection procedure. The report will identify plant-specific, fire risk significant plant areas, structures, systems, components (SSCs), and operator actions developed from risk information such as Individual Plant Examinations of External Events (IPEEEs). The fire risk results report may also consider:

o resident and regional inspector developed inspection results, e plant-specific and non-plant specific fire event information, e licensee developed fire hazards analyses and plant srecific post-fire safe shutdown operating procedures, e licensee developed fire protection program self-assessment documentation.

The SRA's report will not focus on the validity of the modeling assumptions of i the IPEEE.

1 1

• II. son WNtney- Basehne.wpd Paga 5j l 4

s-The development, presentation and finalization of the plant specific fire risk

. report may require approximately one week of effort. It is expected that in many cases the plant specific fire risk report for the previous triennial inspection may only need minor updates to incorporate plant modifications or changes to fire protection methodology implemented since the previous inspection.

c. Information Gathering Site Visit and inspection Plan:

The inspection team leader will manage and coordinate a 2-3 day information gathering site visit accompanied by the team members and the senior reactor  ;

analyst (SRA) designated to support the team. The purpose of the information  !

gathering site visit is to gather the site specific information important to l inspection planning and for the SRA to clarity risk assumptions with licensee risk analysts, in advance of the information gathering site visit, the team leader should provide the licensee a list of information and documents that will be needed to prepare for the inspection, and the inspection team should already have reviewed previous inspection information and docketed information about the licensee's fire protection program in order for the on-site information l exchange to be as effective as possible.

I The inspection plan developed by the team leader for the triennial inspection should take into account:

e the results of licensee self-assessments; I e the results of previous fire protection inspections including resident, regional, and team inspections; e- the fire risk report, e previous industry problems and reactor events.

The plan should also consider the scope of previous fire inspections, so that the inspections are not repetitively focused on the same area or fire zones,

d. On-site Inspection: The inspection team will confirm post-fire safe shutdown capability for fire scenarios in from one to five selected and prioritized plant areas, rooms or zones. In addition, the team will assess the design adequacy of fire protection systems and features for the areas, rooms or zones.

The inspection will be conducted during a one week onsite inspection period.

Based on input from the SRA and team members, the areas, rooms, or zones selected for review will be determined by the team leader and documented in the inspection plan. For those features applicable to the selected area, rooms, or zones, specific inspection guidance pertaining to the features is contained in the referenced draft FPFI inspection procedure.

In assessing risk significance of inspection findings in one or two of the three areas of defense-in-depth, some effort will need to be applied in the other defense-in-depth areas to support processing the issue through the (inspection findings) Significance Determination Process.

p '

~ '

ILeon Whitney - Baseline.wpd P:ge 6) '

i ?

a.

~

RESOURCE ESTIMATE: l l This procedure is estimated at 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> per year for routine inspection and 108 hours0.00125 days <br />0.03 hours <br />1.785714e-4 weeks <br />4.1094e-5 months <br /> every 3 years for the triennialinspection .

I

REFERENCES:

I IN 97-48, " inadequate or inappropriate Interim Fire Protection Compensatory Measures," July l 9,1997. l l

NRC " Fire Protection Functional Inspection (FPFI) Draft for Prairie Island inspection, April 6, )

1998.

ATTACHMENT ROUTINE INSPECTION GUIDANCE TABLE CORNERSTONE RISK PRIORITY EXAMPLES INITIATING EVENTS (10) Equipment or actions that could cause Transient combustibles (rags, wood,6on or contribute to initiation of fires in high exchange resin, lubricating oil, or fire risk areas or near equipment Anti-Cs) are not in areas where required for safe shutdown. transient combustibles are prohibited.

Transient combustible amounts in other areas do not exceed administrative controls.

Ignition sources (welding, grinding, i brazing, flame cutting) have a fire

( watch. Planning includes precautions and additional fire prevention measures where these activities are near combustibles.

[Leorr WNtney - Baseline.wpd p,tg3 7 l l

l v.

l 4-MITIGATING SYSTEMS (90) Fra Barriers in Ngh Are nok areas. Doors and dernpers that prevent the spread of Ares to/or between high fire Detection Systems for high fire risk risk areas remain in place and are areas funcbonal Autornatic suppression systems for high Electrical cable fire wraps and fire risk areas penetrabon seals that protect the post-fire safe-shutdown train are not l Manual suppression from fire bngado damaged.

Compensatory measures for degraded Fire detection and alarm system is fire detection equipment, suppression funcbonal for high fire risk areas, features and fire propagation bemers.

Automabc suppression system spnniders are not blocked.

Fire brigade perfonnance indicates a prompt response with proper fire fighting techniques for the type of fire encountered.

Manual fire suppression equipment is of the proper type and has been tested. l Degraded fire detection equipment, suppression features and fire l

propagation baniers are adequately t

compensated for on reasonably short-term bases. ]

I'

)

l 9

l i

i

F

- rit(11 l, -

SIGNIFICANCE DETERMINATION PROCESS lt INFORMATION SHEET FOR GENERIC PWR (4 Loop)

FOR TRAINING PURPOSES ONLY Plant Ecuipment (Single Unit Station - no cross-tie capability for any systems) 3 train AFW (2 motor and 1 turbine driven pumps) 2 train centrifugal charging (used for high head safety injection (HHSI))

2 train intermediate head safety injection (IHSI) (-1400 psi head) l 2 train low pressure safety injection (LPSI) and RHR (~300 psi head) 2 train containment spray (containment pressure suppression only, no DHR function) 2 trains emergency diesel generators and 1 train gas turbine generator 3 trains motor-driven condensate pumps (-300 psi head) 2 trains turbine-driven main feedwater pumps (60% of full power each) 2 trains Service Water (2 pumps per train, each 100%)

2 trains CCW (2 pumps per train, each 100%)

1 Atmospnede Dump Valve per Steam Generator (Class 1E de power)

Plant fire'nain system supplied by 3 motor-driven and 1 diesel-driven fire pumps (provides emerponey AFW suction source and RCP seal cooling backup - determined to be risk-sigraficant under maintenance rule)

Core Heat Removal Paths Normal Decay Heat Removal: RHR flow to RHR heat exchangers (1 per train) to CCW to Service Water to ultimate heat sink (river)

Initial Accident Heat Removal: (1) AFW to steam generators to either main condensor or via ADVs to atmosphere, (2) safety injection (high, intermediate, low head)

Lor.0-term Accident Heat Removal: (1) normal DHR path for non-LOCA accidents, (2) for LOCA accidents the same heat removal path is used as for normal DHR except that RHR takes suction from the LOCA sump in containment and retums coolant to the vessel either directly or via the IHSI or HHSI system, depending on primary pressure j Station Blackout Procedures Turbine-driven AFW pump is operated normally until battery power depletes (6 hour6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />  !

I battery life assuming non-essential load stripping has been accomplished per procedure), j then can be operated manually / mechanically per procedure. RCPs have high- i tempera +ure seal O-rings installed. Diesel fire pump can supply emergency RCP seal )

cooling per procedure using pre-staged hoses / couplings. If these procedures are credited, then " recovery of failed train" can be credited in thn SDP Table 2 assessment of remaining mitigation capability.

l 1

. , , Suct. ass Paths 4 Initiiting Ev:nt: Tr:nsi:nt 1" path: 1/2 MDA.FW or TDAFW train or 1/2 MFW 2"8 path (1" path lost):

(1/2 HHS1 or 1/2 IHSI train) and (2 of 2 PORVs) and (1/2 LPSI/RHR train for HP recirc) to provide feed / bleed cooling initiating Event: LOOP 1" path:

(A EDG or GTG) and (A MDAFW train) or (A HHSI or lHSI train (feed / bleed] with 2/2 PORVs and A LPSI/RHR for HP recirc capability) 2"8 path (1" path lost): (B EDG or GTG) and (B MDAFW train) or (B HHSI train [ feed / bleed] with 2/2 PORVs and B LPSI/RHR for HP recire capability) 3'8 path (above paths lost): TDAFWis not a long term succ.ess path, therefore its availability (along with implemt.nting all other SBO procedures) can be credited as providing sufficient time for recovery of power (from either onsite or offsite sources)

Initiating Event: SGTR 1" path: (1/2 MDAFW trains or [TDAPN or 1/2 MFW] with ruptured SG isolated) used to cooldown/depressurize (stops leak) and (1/2 HHSI or 1/2 IHSI train) used to maintain coolant inventory for effective secondary heat transfer 2"8 path:

(1/2 HHSI or 1/2 IHS! train) and (2 of 2 PORVs) used to cooldown/depressurize (stops Icak) and (1/2 LPSl/RHR train for HP recirc) and (ruptured SG isolated) to provide feed / bleed cooling Initiating Event: Small LOCA 1" path:

(1/2 HHSI or 1/2 IHSI train) and (1/2 LPSI/RHR train for HP recirc in " piggy back" mode) 2"8 path (1" path lost):

(1/2 MDAFW trains or [TDAFW or 1/2 MFW) for cooldown) and (1/2 HHSI or 1/2 IHS! train) and (1/2 RHR train for normal RHR)

Initiating Event: Medium /Large LOCA Only path:

(4 accumulators) and (1/2 HHSI train) and (1/2 lHSI train) for injection and (1/2 LPSI/RHR train) for LP recirc t

,, initiating Event: ATWS (actu lly a transiint with concurrsnt RPS f2ilurs) 18 path:-

M nual opining or de-snergizing raactor trip bretksrs

. (electrical RPS failure) 2"8 path (1" path lost):

[(1/2 MFW) or (2 of 2 MDAFW) or (TDAFW)) and (1/2 HHSI train) for emergency boration initiating Event: MSLB (inside containment) and MFLB

' 1" path:

Isolation of feedwater to faulted SG and (1/3 AFW or 1/2 MFW trains) 2"d path (1" path lost):

(1/2 HHSI or 1/2 IHSI) and (2/2 PORVs) and (1/2 LPSI/RHR for HP recirc)

Initiating Event: MSLB (outside containment):

1" path:

(Isolation of feedwater and main steam to faulted SG) and (1/3 AFW or 1/2 MFW trains) 2"* path (1" path lost):

(1/2 HHSI or 1/2 IHSI) and (2/2 PORVs) and (1/2 LPSI/RHR for HP recirc)

Extmpla inititting EvInt Sc ntrios To B3 Considirsd Affected System Support Systems initiating Event Scenarios AFW 125v DC transient', LOOP, MSLB (outside contmt), SGTR, SLOCA from PORV/SV/RCP, MFLB, MSLB (inside HHSI 125v DC, CCW contmt), SLOCA from pipe breaks, ATWS, M/L LOCA IHSI 125v DC, CCW transient', LOOP, MSLB (outside contmt), SGTR, SLOCA for PORV/SV/RCP, MFLB, MSLB (inside LPSl/RHR 125v DC, CCW, SWS contmt), SLOCA from pipe breaks, M/L LOCA EDG 125v DC, SWS, LOOP EDG HVAC CCW 125v DC, SWS transient', LOOP, MSLB (outside contmt), SGTR, SLOCA for PORV/SV/RCP, MFLB, MSLB (inside SWS 125v DC contmt), SLOCA from pipe breaks, ATWS, M/L LOCA

' Note: Loss of PCS will always occur during a reactor trip from 100% power, due to MSIV closure on low RCS temperature. Therefore, transient scenarios should generally be developed using loss of PCS, unless the reactor was operating at 50% or less power during the time period being considered. However, operating procedures exist to reopen an MSIV for the purpose of re-establishing a MFW supply to the steam generators and the main condensor as a heat sink following a reactor trip.

l l

Other information I Leak within makeup capability: < 0.5 inch diameter SLOCA break size: 0.5 - 2.0 inch diameter MLOCA break size: 2.0 - 6.0 inch diameter LLOCA break size. > 6.0 inch diameter l

l

l

. Credit Allowed for SBO recovery (NOT CURRENTLY USED)

Westinghouse Old (low No emergency SBO Emergency SBO RCP seal cooling temp) O-rings RCP seal cooling within within first 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> first 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> TDAFW train available No credit allowed for Credit allowed for recovery of power recovery of power if licensee SBO procedures are Battery life 5 hrs or greater judged to be adequate TDAFW train available No credit allowed for Credit allowed for recovery of power recovery of power if licensee SBO procedures are Battery life less than 5 brs judged to be adequate AND if TDAFW pump can be operated manually without electrical power TDAFW train not available No credit allowed for recovery of power Westinghouse New (high- No emergency SBO Emergency SBO RCP seal cooling temp) O-rings AND RCP seal cooling within within first 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> Byron-Jackson (3 or 4 first 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> equal stages of pressure breakdown)

TDAFW train available Credit allowed for recovery of power if licensee SBO '

procedures are judged to be adequate Battery life 5 hrs or greater TDAFW train available Credit allowed for recovery of power if licensee SBO procedures are judged to be adequate AND if TDAFW pump Battery life less than 5 hrs can be operated manually without electrical power TDAFW train not available No credit allowed for recovery of power

Reference:

Draft Technical Report " Perspectives Regarding RCP Seal LOCA Modeling in the IPEs", revision 1, October 1997, Office of Regulatory Research, Division of Systems Technology, Probabilistic Risk Analysis Branch I

~.

n -

)

p R )o Hi -

o o Rt a

/ l L I S

u c

r 4

( P i L c R 2 e

r _

W )

S f P P y C oH 1 (

cl P r n i

f eO n s o e e s s

Gs o o s a p L V) r -

Rd e fow p

i W Oe n Pg r F t Pl b o n x A W F 2/ d i

t i R D f a n ni

(

s M W M o ef n ra t n 2 F 2 2(

foT e f A f o

n r i

s o D I o n 1 T l hF a r

t a T I S I P S H

s H -

s e H I c 2 2 c f f u o o S l 1 _

P -

D -

S _

) A B p

o R R o H H L R /

R

/ I 4

(

I S

S R P P L L-W P

cl y )

n sd Ve e i

r eO n Rl e se Ob/

Gs o r e

Pd e

a p 2 e r w / f f om P o 2(

n P e og n t

i t i a n i

f s A A A B B B f

oi r a r r O WI F S I S

I S S I

W F

foT nr f

o A H H HI HH HI A I o s s D - - - - D hF o M M t

a L P

s s

e c

c u A B S

P G G G D D T D S E G E

)

p o

o L I 4

( S I H S R H H I

G

/

W P

2 2 S y f f d cl o o e i

r n 1 1 r

u eO n

t p

e es u Gs o r a p t e

r r l a

fou o n P s

n R )o i

ogn R Hi o i T t t i a n G Rt a

/l y -

mi S I u t r a i

r W F

• S P

c r

l i

b foT i

W L c a nr A e p I o D

• F 2 r a c

hF M W M f P t

P a

2 F 2 oH 1 (

s e

r s f A f o

i u

s e

o D q T 1 s e c 1 V) c R u Rd e

• S Oe P Pl b D 2/ d S f oefe 2(

)

p o

o s L r 4 e

( k a

R e r

W P

B I S

p cl y i H )n i

r n T r Ho eO i

t r 2 a n o f r e s e t oob Gs o c a 1 (

a p e r r R fou S g n P n ogn W i z

i t i T i g

a n A r e

mi r r a n e

foT e nr d I o r W hF t o F P

a g A s i n D s n e

c c

e p

o MFWA 2

f u l o D S a '

P u 2 l n

D a S M

)

p o R o H L R 4

( 2 f

Pt o W

P 1

cl y -

i r n eO n

e se n Gs o R Hi )o W a p A Rt a F r r C /l A W .

O I u D F fou S c n P L P i r M W M og l l

a L c e 2 F 2 in t i a n m 2 r f A f o

f P o D h S

mi r ar oH 1 (

1 T 1 foT n r I o hF t

a P

s I s S e I _

H S c

c H H u I S 2 2 P f f o o D 1 S

1 _

m

4 R

) H p

o R

/

I o S -

L P -

4

(

L .

2 R /

1 W

P cl y -

i r n eO n

A I S

e e s C O H -

Gs o L I _

a p 2 -

r e g

/

1 fo m r n Pg a

L _

i o n d I t i a n n S a H mi r a m H r

foT i u 2 nr o d /

1 -

I e

hF t M a -

P s r -

s s t o

e a c l u

c u m S u -

P c c -

D A S -

4

/

4

)

p o

o R )"

L H 4 R*

/

( I "

R S .

W B W F

P L

P L #

cl y F A W 2 f

D F o"

i n M M FWM r

eO n d 1 (

e e s n a 2 A 2 Gs o f o D f

o a p )

t T 1 n

ow r

1 _

e s f V n Pg m n R )* -

o n i O*

i a r t i a n t n o P t

mia r r o

c a

r 2

f foT e )

A

)

B e

n o**

nr d e 2(

I o i s n n G hF t i n i a

r i

a r

a ( t t m P ( ( a s

B *

• t e

s L s s S I e S e e S I S

c c M t

a t

a d e H H u

l o

l o t c H I S i s i s e 2 2 f

P f f f D W W A o o S F F

• 1 l

l llI W

F

)

A n p

o D FW R Hi )o o Rt a L MA /l I u 4

(

2

/ D S P

c r

1 T i R - ~ L c e

W P )

)

2 f P r

cl y

n

)t n

A n

B oH 1 (

i r e n eO n m i

a i

a s r r e e n (

t

(

t Gs o i t

a

• s s

a p n e s e

r o V) fow t t c l a l a Rd e n P e o s

o Oe d s o g n i

s i i r Pl b i

t i t S S o 2/

a n u M M t

a f d e mia (

o r o fe r r - - e foT B n 2(

e nr L - -

G I o S ) )

e hF t M A B m a

a n n e P i a

i t I

S s r a r S I

S s

e (

t

(

t d H H c

• s s t e H I c e e c 2 2 u t t e f f S l a l a f f

o o m P o s

o s A 1 1 D i i*

S W

F W

F 1l

[ August Spector - FPE_ INS.wpd Pags 1l t

Pre-decisional Draft (Workin progress)

Revision Draft 1 (May 6,1999, Master, rev I)

Evaluation Guidance Determining Potential Risk Significance of Fire Protection and Post-fire Safe Shutdown Inspection Findings Fire Protection Engineering Section (FPES)

Plant Systems Branch (SPLB) and Safety Programs Section Probabilistic Risk Assessment Branch (SPSB) ,

D!v'.sion of Systems Safety and Analysis (DSSA)

Office of Nuclear Reactor Regulation (NRR) )

{

l 1.0 Introduction (

l l

The fire protection cornerstones are the accepted fire protection defense-in-depth principles. These l principles are:

I

1) Prevent fires from starting;

2) Detect and suppress those fires that do occur; and

3) Provide protection for structures, systems, and components important to safety so that a fire that is not promptly extinguished by fire suppression activities that it will not prevent the safe shutdown of the plant'.

A fire protection program finding can generally be classified as a weakness associated with meeting the objectives of one of these defense-in-depth (DID) principles. As a result, this screening methodology was developed to evaluate the significance of potential fire risk due to an identified fire protection DID weakness or inspection finding. Since potential fire protection DID weaknesses or findings for a fire area, zone or room under consideration have a synergistic impact on risk, those findings for a given fire area are treated collectively by this Fire Protection Risk Significant Screening I Methodology (FPRSSM) to arrive at an overall quantification of the relative impact these findings may have on fire risk.

2.0 Purpose

' Fire protection features sufficient to protect against the fire hazards in the area, zone or room under consideration shall be capable of assuring that necessary structures, systems, and components need to achieve and maintain safe shutdown are free of fire damage (see Section Ill.G.2a, b, and c of Appendix R to 10 CFR Part 50),

that is, the structure, system, or component under consideration is capable of performing its intended function during and after the postulated fire, as needed.

o l August Speder- wt:_ INS.wpd Page 2]

Pre-decisionalDraft (Work in progress)

In order to determine the relative significance of potential fire risk due to an observed weakness or weaknesses in the licensee's fire protection program, the inspector / reviewer needs to perform a screening evaluation which considers each mitigating element of the fire protection DID philosophy (e.g., detection, suppression, and passive protection separating post-fire safe shutdown functions).

, These evaluations are a qualitative assessment of the degree of change within each of the fire mitigation DID principles. Based on the results of these evaluations an integrated assessment of the potential fire risk significance of these inspection finding (s) can be performed using the FPRSSM described in Section 4.0.

The purpose of this evaluation guidance is to assist the regional inspector or the NRR fire protection reviewer in determining the potential risk / safety significance of: 1) a set of fire protection inspection findings relative to a given fire area, zone, or room; 2) a fire protection design condition which deviates from the intent of the facilities licensing / design basis; 3) an engineering evaluation documenting a change in the licensee's fire protection program; or 4) a licensee's request to change

. certain post-fire safe shutdown fire protection features or capabilities through an amendment to its operating license fire protection condition or through an exemption to certain fire protection regulatory requirements. For the purpose of this guidance, findings will be defined as conclusions or factual observations of those *in-p! ant" conditions which do not meet regulatory requirements, do not conform to the facilities operating license fire protection condition, or are considered to have risk implications due to a inherent fire protection / post-fire safe shutdown system design weakness.

3.0 Scope The FPRSSM and its implementing guidance evaluates the potential risk significance of a weakness (es)in a fire protection program and its implementation by focusing on these areas:

e IPEEE fire risk analysis e safe shutdown capability

. fire barrier effectiveness-e fire detection / automatic suppression systems e manual suppression effectiveness The licensee's IPEEE fire risk assessment provides insights on the risk associated with a limited set of postulated fire scenarios thet generally evaluated the risk impact associated with relatively small, non-bounding, type fire in defined areas of the plant. The IPEEE for a specific plant was performed as a one time analysis and its intent was to determine if any additional fire vulnerabilities that were not addressed by compiiance with the fire protection requirements existed. Because of methodological issues affecting the current fire risk assessment (FRA) process, IPEEE results and the insights from predicted probabilities of fire-induced damage should be used with caution when

' evaluating the significance of the plant's inability to cope with a challenging fire. In addition, when using the IPEEE results to make risk judgements, the basis of the assumptions used in the analysis and the screenings made should be weli understood .

Additionally, guidance is provided for making qualitative judgements and evaluating identified 0 technical issues associated with fire protection DID principles important to controlling and suppressing a fire and protecting safe shutdown capability. Section 5.0 provides guidance for

' l August Sped (4 - wt:_ INS.wpd

~

Page 3l Pre-decisionalDraR (Wortin progress) developing the basis for potential fire scenarios leading to conditions that could challenge fire mitigation features associated with the protection of safe shutdown capability and maintaining at least one train of this capability free of fire damage.

Section 8.0 provides guidance for assessing the significance of findings associated with fire barriers separating redundant safe shutdown trains within a fire area. This guidance includes qualitative assessment criteria for determining the potential significance of findings associated with a fire area

~

boundary barrier separating post-fire shutdown paths.

Section 7.0 establishes guidance for assessing the effectiveness of automatic fire sprinklers, Carbon Dioxide (CO2), and Halon suppression systems. Certain fire suppression systems are actuated, or given the permissive to actuate by the fire detection system. Section 7.1 provides guidance on assessing the effectiveness of a fire detection system by performing a general evaluation of the placement and spacing of fire detection devices within an area under consideration. When automatic fire detection systems are used in conjunction with automatic pre-action sprinkler, water spray or water mist, CO2, and Halon systems, the evaluation of automatic fire suppression effectiveness is dependent on the adequacy of the fire detection system design and

~

its ability to react to a potential fire. In addition, the overali plant fire detection capability and the acknowledgment of a potential fire by plant operators, will affect the ability of the fire brigade to exting'lish a fire, even if it is controlled by an automatic fire suppression system. Section 8.0 provides the guidance for evaluating the fire brigade and making judgements regarding their skill and ability to successfully extinguish the fire or confine it to the room of origin.

4.0 Fire Protection Risk Assessment Methodology The FPRSSM is an integrated process that can be used to assess the relative risk significance of identified weaknesses in the fire protection DID principles in a given fire area, zone, or room under consideration. The following steps describe the general process that should be followed when implementing this methodology (see Figure 4-1, Firs Protection Risk Significance Screening Methodology - Process Diagram).

Sten 1 Grouping of Fire Protecdon and Post-Rrw Safe Shutdown Findings The specific fire protection inspection findings affecting the fire protection mitigation DID features are grouped together for each specific fire area, zone or room under consideration. Then a area specific fire scenario needs to be considered and postulated to occur. Step 2 provides guidance for defining fire scenarios. Step 1 and Step 2 should be performed during an inspection in an integrated manner

- (i.e., observations of a fire protection degradation and the related fire hazards in the area of

. concem).

Step 2 DeRne the Fire Scenario in order to properly support the FPRSSM risk estimates, the inspector or the reviewer will need to develop a postulated fire damage scenario which describes the fire and its potential for propagation (see Section 8.0, Fire Scenario Considerations) within the fire area, zone or room under consideration. Under this postulated scenario, the inspector or reviewer will need to make deterministic / qualitative judgements regarding the effectiveness of various degraded fire protection mitigation feature or systems and their ability to protect a post-fire safe shutdown path and maintain it free from fire damage.~ Postulated fires involving fuel sources in an area under consideration are

j August spees- we:_ INS.wpd Page 4l Pre-decisional Draft (Workin progress) deemed meaningful if they are capable of developing a plume and/or a hot gas layer that has the potential to directly affect components of equipment that are important to safety.

l ' Step 3 Qualitative Evaluation of Findings j

' Once the fire area, zone, or room affiliation for the various inspection DID findings and a meaningful l

. fire scenario have been established, the individual findings have io be evaluated with respect to their t

ability to satisfy the performance objective established by the applicable DID principle. Upon making j the determination of which DID principles have been affected by the specific fire protection weakness of finding, a qualitative avaluation of each finding and its effects on accomplishing the DID i objective is performed. It should be noted that many inspection findings can contribute to a l degradation in a DID principle. For example, poor training, poor fire brigade / operational drill performance, improperly installed detection, and inadequate hose coverage of a fire area can all contribute to the degradation rating assigned to manual suppression. Therefore, in order to perform this step, the existing plant conditions noted by the inspection finding are evaluated against the deterministic / qualitative evaluation guidance and degradations categorization criteria established by Section 6.0, Passive Fire Protection Features, Section 7.0, Detection / Automatic

- Suppression Effectiveness, Section 8.0, Detection / Manual Firefighting Effectiveness, and /or -

Section g.0, Safe Shutdown Capability.

The output from this deterministic / qualitative evaluation, results in a degradation qualitative rating (DQR) (e.g., High, Medium, or Low)2 being assigned to each DID principle inspection finding).

' Step 4 Assignment of Quantitative Values From the Step 3, " Qualitative Evaluation of the Findings," a DQR is assigned to each DID principle. Once the DQR for the findings have been determined, they are quantified by  ;

assigning a value from lookup Table 4.1 that corresponds to the DQR for each DID principle l category. '!

Step 5 Integrated Assessment of DID Findings Once Step 4 has been completed, then the respective DID findings for a given fire area, zone, or I

. room are assessed collectively by using lookup Table 4.2.

The output from the process described by lookup Table 4.2 is an estimation of the potential significance of the fire protection DID inspection findings associated with the specific fire area,  ;

zone, or room being evaluated. This significance, at this point. is based solely on the  !

, quantification of the DID finding and its perceived degradation, and does not consider the fire i

l 2

Note that the ratings for safe shutdown are extra high, high, rnedium, and low degradation qualitative )

t ratings._ Fire barners, detection / automatic suppression, and detection / manual suppression only have high, medium, or low degradation qualitative ratings.

)

l l

1 b l August Spector- wt:_ INS.wpd Paga 5]

l

\

. Pre-decisionalDraN (Workin progress) 1 ignition frequency (IF). The significance is called the Overall Degradation Qualitative Rating for i

. the area and is designated "Overall DQR."

The overall DRQ for the fire area has been determined in lookup Table 4.2 in two parts. First of

all, the individual quantified values for each of the DID findings are summed. Secondly, the sum is compared with the criteria in lookup Table 4.3. Table 4.3 provides criteria for determining the

)

overall DQR for the fire area being evaluated as extra-high, high, medium, low, or insignificant, j i Both the DQR for the individual DID findings and the evaluation criteria for scoring an overall '

DQR are exponents of 10.

Adjustments have been made in lookup Table 4.2 to the score of the fire area overall DQR to account for dependencies between DID elements. The first adjustment is made to the score of a fire area when a high degradation of the fire brigade occurs with low and medium degradations l . of automatic suppression. When automatic suppression degradation is low or medium, and the l fire b'igade degradation is high, -0.75 is added to the score for the fire area. .This adjustment l has the.effect of providing partial credit for automatic suppression when it has a low degradation and is paired with high degradation of fire brigade. In this manner, no credit is provided for automatic suppression when it has a medium degradation and is paired with a high degradation l of the fire brigade.

The other included dependency exists between cases where no degradation exists in either the

~

automatic suppression or fire brigade. In this case, -0.5 is added to the score for the fire area.

l This adjustment is made since common cause mechanisms (common piping, valves, and L similarities between pumps) may exist between the automatic suppression system and l standpipe for the fire brigade. Since overall suppression effectiveness is underestimated under some conditions, this correction is also includsd for all combinations of low automatic j suppression degradation and low manual (fire brigade) degradation, i

l l ' Step 6 IdendRcadon oflanidon Frequency and Agustment by Degradation Time The next step.is to select the ignition frequency in lookup Table 4.4a* If several rooms of a single type exist, the ignition frequency should be divided by the number of rooms. For 3

Generic ignition frequencies for specific buildings or rooms are provided in lookup table Table 4.4a l (taken from AEOD data base, NRC *Special Study: Fire Events O- Feedback of U.S. Operating Experience - Final I Report,." June 19,1997) ]

)

i l

t1

i l August Sp.c.ici- wt:_ INS.wpd Pag 66]

. Pre-decisionalDraN (Workin progress) example, if two cable spreading rooms exist per unit, the ignition frequency per cable spreading room is 5E-3/ reactor-year divided by 2. Note that the ignition frequencies for a building will be conservative when assigned to a fire area within the building since several fire areas can

- comprise a building.

' After the ignition frequency for the fire area is determined, lookup Table 4.4 is entered. Each

. range of ignition frequencies in lookup Table 4.4 is represented by the uppermost value in the ,

band for calculational purposes. For example, for purposes of performing manual calculations  !

with this methodology, the ignition frequency is rounded up to the next decade. However, if the l frequency is a only a factor c' 2 greater than the lower decade, the lower decade can be used 1 l for the ignition frequency.' in this manner, most risk significant areas will have comparable risk <

ifor the same degradations in DID.

The ignition frequency is then adjusted in lookup Table 4.4 by the time that the degradation existed. In practice, we expect the inspector to assume that the degradations are simultaneous,

! and all occur for the length of time associated with the longest degradation. This is a l, conservative approach, and if desired, can be refined as part of a more detailed analysis. The

degradation of 3-30 days decreases the frequency by 10, and the degradation of less than 3 i days decreases the risk by 100.

l: Step 7 Correlation of Fire Protection Findings to the Inspection Finding Risk

l. Characterization Process The overall DQR which represents the relative risk significance of the DID findings is then combined with the adjusted ignition frequency in lookup Table.4.4 to obtain an overall potential risk significance estimation for the fire area, zone, or room under consideration.

1.

In the FPRSSM, the CDF essociated with the impact of the DID findings is strictly what is calculated. However, for purposes of using this model, the CDF due to DID findings will be considered as the ACDF. This is conservative since the CDF due to the DID findings is greater than ACDF.

The colors which represent changes in CDF in the FPRSSM are correlated to the identical changes in CDF associated with the inspection Finding Risk Characterization Process. The colors noted in lookup Table 4.5 represent a ACDF value. These values correspond to the following ACDF:

ColorCharecearianglen Delenin Core Damese 7;; mi(CDP)

RED ACDF > 10 4 YELLOW - 104a ACDF > 10 5 WHITE 10-5 = ACDF > 104 GREEN ACDF s 10 4 The uppermost value in the range of ACDF belongs to that range to allow for a reduction due to fire that ignite, flame, smolder, and bumout without mitigation or intervention.

f

[ August Spector-FPE INS.wpd Pagg 7]

l Pre-decisionalDraft (Workin progress) l The FPRSSM is consistent with R.G.1.174, "An Approach for Using Probabilistic Risk l Assessment (PRA) in Risk-informed Decisions on Plant-Specific Changes to the Licensing Basis," July 1998.

Step 8 - General Rules for Applying FPRSSM Since a fire barrier failure is represented by a probability, the ACDF is a combination of two contributions: a contribution where the barrier fails, and one where the barrier succeeds. Table 4.1 can be used to calculate both of these terms. For purposes of discussion, the term referring to the case where the barrier fails will be called the double room term (DRT) and the case where '

the barrier succeeds will be called the single room term (SRT). The SRT and DRT are shown by the figure below.

Single Room Term (SRT) Fire Barrier Prevents Fire / Smoke Propagation j l

Fire Area B Fire Area C SSD Train A SSD Train B Fire affected area 3-hour fire barrier (fire barrier successful. No fire / smoke impact on Fire Area B 4

Double Room Term (DRT) Fire Barrier Falls to Prevants Fire / Smoke Propagation Fire Area B Fire Area C SSD Train A SSD Train B Fire affected area l

N 3-hour fire barrier fails (Fire / smoke impacts Fire Area B)

The safe shutdown (SSD) vulnerability for the DRT is the combination of mitigating equipment and actions in fire areas B and C. The SSD vulnerability for the SRT is the combination of mitigating equipment and actions in C alone.

As a result, the SSD DQR in Table 4.1 can be different depending on whether the SRT or DRT l

F1 l August Spector - r-Pt:_ INS.wpd Paga 81 i

i Pre-decisionalDraft (Workin progress) is calculated. Note that the mitigating equipment for the DRT is a subset or can be equal to the equipment for the SRT.

However, both the SRT and DRT are not needed in all cases. The following rules provide guidance on when to use these terms, i

a. For the case where the fire barrier has a high degradation, only the DRT is needed to calculate delta CDF.

b. For the case where the fire barrier has a medium degradation, only the DRT is needed to calculate the delta CDF. (Note that this is an approximation though.)

c. For the case where the fire barrier has a low degradation, both the SRT and DRT must be considered. In the case where one term dominates the other, naturally the total delta CDF can be considered as the dominant term. In cases where the SRT and DRT are approximately equal, it is suggested that these contributions be added. (See Appendix 1)

d. If no SSD equipment, components, or cable are on side of the fire barrier away from the origin of the fire, only the SRT is needed to perform the calculation. No reduction in CDF should be given by the barrier since it protects no SSD equipment. Note that adding the DRT in this case will not impact the results. (See Appendix 1.)

' Calculation of the DRT and SRT in Table 4.1 is as follows: The DRT is calculated by assigning ,

the DQR to the fire barrier condition that is found during the inspection. The SRT is calculated by assigning a value of high to the DQR for the fire barrier of concern. No fire barrier is credited ,

for the SRT since the equipment in the room has no fire barrier to prevent fire damage to the I equipment or component of concern. As indicated in the above discussion, the SSD mitigating capability for the SRT will be greater than or equal to that for the DRT. I i

Step 9 Agusdng the Fire Area Risk SignlRcance Under Spurious Actuations l Inspection findings regarding fire-induced spurious equipment actuations may occur. In order to evaluate these findings, the risk significance of the fire area will need to be adjusted due to the fact that spurious actuations may not necessarily occur in the presence of a fire. To evaluate the risk significance of a finding related to a single spurious actuation, the frequency of the area i' must be reduced by 10. To evaluate the impact of multiple spurious actuations, the frequency of the fire area is retained at the value representative of the single spurious actuation. In this manner, an upper bound for the risk significance of the multiple spurious actuations is attained.

Further reductions in the score of the fire area to reflect lesser dependence between single and multiple spurious actuations, or to reflect recovery, should be supported by engineering judgement and further detailed evaluations.

5.0 Fire Scenario Considerations

(

I in order to perform a screening risk significance estimate of the fire protection DID weaknesses / findings, a reasonable fire scenario, based on in-situ conditions and allowed l

l i

n

. l August Specisi- rPt:.JNS.wixi ' Prga 9]

i l

Prw4ecisionalDraN (Workin progress) operational practices, shall be d' eveloped. Therefore, it will be the inspector's responsibility to develop a fire scenario for the fire area, zone, or room under consideration. This will include evaluating the fuel availability, its distribution, and its relationship to post-fire safe shutdown -

mitigation systems, equipment, and components, and potential ignition sources. Do not

' assume that all equipment is arbitrarily lost by a postulated fire. A more realistic impact of fire on the equipment will need to be considered. This fire scenario should consider the relative location of fire sources and their relationship to safe shutdown and accident mitigation equipment, the heat release rates of these combustible, and it the amount of material available to burn is sufficient to sustain a fire for an appreciable duration.

The following are general assumptions should be used as guidance to assist with the development of a postulated fire scenario:

GeneralAssumptions

1. An ignition source is present that is sufficient to ignite the initiating fuel package and the fire achieves its peak heat release rate. Under most cases, it will be argued that an ignition of a fire source occurs with a low frequency however, its occurrence may result in the greatest loss or consequence. Therefore, for this screening method, a conservative approach will be taken and it will be assumed that ignition occurs.

2. - The maximum transient fire loads allowed by administrative controls, if not restricted from the fire area, zone, or room under consideration, can be considered a initiating fuel package.

3. The presence of extemal ignition sources (e.g., welding, cutting, grinding, temporary wiring) allowed by administrative controls can be present and considered a potential ignition source.

4. For those fire areas, zones, or rooms under consideration where the in-situ fuel sources do not present a hazard (e.g., no exposed cables or other fuel sources) assume a transient fuel and external ignition source as the fire scenario.

5. Plant electrical equipment (e.g. motor control centers, switchgear, relay panels, termination cabinets, motors, MG sets, transformers) are assumed to be ignition and fuel source.

6. Fires (plume and the ceiling jet) a#ecting cable trays of two or more, that are in close proximity to the ceiling, are assumed to propagate.

7. Fires in electrical cabinets, if the tops have ventilation openings or are unsealed, are then assumed to be open. Therefore, a fire can expose and ignite the cables directly Labove the cabinent of concem.

. 8. . Faults in high voltage switchgear can breach a metal cabinet, and are assumed to cause faults in adjacent switchgear and ignition of adjacent fuel sources.

I i

i

.]

i. . l August Sp60wr - FPt_ INS.wpd Pcgg 10]

Pre-decisionalDraN (Workin progress)

9. Exposure fires involving transient combustibles are assumed to have an equal probability of occurring anywhere in the space being evaluated or inspected. Fires involving fixed (in-situ) combustibles are assumed to occur at the site of the combustible and propagate accordingly within and along a contiguous fuel package.

10. At a ventilation flow rate of 10 room air changes per hour, assume that a dense layer of smoke (from floor to ceiling) can develop within a fire area, zone, and room under consideration in 5 to 15 minutes after fire initiation.

11. Since fire damage data of SSD and recovery system equipment and components is extremely limited, it is assumed that if unprotected (no fire resistive barrier) SSD and recovery equipment / components that are in the fire's plume or located in the ceiling region are likely to be damaged.

Guidance The first step is to identify a fire scenario for a given fire area, zone or room under consideration.

In addition, this will require that the general location of the post-fire SSD systems, equipment, and components and any recovery (EOP type) systems, equipment, and components be identified within the given fire area, zone or room under consideration. Proximity of combustibles and their relationship to SSD and/or recovery equipment will need to be observed.

For example, the SSD and/or recovery equipment of concem are located near the ceiling and the in-situ fuel packages (combustibles), such as cables in cable trays, are located within this same region of the room'.

SSD and/or systems, equipment, and components are considered to be targets which are subject to fire damage. These targets can be either in the ceiling jet layer (upper hot gas layer i portion of the room) that forms directly beneath the ceiling, or in the fire's plume region, or in the i sub-layer that is beneath the ceiling jet layer.  !

Generally, a fire presents the greatest challenge when it is located directly beneath a target such as the case of a floor based exposure type fire involving transient combustibles. For fixed (in-situ) combustibles, actual geometry of the fuel packages (along the wall, in the corner, near the ceiling) between the fire source and the target needs to be assessed in order to determine if the targets of concem are located within the fire plume or in the ceiling jet region.

Basic ignition of secondary fuel packages is generally attributed to convection, conduction, radiation, or a combination of these energy (heat) transfer methods. Conduction occurs when l the fuel packages are in direct contact with each other and heat directly transfers from one l package to the other. Convection occurs when the heat in the fire plume carries heat to the i

• It should be noted that this type of assessment of the fuel configuration and distribution is not the same as the fire loading (btus/sq. ft) calculations performed for the plant specific Fire Hazards Analysis (FHA).

I

l August Spector - FPE_ INS.wpd ^

Pags 11l l l

1 Pre-decisionalDraft (Workin progress) secondary fuel packages and radiation is typically the main mode of heat transfer to adjacent

~ fuel packages (e.g., two adjacent fuel packages at the floor level and not in contact with one

. another); Radiation is dependent on the size of the flame, temperature, emissivity of the flame,

, absorptivity of the fuel package (combustible) surface, geometric viewing factor between the -

flames and the fuel package surface, and the fuel package ignition characteristics.

Assuming ignition, the combustion characteristics of the fuel packages needs to be evaluated.

For example, the state of the fuel (solid, liquid, gas), type and quantity, configuration and location, rate of heat release, rate of fire growth, and production rate of combustion products must be considered. As a fire, due to the initial fuel package, begins to grow in intensity, it can produce sufficient convective, conductive, and radiant energy to ignite adjacent fuel packages l (e.g., Floor based fire exposes one bank of cable trays in the upper regions of a room; then this l burning bank of cable trays ignite, a second and adjacent bank of cable trays within the same  !

upper region). Thus, the fire scenario is deve!oped.

.6.0 Fire Barrier integrity I

l The following evaluation guidance is to be used for making qualitative judgements relating to the l

general effectiveness of passive fire protection features used to protect post-fire safe shutdown capability or prevent a fire from spreading from one fire area, zone or room to another.

1

. General Assumptions- l l

1. The fire wall, ceiiing, floor or raceway / equipment fire barrier of concern is known to i separate redundant trains of systems, components, or equipment required for plant  !

shutdown.

2. The in-situ (fire load) is meaningful and is in a configuration that would directly challenge l the passive fire barrier or fire resistive device under consideration (e.g., A set of cable I trays in the overhead penetrating a fire wall and a combustible liquid is spilled in close  !

' proximity of this barrier penetration system).  !

3. Compensatory measures are not viewed as risk equivalent functions and are not credited by this methodology. (e.g., Blocked open or missing fire door which can not be closed during a fire and is under an hourly fire patrol does not assure that the risk is reduced).

Evaluation Guidance L For degradations related to silicone foam based penetration seals5 see the table at the end of L this section.

! i r-5 The guidance table for penetration seal degradations assumes that the silicone material is mixed and its cell structure is in accordance with the manufacturers recommendations and guidelines.

- [ August Spector - FPE_JNS.wpd P9312]

l l

Pre-decisional Draft (Workin progress) l High Degradation Categories The following are examples of high degradation categor%s:

(a) Removed or missing fire barrier protecting or separating redundant safe shutdown l systems or components.

(b) Breach in a electrical raceway fire barrier system which is contained with in a fuel package (e.g., barrier system is in a cable tray stack or within the postulated fire's plume i or ceiling jet).

(c) Fire barrier system design which is mis-applied or with a indeterminate fire resistive rating (d) Ceiling fire barrier system with unsealed openings. The lower room or fire area under consideration must have a meaningful fire load.

'(e) Un-analyzed; unprotected openings in a fire area / barrier wall. These openings fall within the upper half of the wall. The fire area / room under consideration has a meaningful fire load and it is located in the area of the opening.

(f) In operable fire door or damper in a fire areal barrier wall. The fire area / room under considerafon has a meaningful fire load and it is located near the area of the opening.

(g) Blocked open fire doors.

Medium Degradation Categories i

The following are examples of Medium degradation categories:

(a) Fire dampers installed in fire barrier assemblies which are not qualified to close under anticipated ventilation system air flow conditions. '

(b) Fire dampers installed in fire barrier assemblies whbh are not installed with the required thermal expansion clearances as determined by the conditions of its qualification testing.

(c) Improper temperature fusible link setpoint or installation. These links are generally used to activate fire door / damper closure.

(d) Bent or warped fire door.

(e) Fire door with a single side through hole.

(f) Excessive fire door to frame and door to floor clearance gaps.

[ August Sgtor - FPE_ INS.wpd page 13]

Pre 4ecisionalDratlt (Workin progress)

'(g). . Improperly installed or qualified fire door hardware.

'(h) Raceway or equipment fire barrier assembly which has been mechanically damaged and the fire barrier wall thickness has been reduced by 25 percent.

(i) .. Penetration seal assembly which are not qualified by test or analysis (e.g., thermal penetration mass is greater than that tested).

Low Degradation Categories' The following are examples of low degradation categories:

l(a) Fire door installed and maintained in accordance with industry standards.  ;

~

(b)l Fire damper installed and maintained in accordance with industry standards.

'(c) Fire barrier penetration seal installed in accordance with th'e construction attributes and conditions qualified by fire tests.

(d) Raceway and equipment fire barrier assemblies installed in accordance with the construction attributes and conditions qualified by fire tests.

(e) Fire walls / barrier assemblies installed in accordance with tne construction attributes

and conditions qualified by fire tests

, QUnwnCE PORDE1EnanNn00 MntBARRER PanEMATION SEAL TNMKNESS w.

t'

~ 4 OEGRADATNW CATNGORES M' <<

W 1

's , 'w

.m

• ?r '

. , n y .f

,,g

_ 31

• h <

~

. Tw .

4 0 TO 30 PERCENT 30 TO 80 PERCENT 80 TO 100 PERCENT PERCENTAGE OF PENETRATION SEAL MATERIAL (REQUIRED)

l. THICKNESS REMAINING IN PENETRATION OPENING 7.0 Detection / Automatic Fire Suppression Effectiveness

. 7.1 Automatic Fire Detection Effectiveness The following evalu' ation guidance is to be used for making qualitative judgements relating to the general effectiveness of certain automatic fire detection features used to promptly detect a fire within the fire area, zone, or room under consideration.

GeneralAssumptions i l

i

[ August Spector - FPE_ INS.wpd Pag 314]

l t

Pre-decisionalDraft (Workin progress)

1. The performance objective of the fire detection system is to rapidly detect a fire. It is assumed that detector spacing that meets the minimum spacing requirements and l performance-based guidance specified by current industry codes will assure that the performance objectives are met.

2. It is assumed that all fire detection (initiating) devices are installed at or near the ceiling.

3. It is assumed that the fire alarm system code requirements have been met for the fire detection system and notification devices.

4. Compensatory measures are not viewed as risk equivalent functions and are not I

credited by this methodology. (e.g., inoperable fire detection under an hourly fire patrol does not assure that the risk is neutral).

Evaluation Guidance

• The fire detection element is critical in that it senses a potential fire condition and completes the logic of the system that provides notification to the control room to alert the operators of a pending fire condition and in some cases it also actuates the fire suppression system in the area, zone, or room under consideration in order to judge the effectiveness of the fire detection and notification capabilities it is important to understand the application and placement of fire detectors. The following specific guidance is provided in order to establish a basic understanding. A review of the layout and placement of the detection (initiating) devices within those fire areas, zones, or rooms under consideration will be required. The following guidance is provided to establish a basis and general rules for making these judgements regarding the effectiveness of the design.

Generally, two basic types of fire detection devices are used. They are products of combustion (POC) type detector or thermal detectors. The majority POC detectors are ionization or photoelectric spot (smoke) type and they are listed by Underwriter's Laboratories (UL) to be placed on a smooth ceiling, with a ceiling height that does not exceed 15 feet 9 inches and a maximum spacing of 30 feet between detectors. The detectors, during their listing approval tests, are not subject to any air movement and there are no physical obstructions between the detector and the fire source.

With respect to thermal detectors, generally there are two types fixed temperature and rate compensated. The UL listing for a fixed temperature and rate compensated is related to an area of coverage. For example, a fixed temperature detector can generally be used to protect a maximum of 225 square feet and a rate compensated detector can be used to protect a maximum of 2500 square feet with a 50 foot spacing factor.

s Refer to the NFPA National Fire Alarm Code Handbook,1993 Edition, for further guidance l

l

I [ August Spector- FT51E_. INS.wpd Paga 16]

l Pre-decisional Draft (Workin progress)

Judging Adequacy of Spot Type Thermal Detector Placement - Minimum Factors

1. Spot type detectors shall be located on the ceiling not less than 4 inches from the side wall or on the side walls between 4 and 12 inchas from the ceiling.

2. Reduced spacing shall be considered and may be required due to structural obstructions and characteristics of the area being protected. For smooth ceilings the distance between detectors shall not exceed their UL listed spacing and there shall be a detector within one half of the listed spacing, measured at right angle, from all walls or partitions extending to within 18 inches of the ceiling, or all points on the ceiling shall have a detector within a distance equal to 0.7 times the listed spacing.

3. The maximum linear spacing on smooth ceilings for spot type heat (rate of rise or rate compensated) detectors are determined by full scale fire tests. These tests assume that the detectors are to be installed in a pattern of one or more squares, each side of which equals the maximum spaced as determined in the test. The distance from the detector to the fire sha!I be maintained always at the test spacing multiplied by 0.7. (See table below)

TEST SPACING MAXINUM TESTDISTANCE FROM FIRE TO DETECTOR 50 X 50 FEET 35 FEET 40 X 40 FEET 28 FEET 30 X 30 FEET 21 FEET 25 X 25 FEET 17.5 FEET 20 X 20 FEET 14 FEET 15 X 15 FEET 10.5 FEET

4. On ceilings 10 feet to 30 feet high heat detector spacing shall be reduced in accordance with the table below:

CEILING HEIGHT '

UP TO PERCENT OF UL LISTED (FT) . SPACING -

0- 10 100 10 12 91 12 14 84 14 16 77 16 18 71 18 20 64 20 22 58 22 24 52 24 26 46 26 28 40  !

28 30 34 l 1

5. A ceiling shall be treated as a smooth ceiling if the beams project no more than 4 inches below the ceiling. If the beanis project more than 4 inches below the ceiling, the spacing )

l of spot type heat detectors shall be at right angles to the direction of the beam travel and I shall not be more than 2/3 of the smooth ceiling spacing. If the beams project more than

i l August Spector - FPE_ INS.wpd Pags 16]

Pre-decisional Draft (Workin progress) 18 inches below the ceiling and are more than 8 feet on center each bay formed by the beams shall be treated as a separate area and have at least one detector installed within the bay.

Location and spacing of heat detectors should consider beam depth, ceiling height, beam spacing, HVAC vents and effects, obstructions, and fire size.

(a) If the ratio of beam depth (D) to ceiling height (H) (D/H) is greater than 0.10 and ratio of beam spacing (W) to ceiling height (H) (W/H) is greater than 0.40, heat detectors should be placed in each beam pocket.

(b) If either the ratio of beam depth to ceiling height is less than 0.10 or the ratio of beam spacing to ceiling height is less than 0.40, heat detectors should be installed on the bottom of the beams.

Judging Adequacy of Spot Type POC Detector Placement - Minimum Factors

1. Spot type detectors shall be located on the ceiling not less than 4 inches from the side wall or on the sidewall between 4 and 12 inches down from the ceiling to the top of the detector.

2. On smooth ceilings, spacing of 30 feet shall be permitted to be used as an initial criteria.

All points on the ceiling shall have a detector within a distance equal to 0.7 times the selected spacing. General guidance for spacing of spot type smoke detectors on smooth ceilings 10 feet to 30 feet high' is provided in the table below:

Fire Size (Btus/second) and Growth ' Maximum Celling ' ' Maximum Rate .

- height (feet) Spacing (9t) 100 btus/sec - fire growing at a slow rate 10 22 15 15 18 12 250 btus/sec - fire growing at a slow rate 10 40 15 35 18 30

, if1...HS,yy3,C:W M; M % p;4 7:yp h yg? Q 100 btus/sec - fire growing at a medium rate 10 18 15 12 18 N/A 8

it is assumed that the ratio of the gas temperature rise to the optical density of the smoke is a constant and that the detector will actuate at a constant value of temperature rise equal to 20*F, which is considered indicative of concentration of smoe from a number of common fuels that would cause detection by a relatively sensitive detector.

I i

. [ August Specs - FPt:_ INS.wpd Page 17l Pre-decisionalDraft (Workin progress)

- 250 blueleec - fire growing at a medium rate 10 35 15 28 20 24 25 18 28 12

L + J ... . / . . . . J;; %

g.4

..9.,. .

100 btuelsec - fire growing at a fast rate 10 12 15 N/A

_ _ q :; ;g 3.j; q. ; , ;y .3,; y_[:.y .y, p.gq:p

7 ;. - .., ,

250 btusisec - fire growing at a fast rate 10 28 15 20

' 20 14

3. Ceiling construction where beams are 8 inches or less in depth shall be considered equivalent to a smooth ceiling. If the beams are more than 8 inches in depth the spacing of spot type detectors in the direction perpendicular to the beams shall be reduced. If the beams are less than 12 inches in depth and less than 8 feet on center spot type detectors shall be permitted to be installed on the bottom of beams.

4. If the beams project more than 18 inches below the ceiling and are more than 8 feet on center each bay formed by the beams shall be treated as a separate area and have at least one detector installed within the bay. 1 Location and spacing of heat detectors should consider beam depth, ceiling height, beam spacing, and fire size. To detect a flaming fire (strong plumes), detectors should i be installed as follows:

(a) If the ratio of beam depth (D) to ceiling height (H) (D/H) is greater than 0.10 and ratio of beam spacing (W) to ceiling height (H) (W/H) is greater than 0.40, heat detectors should be placed in each beam pocket.

(b) If either the ratio of beam depth to ceiling height is less than 0.10 or the ration of beam spacing to ceiling height is less than 0.40, heat detectors should be installed on the bottom of the beams.

To detect smoldering fires (weak or no plumes'), detectors shall be installed as follows:

(a) If air mixing into the beam pockets is good (e.g., air flow parallel to long beams) and condition (a) exists as above, detector shall be located in each beam pocket.

(b) If air mixing into the beams pockets is limited or condition (b) exists above, 8 - Most fires modeled as part of the plants IPEEE were of these type of fires. Therefore, based on the

. plant "as-built

• conditions (detection spacing greater than recommended to detect weak fire conditions) the assumptions regarding detection effectiveness and fire brigade or automatic suppression system response may have been overty optimistic.

l

[ Augu;t Spector - FPE_ INS.wpd Prg318]

Pre-decisional Draft (Workin progress) detectors should be located on the bottom of the beams.

5. The radius of a fire plume where it impingies on the ceiling is approximately 20 percent of the ceiling height (0.20H) above the fire source and the minimum depth of the ceiling jet is approximately 10 percent of the ceiling height (0.10H) above the fire source. For ceilings with beams deeper than the jet depth and spaced wider that the plume width, detectors will respond faster in the beam pocket because they will be in either the plume of ceiling jet. For ceiling with beams of less depth than the ceiling jet or spaced closer that the plume width, detector response will not be enhanced by placing detectors in each beam pocket, and the detectors may perform better on (for spot-type detectors) the bottom of the beams.

6. Where plumes are weak, ventilation and mixing into the beam pockets will determine detector response. Where beams are closely spaced and air flow is perpendicular to the beam, mixing into the beam is limited and detectors will perform better on the bottom of the beams.

High Degradation Categorization The following are examples of high degradation categories (a)- The detection system for the fire area, zone, or room under consideration is inoperable.

(b) Insufficient number of detectors (c) The placement and spacing of 25 percent of the detectors within the fire area, zone, or room under consideration do not meet the spacing / placement conditions of their UL listing or do not meet the general assessment guidance specified above.

Medium Degradation Categorization The following is an example of a medium degradation category:

(a) The placement and spacing of 10 percent of the detectors within the fire area, zone, or room under consideration do not meet the spacing / placement conditions of their UL listing or do not meet the general assessment guidance specified above.

Low Degradation Categorization The following is an example of a low degradation category:

(a) The layout and placement of fire detection devices within the fire area, zone, or room under consideration meet industry codes and the conditions of the detection device's Underwriter's Laboratories listing and testing approvals.

7.2 Fixed / Automatic Fire Suppression Systems

[ August Spector- wt:._ INS.wpd Paga 19]

Pre-decisionalDraft (Workin progress) 7.2.1 Automatic Sprinkler Protection GeneralAssumptions Compensatory measures are not viewed as risk equivalent functions and are not L credited by this methodology (e.g., Additional manual fire fighting equipment staged in the area of the degraded or inoperable automatic fire suppression systems does not assure that the risk is neutral)..

GeneralEvaluation Guidance '*

The following evaluation guidance is to be_ used for making qualitative judgements relating to the general effectiveness of certain automatic fire suppression features used to promptly suppress and control a fire within the fire area, zone, or room under consideration.

1. Sprinklers shall be installed in accordance with their UL listing.

2. Ordinary-temperature-rated sprinklers shall be used throughout Nuclear power plant buildings. Where maximum ceiling temperatures exceed 100*F, sprinklers with temperature ratings in accordance with the maximum ceiling temperatures" noted below shall be used

• " glaximum? ' ; Sprinider *

, 1Sprinider/ . " Sprinkler ^ '

! Glass bulb - -W

a

Coning 1 (Temperature temperature O , temperature rating i colors :

e Temperature (F)12 reting(F)i,, claeolfication . _._,_a b._,2 100 135 to 170 Ordinary Uncolored Orange or Red 150 175 to 225 Intermediate White ' Yellow or Green 225 250 to 300 High Blue Blue 300 325 to 375 Extra High Red Purple 375 400 to 475 Very Extra High Green Black

3. Early suppression fast response sprinklers shall be used only in wet pipe sprinkler  ;

systems.

l 4. . The distance from sprinklers to walls shall not exceed one-half of the allowable distance between sprinklers. Sprinklers shall be located a minimum of 4 inches from wall.

f y - 5.~ " ~ Non-continuous obstructions at or very near the ceiling and close to the sprinkler such as i

' " Refer to Automatic Sprinkler System Handbook, Sixth Edition, for additional guidance l The maximum ceiling temperature is equated to the temperature that would be experienced at the ceiling on the hottest summer day (Summer High) i i

s

n 1 l [ August Spector - FPE_lNS.wpd P g3 20]

Pre-decisionalDraft (Workin progress) columns, cable trays, light fixtures, large pipes, HVAC ducts shall be treated as vertical obstructions. The minimum separation between vertical obstructions and a sprinkler shall be as follows:

' Minimum distance from vertical obstruction

' Maximum dimension of obstruction' Maximum horizontal distance sprinkler shall be

. placed away from obstruction

% to 1 inch 6 inches Greater than 1 inch and less than 4 inches 12 inches Greater than 4 inches 24 inches

6. The minimum separation of a sprinkler from a horizontal obstruction (beams, HVAC ducts) shall be determined by the height of the sprinkler deflector above the bottom of the obstruction shall be as follows:

Position of sprinkler deflector when located above bottom of obstruction Distance from sprinkler to side of. Maximum allowable distance of deflector above  ;

obstruction. bottom of obstruction less than 1 ft. O in.

1 ft to less than i ft-6 in. 1 in. l 1 ft-6 in. to less than 2 ft. 1 in.

2 ft. to less than 2 ft-6 in. 2 in. j 2 ft-6 in. to less than 3 ft. 3 in. '

3 ft. to less than 3 ft-6 in. 4 in.

3 ft-6 in. to less than 4 ft.. 6 in.

4 ft. to less than 4 ft.-6 in. 7 in.

4 ft-6 in. to less than 5 ft 9 in.

5 ft. to less than 5 ft.-6 in. 11 in.

5 ft.-6 in, to less than 6 ft. 14 in.

7. Under obstructed construction, the distance between the sprinkler deflector and the ceiling shall not be less 6 inches and more than 12 inches

8. Sprinklers shall be positioned with respect to lighting fixtures, cable trays, pipes, ducts and obstructions more than 24 inches wide and located entirely below the sprinkler so that the minimum distance from the near side of the obstruction to the center of the sprinkler is not sless than the value specified below:

Pooldon of sprinklers in reladon to obstrucdon located endroly below the sprinklers Distance of deflector above the bottom of - Minimum distance to side of obstruction (ft) the bottom of the obstruction Less than 6 inches 1 % feet 6 inches to less than 12 inches 3 feet 12 inches toless than 18 inches 4 feet i 18 inches to less than 24 inches 5 feet i 24 inches to less than 30 inches 6 feet

9. Where the bottom of the obstruction is located 24 inches or more below the sprinkler deflector: (a) Sprinklers shall be positioned so that the obstruction is centered between adjacent sprinklers; (b) The obstruction shall be limited to a maximum width of 24 inches.

Where the obstruction is greater than 24 inches wide, one or more lines of sprinklers

Mugust Spector - FPE_ INS.wpd P::g3 21]

Pre-decisional Draft (Work in progress) shall be installed below the obstruction; and (c) The obstruction shall not extend more than 12 inches to either side of the midpoint between sprinklers. When the extensions of the obstruction exceed 12 inches, one or more lines of sprinklers shall be installed below the obstruction.

10. In the special case of an obstruction running parallel to and directly below a branch line:(a) The sprinkler shall be located at least 36 inches above the top of the obstruction; (b) The obstruction shall be limited to a maximum width of 12 inches; and (c) The obstruction shall be limited to a maximum of 6 inches to either side of the centerline of the branch line.

11. A minimum of 18 inches of clear space below the sprinkler deflector shall be maintained.

High Degradation Categorization The following are examples of high degradation categories (a) The system is out of service or inoperable (b) Sprinkler head distance from the ceiling, at least 2 or more of the heads exceeds the I limits specified above.

(c) Two or more adjacent sprinkler heads in the area under consideration are affected <

obstructions (horizontal, vertical, or obstructions located below) and obstruction heads are not provided to compensate for the specific obstruction. See evaluation guidance above.

(d) Based on the specified ceiling temperature limits, the sprinkler head temperature ratings exceed the maximum temperature set-points recommended. See evaluation guidance above.

• l (e) The placement and spacing of 25 percent of the sprinklers within the fire area, zone, or room under consideration do not meet the spacing / placement conditions of their UL listing or do not meet the overall sprinkler head spacing assessment guidance specified above.

(f) If the system is actuated by the fire detection system and the degradation categorization of the fire detection systems is high, then the degradation categorization of the suppression syste,m would be high even when the degradation to the suppression system is mediu'n or low.

l Medium Degradation Categorization The following are examples of medium degradation categories:

r 1 l August Spector - FPE_ INS.wpd Pags 22]

l i Pre-decisionalDraft (Workin progress) l l

(a) Improper assessment of system performance or evaluation of internal system corrosion. '

l (b) The placement and spacing of 10 percent of the sprinklers within the fire area, zone, or

room under consideration do not meet the spacing / placement conditions of their UL l listing or do not meet the general assessment guidance specified above 1

(c) If the system is actuated by the fire detection system and the degradation categorization of the fire detection systems is medium, then the degradation categorization of the l suppression system would be medium even when the degradation to the suppression system is low.

Low Degradation Categorization The following is an example of low degradation category:

1 (a) The sprinkler system layout and head placement within the fire area, zone, or room l under consideration meets or exceeds the minimum industry code requirements and the  !

conditions of the sprinkler head UL listing and testing approvals.

l 7.2.2. Automatic Halon Systems (LATER - UNDER DEVELOPMENT) i l

GeneralAssumptions l l

General Guidance High Degradation Categorization Medium Degradation Categorization i

Low Degradation Categorization )

i 7.2.3 Automatic Carbon Dioxide Systems (LATER - UNDER DEVELOPMENT)

General Assumptions  ;

j i

General Guidance High Degradation Categorization Medium Degradation Categorization

l[ August Spector - FPE_ INS.wpd - Paga 23]

Pre-decisional Draft ^(Work in progress)

J j

Low Degradadon Categorizadon 8.0 Detection / Manual Firefighting Effectiveness j i

For guidance regarding how to evaluate the minimum effectiveness of the fire detection system and its ability to adequately react to a fire See section 7.1.

Manual fire fighting effectiveness under severe fire conditions is complex and difficult to assess, 1

Generally, event history has demonstrated that when faced with a challenging fire condition the  !

' effectiveness of plant fire brigades, in the absence of assistance from either fixed plant fire i protection features or offsite fire fighting support, have shown conditional limitations which have

)

impeded their ability to be effective. For example, weaknesses in actual fire brigade  !

performance is often a reflection of ineffective training, minimal fire brigade drill performance i expectations, incomplete fire fighting strategies (pre-plans), poor fire ground communications,  !

improper or inappropriate specialized fire fighting equipment and extinguishing agents, poor  !

application and logistics /stagging of specialized fire fighting equipment, inappropriate staffing, poor fire ground command and control, physical limitations of individual fire brigade members, etc.

In addition, manual fire fighting is affected by several time factors. Manual fire fighting l effectiveness is directly affected by how long (time) it takes for plant operations to accept or '

acknowledge the fire alarm and confirm that there is a fire. Once, plant operations has made the decision to respond the fire brigade (5 -10 minutes), the fire brigade has to react and then report  !

to the fire brigade equipment locker (s) (5 -10 minutes) and don protective clothing, SCBA, and prepare the appropriate special fire fighting equipment to take with them to the fira area, zone or room under consideration (7 -15 minutes). Upon completing the donning of the appropriate ,

protective equipment and selecting the initial fire fighting equipment to responded with, the i brigade responds to the area of concem (5 -15 minutes before the complete team is assembled

. near the area of concem). Once in the area, the fire brigade deploys and readies its equipment to fight the fire (5 -15 minutes). Once the equipment is setup, the brigade then make its an effort to control and suppress the fire (7-30 minutes under ideal condtions). Once the fire has been placed under control complete fire extinguishment can be accomplished (30 minutes - 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />).

Therefore, it is assumed that it takes from 34 minutes to 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> and 35 minutes for a fire brigade to control a challenging fire under ideal conditions. Time is a factor for fire growth and smoke development. For example, depending on the room size and the fuel burning, a dense layer of smoke (from floor to ceiling) can develop within a fire area, zone, and room under consideration in 5 to 15 minutes after fire initiation at a ventilation flow rate of 10 room air changes per hour.

Time is an important factor that needs to be considered. In addition to time, judgements will have to be made with regard to the skill of the fire brigade under strenuous conditions. Their ability to cope with the stress of a serious fire challenge and implement the guidance provided by the fire fighting (pre-fire plan) strategy are an equally important factors. These integrated factors (time, skill / equipment utilization) are best evaluated by witnessing a unannounced fire i brigade drill.

- [ August Spector - FPE.,JNS.wpd Page 24]

Pre-decisionalDraft (Workin progress)

GeneralEvaluation Guidance

a. Communications Evaluate the adequacy of the fire brigade' communications equipment.

. - Individual radios with lapel microphones through a repeater (best) e- Cell phones, regular phones, and message runners (minimal)  !

b. Fire Brigade Equipment Evaluate the adequacy of the fire brigade equipment.

e ' Appropriate specialized fire fighting nozzles, hose, and fittings provided.

• Are the site wide fire hazards identified and the appropriate fire fighting and specialized extinguishing agents provided in the vicinity of the subject fire

.g hazards.

• ' Smoke removal equipment provided.

• Specialized equipment, such a thermography equipment provided.

• Appropriate search and rescue equipment provided.

• Adequate SCBA and spare air cylinders.

• Personal Protective equipment (turnout coats, pants, and helmet) meet industry i

and OSHA standards

-* - Standpipe installed hose capable of reaching all areas

c. Fire Fighting (pre-fire plans) Strategies Evaluate the fire brigade fire fighting (pre-fire plans) strategies. These fire fighting strategies should as a minium address the following for each fire area containing safety-related equipment or components:

• Fire hazards e Extinguishants

. Direction of attack e Systems to be managed to reduce loss e Heat sensitive systems

.. Fire brigade specific duties

• Potential hazards toxic radiation e Smoke control managementNentilation systems e Special operationalinstructions e Instructions for general plant The fire fighting (pre-fire plans) strategies should included a smoke removal / venting plan.

)

i i

l

r l August $pector- FPR_ INS.wpd l

l Peg 3 25l Pre-decisional Draft (Workin progress)

Assess how this plan will protect the redundant shutdown path to a harsh smoke environment as result of the plans implementation and that the plan takes into consideration on how the areas immediately adjacent to the fire area, zone, or room under consideration will be maintained tenable.

d. Fire Drill Witness a unannounced fire brigade drill and evaluate the following:

Verify that the fire brigade response is satisfactorily demonstrated, including fire brigade leader command and control, teamwork, communications techniques, utilization of support from other resource groups, and proper selection of suppressant. Review the adequacy of the fire brigade's capability to locally control HVAC systems / dampers in the fire area. Review the licensee planning for

> post-fire habitability of important operating spaces (ventilation, room cooling).

. The drill should comprehensive and simulate the use of fire fighting equipment required to cope with the type of fire and challenging environmental conditions presented by the fire and the burning materials under consideration. Observe the J following:

)

Protective clothing properly utilized.

SCBA properly utilized. (Including wearing face pieces and using breathing air)

Hose lines properly deployed.

Entry into the fire room done properly.

Assess fire brigade leader's direction, thoroughness, accuracy and J effectiveness during the fire fighting effort.

Radio communications with control room adequate?

Did fire brigade check for propagation into another area?

Did fire brigade utilize the fire fighting (pre-fire plans) strategies?

Did the fire brigade perform smoke removal operations?

Did the fire brigade bring sufficient equipment to the scene to properly perform fire fighting operations?

Established back-up hose lines?

Note - The time of the alarm, the time the fire brigade is fully assembled, and the time the fire is placed under control.

l

. Verify that drills make effective use of the pre-fire plans, and that the pre-fire ,

plans accurately depict the conditions in the identified risk critical fire areas. )

i

• Determine whether communications between the control room and fire brigade are adequate to both fight the fire and conduct post-fire safe shutdown.

High Degradation Categories

l August Spec,tc4 - r Pt:_ INS.wpd Pag 3 26] l I

Pre-decisionalDraN (Workin progress)

The following are examples of high degradation categories:

(a) _ Drill was observed and the fire scenario was not comprehensive  !

. (b) Drill was observed and demonstrated weaknesses2in the following areas: l l

Excessive fire brigade response time (e.g., greater than10 minutes);

Fire brigade did not perform satisfactory as a team;  ;

Weaknesses associated with the proper use of personal protective equipment l and fire fighting equipment and its deployment; Fire brigade did not use proper fire fighting techniques or agents to fight the simulated fire; Fire brigade did not use full protective equipment inck. ding SCBA ;

Pre-fire plans and their goals were not fully implemented;  !

Communications was not satisfactory  !

I Medium Degradation Categories - l The following are examples of medium degradation categories:

(a) If a drill can not be witnessed then assume that the fire brigade skill level and their response to a challenging fire condition would result in medium or average performance. j I

(b) Fire fighting (pre-fire plans) are less than comprehensive and do not establish the minimum guidance needed to support the necessary fire fighting operations.

(c) Fire brigade equipment not st_ ate-of the-art, specialized fire fighting agents not provided for special hazards or adequately stagged, response and transport schemes for fire 3 fighting equipment not well defined, and noted weaknesses in the material condition of fire brigade equipment.

Low Degradation Categories -

The following are examples of low degradation categories: l

'(a) Drill scenario was well planned and the observed fire brigade performance was satisfactory when evaluated against the guidance above. I (b) No apparent weakness in fire brigade equipment or the stagging of this equipment, I specialized fire extinguishing agents for special hazards are maintained in the i

12 Note . Weaknesses in the fire bngade skill levels (i.e., team work, equipment deployment and utilization,  !

and techniques) are the critical elements of this effectiveness evaluation and should be considered to be critical indicators of reactive performance.

l l

' l August Spedor - wt:_ INS.wpd Pag 3 27]

Pre-decisionalDraft (Workin progress) appropriate areas of concern.

(c) Fire fighting (pre-fire plans) strategies are comprehensive and exceed minimum NRC guidance.

9.0 . Safe Shutdown Capability Degradation to post-fire safe shutdown capability are generally caused by direct fire damage to systems or components being used to achieve and maintain post-fire safe shutdown conditions, Lunsuitable environment for operator actions, safe shutdown equipment unable to provide injection at flow / pressure necessary to meet function and the unavailability of safe shutdown equipment due to fire induced circuit failures.

General Assumption it is assumed that there are no other equipment failures except those identified below. If the reliability factors for the post-fire safe shutdown / recovery systems or components being taken credit for by this methodology are not satisfactory the risk significance of the fire area may need to be adjusted in a more detailed analysis.

Extra High Degradation Redundant SSD trains are located in the area (s), zone (s), or room (s) under consideration. No recovery capability exists for performing essential functions external to the area, sone, or room under consideration.

High Degradation Redundant SSD trains are located in the area (s), zone (s), or room (s) under consideration.

Manual recovery of one fire affected SSD train is credited (e.g. altemative shutdown method for the control room) for providing the essential safe shutdown function (s).

Medfum Degradation Redundant SSD trains are located in the area (s), zone (s), or room (s) under consideration. One protected train or a recovery train available remains unaffected by the fire and immediately available.

Low Degradation Redundant SSD trains are located in the area (s), zone (s), or room (s) under consideration. One protected train and the manual recovery of the fire affected SSD train or one system with redundancy (remaining trains subject to common cause failure (CCF)) remains unaffected by the fire and immediately available.

Note that all degradations in safe shutdown oflesser severity than tow Degradation "have

r l August Spector- FPQNS.wpd Paga 28]

Pre-decisionalDraft (Work in progress) minimal risk significance. As a result, safe shutdown capability with a minimum of two independent trains, each of which can perform the essential function, is minimally risk significant.

1 i

Fire Protection Risk Significance Screening Methodology Lookup Tables l i

i l

1 l

' ' Table 4.1' Quantification of Degradation Qualitative Ratings of the Individual DID-

- Findings'8

.. . . . Automedc Fire., ntanualFire Fighdng

. Level of- ' Safe ~ 3-NourFire ta Hour Fire -- Suppression EWectiveness - l Degradadon Shutdown aarrier Twerier - EWecdveness ' (Fire Brigade)

OUtside inside CAntrol

,. . . . ~ - . . . . , _ . ., . . . ~ Control Room Room

, Extra High - 0 N/A ..N/A N/A N/A N/A i

' High . -1 0 0 0 -0.25 -0.5 Medium -2 -1 -0.5 -0.75 -0.5 -1 Low . -3 -2 (door) -1 -1.5 -1 -1.5 l

~

l

'8 Each of these values in Table 4.1 and in the criteria for determing the potential risk significance are l

exponents of 10.

l

l August Spector- FPE]NS.wpd

~

Pagn29 l Pre-decisional Draft (Work in progress) l Table 4.2, Fire Protection' Defense-in-Depth Degradation Qualitative Rating'(DQR) Ch l Redundant SSD trains are locatedin the area (s), zone (s), or room (s) under consideration. No recove l performing essential functions external to the area, sone, or room under consideration. (EX ,

3-Hour Fire barrier Protection Scheme 1-hour Fire Barrier Protection !

Degradation Quality Ratings Degradation Quality Ratin j Detection and Fire barrier Overall Detection and Detection and Fire b l manual fire DOR automatic fire manual fire suppression for Area suppression suppression HIGH HIGH EXTRA HIGH HIGH HIGH HIGH MEDIUM HIGH EXTRA HIGH HICH MEDIUM HIGH LOW HIGH HIGH HIGH LOW HIGH HIGH MEDIUM HIGH MEDIUM HIGH HIGH MEDIUM MEDIUM HIGH MEDIUM MEDIUM HIGH LOW MEDIUM MEDIUM MEDIUM LOW HIGH 4 HIGH LOW MEDIUM LOW HIGH HIGH MEDIUM LOW MEDIUM LOW MEDIUM HIGH LOW LOW LOW LOW LOW HIGH HIGH HIGH MEDIU HIGH MEDIUM MEDIU HIGH LOW MEDIU !

MEDIUM HIGH MEDIU '

MEDIUM MEDIUM MEDIU MEDIUM LOW MEDIU LOW HIGH MEDIU ,

LOW MEDIUM MEDIU i LOW LOW MEDIU !

HIGH HIGH LOW HIGH MEDIUM LOW l HIGH LOW LOW MEDIUM HIGH LOW MEDIUM MEDIUM LOW MEDIUM LOW LOW LOW HIGH LOW LOW MEDIUM LOW LOW LOW LOW l

Table 4.2, Fire Protection Defense-in-Depth Degradation Qualitative Rating (DQR

Redundant SSD trains are located in the area (s), zone (s), or room (s) under consideration. Manual recov

. SSD train is credited (e.g. alternative shutdown method for the control room) for providing the essential Lf unction (s). (HIGH) .

3-Hour Fire Barrier Protection Scheme 1-hour Fire Barrier Protection Degradation Quality Ratings Degradation Quality Ratin Detection and Fire barrier Overall Detection and Detection and Fire manual fire DOR ~

automatic fire manual fire suppression for Area suppression suppression HIGH HIGH HIGH HIGH HIGH HIGH MEDIUM HIGH HIGH HIGH MEDIUM HIGH

[ August Spector- FPE_ INS;wpd Pags 30]

Pre-decisionalDraft (Workin progress)

LOW HIGH MEDIUM HIGH LOW HIGH HIGH MEDIUM MEDIUM MEDIUM HIGH HIGH MEDIUM MEDIUM MEDIUM MEDIUM MEDIUM HIGH LOW MEDIUM LOW MEDIUM LOW HIGH HIGH LOW LOW LOW HIGH HIGH MEDIUM LOW LOW LOW MEDIUM HIGH LOW LOW MINIMAL LOW LOW HIGH HIGH HIGH MEDIU HIGH MEDIUM MEDIU HIGH LOW MEDIU MEDIUM HIGH MEDIU MEDIUM MEDIUM MEDIU MEDIUM LOW MEDIU LOW HIGH MEDIU LOW MEDIUM MEDIU LOW LOW MEDIU HIGH HIGH LOW HIGH MEDIUM LOW HIGH LOW LOW MEDIUM HIGH LOW MEDIUM MEDIUM LOW MEDIUM LOW LOW LOW HIGH LOW LOW MEDIUM LOW LOW LOW LOW Table 4.2, Fire'Protebtion Defense-in-Depth Degradation Qualitative Rating (DQP) Ch Redundant SSD trains are located in the area (s), zone (s), or room (s) under consideratior:. One protec

_ train available remains unaffected by the fire and immediately available. (MEDIU 3-Hour Fire Barrier Protection Scheme 1-hour Fire Barrier Protection Degradation Quality Ratings Degradation Quality Ratin Detection and Fire barrier Overall Detection and Detection and Fire manual fire DQR automatic fire manual fire suppression for Area suppression suppression HiGH HIGH MEDIUM HIGH HIGH HIGH MEDIUM HIGH MEDIUM HIGH MEDIUM HIGH LOW HIGH LOW HIGH LOW HIGH HIGH MEDIUM LOW MEDIUM HIGH HIGH MEDIUM MEDIUM LOW MEDIUM MEDIUM HIGH LOW MEDIUM MINIMAL MEDIUM LOW HIGH HIGH LOW MINIMAL LOW HIGH HIGH MEDIUM LOW MINIMAL LOW MEDIUM HIGH LOW LOW MINIMAL LOW LOW HIGH t * '

~!

HIGH HIGH MEDIU t

HIGH MEDIUM MEDIU fi HIGH LOW MEDIU

[ August Spector - FPE_JNS.wpd Pagt31]

Pre-decisionalDraft (Workin progress)

MEDIUM HIGH MEDIU MEDIUM MEDIUM MEDIU MEDIUM LOW MEDIU LOW HIGH MEDIU LOW MEDIUM MEDIU LOW LOW MEDIU !

HIGH HIGH LOW

- HIGH MEDIUM LOW HIGH LOW LOW HIGH LOW

' h'J" F3 MEDIUM MEDIUM MEDIUM MEDIUM LOW LOW LOW

' LOW HIGH LOW LOW MEDIUM LOW

, LOW LOW LOW ilable 4.2, Fire' Protection Defense-in-Depth Degradation Qualitative Rating (DQR) Ch Redundant SSD trains are located in the area (s), zone (s), or room (s) under consideration. One protect recovery of the fire affected SSD train or one system with redundancy (remaining trains subject to c LL- . . .

. . _ . . . ; (CCF)) remains unaffected by the fire and immediately avaliable. (LOW) .

3-Hour Fire Barrier Protection Scheme 1-hour Fire Barrier Protection Degradation Quality Ratings Degradation Quality Ratin Detection and Fire barrier Overall Detection and Detection and Fire manual fire DQR automatic fire manual fire suppression for Area suppression suppression HIGH HIGH LOW HIGH HIGH HIGH MEDIUM HIGH LOW HIGH MEDIUM HIGH LOW HIGH MINIMAL HIGH LOW HIGH HIGH MEDIUM MINIMAL MEDIUM HIGH HIGH MEDIUM MEDIUM MINIMAL MEDIUM MEDIUM HIGH LOW MEDIUM MINIMAL MEDIUM LOW HIGH HIGH LOW MINIMAL LOW HIGH HIGH MEDIUM LOW MINIMAL LOW MEDIUM HIGH LOW LOW MINIMAL LOW LOW HIGH HIGH HIGH MEDIU

! HIGH MEDIUM MEDIU HIGH LOW MEDIU

,, .. MEDIUM HIGH MEDIU

(&{g - MEDIUM MEDIUM MEDIU

-y Mj J '

• MEDIUM LOW MEDIU

[ \U -- . LOW HIGH MEDIU

LOW MEDIUM MEDIU R .

t , -4, LOW LOW MEDIU

[ HIGH HIGH LOW h

HIGH MEDIUM LOW

{'- HIGH LOW LOW MEDIUM HIGH LOW

,, MEDIUM MEDIUM LOW l

l August Spector- FPE_ INS.wpd Page 32]

Pre-decisionalDraN (Work in progress)

F " ' _. '

~

MEDIUM LOW LOW LOW HIGH LOW LOW MEDIUM LOW

..m. , < n LOW LOW LOW

[ August Spector - FPE_ INS.wpd Pagf31l !

Pre-decisional Draft (Work in progress) l i

I

' Table 4.3 Overall Fire Protection DID Degradation DQR Characterization for the Fire

.. Area, Zone, or Room being Evaluated .  ;

EXTRA HIGH The score is greater than -1. I High - The score is greater than -2, and less than or equal to -1

. Medium The score is greater than -3, and less than or equal to -2  !

Low . The score is greater than -4, and less than or equal to -3 ,

Minimal ' i The score is less than or equal to -4 '

l Table 4.4a - Generic ignition Frequencies '

. . , ' Plant Buildings or Rooms - .

Building orRoom ignition Frequency (IF)

Control Room 7E-3 Cable Spreading Room SE-3 Diesel Generator Building 6E-2 Switchgear Room 1 E-2 Battery Room 3E-3 to 1E-2 Reactor Building 3E-2 Auxiliary Building 6E-2 Turbine Building 6E-2 Containment 9E-3 I

[ August Spector- FP~EJNS.wpd "aga 34]

Pre-decisionalDraft (Work in progress)

? bis 4.4. Combination of Flie ignition (initiation) Frequency and Overall Fire Protection Defense'-in-Depth (DID i gradatio. Qualitative Rating (DQR) Characterization for a Fire Area (s), Zone (s) ^or Room (s) under Consideratio Overall Fire Protection DID Degradation DQR Characterization for a Fire Area, Zone or Room un gnition Consideration irtion) Adjusted for Time Degraded Condition Existed ency (IF)

EXTRA HIGH ' HIGH , MEDIUM LOW' MINI i

~

~

> 30~ 30-3 <3 days > 30 30-3 <3 days ' > 30 30-3 <3 ' > 30 30-3 <3 days days days days days days days days days days l

> 104 yYELLO Y 7 yp TYELLO73 WHITE T 17 YELLO 7 WHITE' ' WHITE ' " '

i 31WLi 4'./-J 44 1.3W E :.] E;s.W;. d i  !

IF > 10 4 i- GREEN '

l aiW G-N 4 ' e lTE . b W d d-YELLOkJ WHITE ' GREEN l WHITE -

s  !

IF > 10 4  :

ed W D;A.

m ire 2- mite- -

4 etTE 3

• ' GREEN 1

IF > 10

< 4

_10 - m'

- m

GREEN-m, a .s . u.+a 4 au- m~ <

m._ aa.< , , a

., 4 ~ . s . m , <

~ w ,e a 1

l Figure 9.1 Degradation of Post-fire Safe Shutdown Capability EXTRA HIGH DEGRADATION Redundant SSO trains are located in the etwa(s), zone (s), or room (s) under ,

consideration. No recovery capability exists forperforming essential functions  !

external to the area, sone, or room under consideration.

FIRE AREA BOUNDARY Fire area, zone, or room under consideration

.. . TRAIN Ai l

' TRAIN B .

No exists recovery capability exists for performing essential functions extemal to the area, zone or room under consideration ,

i Figure 9.1a

r

[ August Spector- FPllNS.wpd Pig 3 35] l Pre-decisional Draft (Work in progress) l HIGH DEGRADATION Redundant SSO trains are located in the area (s), zone (s), or room (s) under consideration. Manual recovery of one fire affected SSD train is credited (e.g.

alternative shutdown method for the control room) forproviding the essential safe shutdown function (s).

FIRE AREA BOUNDARY TRAIN A TRAIN B Fire Area of Concem RECOVERY OF TRAIN 3 Manual actions outside the fire area of concern Figure 9.1b MEDIUM DEGRADATION Redundant SSD trains are located in the area (s), zone (s), or room (s) under consideration. One protected train ora recovery train available remains unaffected by the fire and immediately available.

FIRE AREA BOUNDARY

' TRAIN A [(Protected from tire) _ _

TRAIN B ^

Fire Area of Concem OR RECOVERY OF TRAIN unaffected by the fire and

, immediately available

[ August Spector- F&lNS.wpd Pag 3 36]

Pow-decisionalDraft (Work in progress)

Figure 9.1c LOW DEGRADATION Redundant SSD trains are located in the ares (s), zone (s), or room (s) under consideration. One protected train and the manual recovery of the fire affected SSD F train or one system with redundancy (remaining trains subject to common cause

failure (CCF)) remains unaffected by the fire and immediately available.

FIRE AREA BOUNDARY l TRAIN AJ(Protectedfrom Mre): ,.

U~ ,, l TRAIN B J ,

l Fire Area of Concem AND E'WUAL RECOVERY OF OR one system with redundancy

. TRAIN 8 remains unaffected by the fire and immediately available l

l l

f

l~ August Spector - FPE_ INS.wpd ~~ Pagy 3Y]

Pre-decisional Draft (Work in progress) l

l August Spector- FPEJNS.wpd Page 38]

1 Pre-decisional Draft (Work in progress)

' Figure 4.1- Fire Protection Risk Significance Screening Methodology (FPRSSM)

Determination Process of Potential Risk Significance of Fire Protection inspection Findings

+

l

+tA t

I h

4-A k I l

l i

[ August Spector- FPlCINS.wpd P:g339]

Pre-decisional Draft (Work in progress)

Attachment 1 Appilcation Fire Protection Risk Significant Screening Methodology to Case Studies Case 1: Cable Spreading Room and its interface with the "A" and "B" Switchgear Rooms Example 1 A.

The cable spreading room (fire area 57) is directly under the MCR and is located in the Reactor Auxiliary Building on elevation 43'-0". The east wall of the cable spreading room (CSR)is shared with the "A" switchgear room (fire area 60) and the "B" switchgear room (fire area 56) shares the south wall. These interfacing walls between the CSR and the switchgear rooms are l 3-hour fire rated except for the Thermo-Lag wall segment (which includes a door) in the south wall at column lines RAJ/RA3 and RAl/RA3. This wall segment is rated for 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> and 48  ;

minutes. In addition, there is a door.in the east wall which interfaces with fire area 60 and a door i from the CSR into the static inverter enclosure and a HVAC air return from the inverter room to the "B" switchgear room. This air return is not protected with a fire or smoke damper. Located in the B" switchgear room are the hot shutdown panels which are used for post-fire safe shutdown and are required to be used to shutdown the plant in the event of a significant fire in  !

the cable spreading room or the MCR.

i The CSR is protected by an automatic Halon suppression system which is provided with cross zoned thermal fire detectors that activate the Halon system and provide fire alarm indication in the MCR.

A. General Assumptions The fire load in the cable spreading room is high. Most of the redundant control circuits I associated with the plant's safety functions are located within the cable spreading room. For the l purpose of this assessment, it is assumed that a fire has occurred in the CSR. The fire is l

assumed to have started by a fault in the pressurizer heater bus and the fault has caused a fire  !

which has propagated to the cable in the room's overhead, The fuel sources in the CSR have i sufficient energy to result in a fire that is capable of developing a hot gas layer.

l The licensee did not establish the fire or environmental threshold conditions (e.g., temperatures l at which cable insulation would degrade) that could lead to functional failures (faults or spurious l signals) of safe shutdown electrical cables. Therefore, for this assessment it is assumed that if l the critical post fire safe shutdown functions within the cable spreading room will be susceptible j to fire damage and loss of function. Therefore, manual operator recovery actions may have be l

taken outside the main control room in order to regain plant control and achieve and maintain j safe shutdown conditioas. j l

i I

' l August Spector - FPiUNS.~wpd Pago 40]

Pre-decisional DraN ' (Work in progress)

I Due to the uncertainties associated with the installation of passive fire protection features (e.g. )

' installation of a fire door in a Thermo-Lag wall when it was fire tested in a concrete wall; untested fire barrier wall joints and changes in wall direction), and the fact that most fire rated assemblies are qualified on the basis of one successful qualification fire test, it is assumed that if the combustible fuel load, fuel geometry, and room configuration could contribute to a fire condition that could exceed the fire rating of the fire barrier in question, the fire barrier is assumed to fail.

It is assumed that fire barriers and passive fire resistive devices are not qualified to preclude smoke from propagation out of the area of concern and into adjacent plant areas.  !

The fire initiation frequency (IF) for the cable spreading room is 7X10E-3/yr.

Due to a fire in the CSR which results in the loss of a majority of the control room functions, the plant would enter into shutdown procedures and implement its alternative safe-shutdown methodology (shutdown from a remote shutdown panel and/or other locations outside of the MCR).

B. Post-Fire Safe Shutdown - Potential Impact and Capability H For a fire in the CSR alternative shutdown capability is provided. This capability is electrically  ;

and physically independent of the cable spreading room. The hot shutdown control panelis  :

located in a room in the "B" Switchgear Room on Reactor Auxiliary Building (RAB) elevation 43'-0". Manual operator recovery actions are necessary to regain control of and operate systems, equipment, and components needed to achieve and maintain post-fire safe shutdown conditions. Due to a number of air leakage paths between the cable spreading room and the i switchgear rooms, a challenging fire in the CSR will most likely result in smoke propagation into the adjacent rooms.

An inspection team concern was room and electrical equipment cooling and control of smoke at remote and local control stations during a control room fire. Specifically, there are several

- findings or weaknesses that could contribute to habitability conditions of the Hot Shutdown Control Panel (HSCP) room adjacent to the CSR. These findings and weaknesses are related to design issues associated with: the CSR automatic Halon suppression system; the lack of a smoke management plan associated with both the fire brigade pre-fire plans and the design and implementation of the alternative shutdown capability. Smoke in the HSCP room and in the "A" and "B" switchgear rooms would inhibit plant operator access to these areas and could inhibit the successful operation of equipment. For a fire in this area, the normal HVAC in the HSCP is lost when the control cables in CSR are damaged. In addition, the normal HVAC to the  ;

switchgear rooms could be lost as a result of a fire-induced loss of off-site power (LOOP) trip signal caused by the fire in the CSR.

C. Conditions Affecting Fire Mitigation Effectiveness During the inspection, conditions were identified with the design of the CSR Halon fire

' l August Spector - FPE.,JNS.wpd P:g341]

l Pre-decisionalDraN (Workinprogress) suppression system that could affect its ability to suppress a fire. Specifically, the system is designed to provide a concentration of 5 to 7 percent for a minimum of 10 minutes. Review of initial testing records for this system indicates that the system delivered and held a room Halon

' concentration more than 6 percent for 4 to 5 minutes and more than 5 percent for 11 to 12 minutes. These concentrations do not meet the criteria of NFPA 12A (1980). Specifically, cable fires can be deep seated. This condition may result from flaming combustion at the surface or from the ignition wkhin the mass of fuel. Smoldering combustion then progresses slowly through the mass. The burning rate of these fires can be reduced by the presence of Halon and they may be extinguished if a high concentration can be maintained for an adequate soaking time.

' However, it is not normally practical to maintain a sufficient concentration of Halon for a suff;cient time to extinguish deep-sested fires. The NFPA code recommends that the use of

- Halon 1301 systems be limited to solid combustibles, which do not become deep-seated.

NUREG/CR-3656, " Evaluation of Suppression Methods for Electrical Cable Fires." Table 6 of the report states that for exposure fires, the minimum soak time required for a 6 percent concentration of Halon 1301 was 10 minutes for IEEE-383 qualified cables and 16 minutes for unqualified cables in the horizontal position. Table 10 of the report states that for fully developed fires, the minimum soak time required for a 6 percent concentration of Halon 1301 was '15 minutes for IEEE-383 Qualified cables and 10 minutes for unqualified cables in the horizontal position.

Other issues with the CSR Halon suppression system were related to the actuation time delays associated with the cross zone thermal fire detection system and its interlock logic interface with the CSR HVAC fans, fire dampers, and fire doors. For example, if a fire door is not completely closed the Halon system will not actuate, or if a fire damper does not close completely the Halon {

system will not actuate. l I

The inspection report did not identify any conditions that would indicate that there were any l technical issues associated with the placement and layout of the thermal detectors in the CSR. l

- As part of the inspection, fire brigade effectiveness was evaluated by reviewing the fire brigade fire fighting strategies (pre-fire plans), by inspection of the fire brigade fire fighting equipment 1 and personal protective equipment, and by witnessing an unannounced fire brigade drill. The  !

inspection noted fire brigade effectiveness weaknesses associated with pre-fire plans (smoke removal / control) and with the equipment used to protect fire brigade members from the hazards ,

associated with fire-fighting.' Based on the observi.tions made during the witnessing of a drill, l

~ the inspection identified several fire brigade performance weaknesses (e.g., basic fire-fighting .

l technique / equipment deployment operations not fully implemented, lack of reliance on breathing j apparatus during drills). J

~ The Thermo-Lag wall segment (which includes a door) in the south CSR wall at column lines RAJ/RA3 and RAl/RA3 is rated for 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> and 48 minutes. The fire loading conditions associated with this room are high. A fire involving the in-situ fuel in the CSR has the potential energy that exceeds the Thermo-Lag wall assembly qualification fire test conditions.

D. Assessment i

L

p l August Spectri- FPE, INS.wpd - Paga 42]

Pre-decisional DraN (Work in progress)

Based on the conditions noted during the inspection, a qualitative screening assessment of the

- potential risk significance of the fire protection DID principles weaknesses was performed.

Assuming that a fire occurs in the CSR , the fire should be detected eventually by the fire detectors in the room. Since, the Halon suppression system is actuated by cross zoned thermal detectors (2 detectors on two separate detector loops have to sense the fire) it is inherent that a

{

time delay in actuation will occur. The time delay associated with having to actuate a cross zoned thermal detection system may provide the conditions for a deep seated cable insulation fire to become more established. In addition, if a fire door or damper in the system does not fully close or an HVAC fan does not stop, the system will not actuate. Considering the potential l system actuation delays and the inability of the system to deliver the required Halon concentrations needed to control and extinguish a deep seated cable insulation fire, the -

degradations in ADIAS effectiveness is considered to be high.

The fire loading conditions associated with the CSR are high and they have potential energy to exceed the Thermo-Lag wall qualification fire test conditions. The Thermo-Lag wall segment (which includes a docr) in the south CSR wall at column lines RAJ/RA3 and RAl/RA3 is rated for i

i hour and 48 minutes in lieu of the minimum 3-hour requirement. Therefore, if the automatic suppression system and manual fire-fighting fails to suppress, control, and extinguish the fire, the Thermo-Lag wall becomes a critical DID element. The degradations associated with the ability this wall to confine the fire to the CSR (FB) and keep it from propagating to the "B" l l switchgear room is considered to be medium. This is based on the likelihood that the fire could be controlled by manual suppression and confined to the room of origin.

i l It should be noted that the fire detection system provides notification of a potential fire condition and performs no active function to mitigate the consequences of the fire. Detection system i design and installation is a critical factor and can delay operator and fire brigade response to the i 1 potential condition.' Therefore, if degradations associated with detection effectiveness are i medium to high, they can negatively impact timely mitigation of a fire event. During the inspection, weaknesses associated with the fire brigade's ability to effectively implement fire fighting operations in a timely and efficient manner were identified. This, coupled with the medium degradation in detection effectiveness indicates an overall medium degradation in 1 D/MS. J l' Therefore, an assumed challenging fire in CSR can cause fire damage to redundant trains of

! shutdown functions. Thus, the post-fire safe shutdown methodology requires the manual i recovery of plant safe shutdown equipment and functions that have been affected by fire damaged to the cables in the CSR. This is accomplished by abandoning the main control room, isolating fire affected circuits and re-aligning power to needed equipment from alternative power sources, manually re-aligning valves in reactor water make-up systems, monitoring reactor shutdown and core cooling performance parameters and controlling them from HSCP.

Since the HSCP is located in the "B" switchgear room which is adjacent to the CSR, the SSD i capability for the case when the barrier fails is extra high. In addition, the independent ventilation system provided for the HSCP room which is located in the "B" switchgear room l

I

![ August spector- FPE_ INS.wpd pig 343}

Pre-decisionalDraR (Workinprogress)

'could be rendered inoperable as result of a CSR fire Smoke from the CSR fire, and subsequent fire fighting actions could affect operator habitability inside HSCP room and the switchgear rooms. The room and electrical equipment cooling for the *B switchgear room, and indirectly for the HSCP room, could be achieved through manual operation of HVAC equipment whose control circuits and components were protected from fire. However, operating the exhaust fan could contribute to the smoke migration from a CSR fire into the switchgear rooms and the

' HSCP room. Should the barrier between the "B" switchgear room and the CSR survive, controlling the plant from outside the control room is normally considered to be a high SSD degradation. However, the smoke complications could have had an impact on the operators making a timely recovery of the plant. Therefore, a high degradation for the case when the barrier succeeds is expected to be over-optimistic, but for this example a high degradation for SSD will be assumed for the case where the barrier succeeds.

The potential change in CDF 'of these identified weaknesses in the fire protection DID principles can be expressed as follows:

PRS = IF + FB + SSD + AD/AS +D/MS

' Therefore, if a challenging fire were assumed to occur in the CSR, the following is an approximation of the potential risk significance (PRS) of the fire protection DID findings:

IF= -2 FB = -1 SSD = -0

'AD/AS = 0 D/MS = -0.5 PRS = -3.5 (red)

This term is the double room term (term when fire barrier fails). The single room term (case where barrier doesn1 fail) is not normally used for medium barrier degradations since it is less than or comparable to the double room term for medium barrier degradations.

l Example 18.

Take the case where the fire barrier between the "B" switchgear room and the cable spreading room is resMred to a three hour barrier. l IF= -2 FB=- 2 .

l r

' [ August $pector - FPE_.lNSx/pd Pagia 44] <

Pre-decisionalDraft (Work in progress)

SSD=0 AD/AS = 0 D/MS = -0.5 PRS (drt) = -4.5 As a result, PRS due to the double room term (if barrier fails) is now -4.5. However, the FPRSSM indicates that the single room term (if barrier suceeds) must also be used to evaluate .

cases with no fire barrier degradation. Therefore the single room term will use SSD= -1 if full I

credit is given to the SSD and not deducted for smoke Note that the single room term utilizes FB=high=0 since no barrier exists in the "B" switchgear room to mitigate the fire damage. The single room term is as follows.

IF= -2 FB= 0 SSD = -1 l AD/AS = 0 D/MS = -0.5 l PRS (srt) = -3.5 The total of the drt and srt is -3.5 (red) for example 1B.

If the PRS for examples 1 A and 18 are compared the argument can be made that no decrease in change in CDF was realized as a result of fixing the fire barrier. However, remember that example 1A was an approximation due to the fact that only the drt was used. If the single room term is calculated for example 1 A, you would get IF = -2 FB = 0 SSD = -1 '

AD/AS = 0 D/MS = -0.5 PRS (srt for example 1 A) = -3.5 Therefore, example 1 A PRS really is 1E-3.5 + 1E-3.5 which is the sum of srt and drt for 1 A.

Therefore the change is CDF for example 1 A is twice as great as example 1B. The only reason the art for example 1 A was calculated is due to the apparent conflict of the change in CDF for examples 1 A and 1B, prior to calculating the srt for example 1 A. Remember that for medium degradation of fire barrier, the double room term is adequate for this screening approach since the single room term is no larger than the double room term.

Example 1C Fix automatic suppression: AD/AS = Low degradation

[ August Spector - FPE._ INS.wpd . Piga 45]

Pre-decisionalDraft (Workin progress) l:

IF= -2 SSD = 0 FB = -2

AD/AS = -1.5 D/MS = -0.5 l l

l. PRS (drt) = -6 )

~

For the single room term,

. SSD= -1 FB = 0

)

Therefore total change in CDF is -5 (White)

CASE 2: AUXILIARY FEEDWATER ROOMS Example 2A l The auxiliary feedwater (AFW) pump rooms (fire areas 31 and 32) are located on auxiliary l

building elevation 695'-0", center section. The Unit 1 turbine driven AFW and the Unit 2 motor-driven AFW pumps are located in Fire Area 32 and Unit 2 turbine driven AFW and the Unit 1 l motor driven AFW pumps are located in Fire Area 31. The boundaries of the two fire areas are l fire rated for 3-hours with all fire barrier openings protected with 3-hour rated assemblies (e.g.,

l doors campers). These fire areas share a common fire barrier wall. The common wall has a

' door opening that is protected by a tin-clad sliding fire door that is held open and is closed by a fusible link counterweight door closer. Both pump rooms are protected by automatic sprinklers and are provided with spot type ionization detection. The pump rooms contain motor control centers, cables, lube oil, air compressors, and various electrical panels. Therefore, these rooms have enough potential ignition and fuel sources available to produce a hot gas layer.

]

A. General Assumptions j l

The cables and equipment for all AFW pumps are located in Fire Area 32. For the purpose of this assessment, it is assumed that a fire has occurred in Fire Area 32 as a result of a compressor fire that ignites the cables. There is sufficient energy to result in the development of l hot gas layer.  ;

The licensee did not establish the fire or environmental threshold conditions (e.g., temperatures at which cable insulation would degrade) that could lead to functional failures (faults or spurious signals) of safe shutdown electrical cables. Therefore, for this assessment it is assumed that, if I l the critical post fire safe shutdown functions are exposed to a fire environment they will be j l susceptible to fire damage and loss of function.

l

..The fire initiation frequency (IF) for the AFW pump rooms is SE-3/yr.

For the loss of all AFW, the EOPs direct the operators to use bleed and feed to cool the reactor t .

I -

- l August Spector - FPEJNS.wpd - P:gs 46]

Pre-decisionalDraN (Workin progress) and remove decay heat. For the purpose of this assessment, it is assumed that the cabling associated with the bleed and feed operation remains free of fire damage and that the charging pumps are unaffected by the fire or by fire fighting operations.

B. Post-Fire Safe Shutdown - Potential Impact and Capability The inspection identified in Fire Area 32 that AFW flow could be lost due to control cable fire-induced damage to AFW Pump 12 motor operated discharge valves (MV32381 and MV32382). The SSA assumed the loss of all Train A AFW equipment for a fire in this area. As a result, Train B AFW components (including valves MV32381 and MV32382) were credited for

, accomplishing the hot shutdown decay heat removal function for Unit 1. Both valves were normally open and were used to control AFW pump discharge flow to either steam generator SG11 (MV32381) or SG12 (MV32382). According to the SSA, to accomplish the decay heat removal function, one of these valves must be free of fire damage in fire area 32. Given this configuration, a fire in fire area 32 could damage both valve control circuits in the unprotected junction box.

In addition, it was identified that No.12 AFW pump suction valve MV32335 control circuits would be vulnerable to fire damage in Fire Area 32. Damage to these circuits could cause the normally open pump suction valve to close, tripping the credited AFW pump on low suction pressure.

C. Conditions Affecting Fire Mitigation Effectiveness

- Fire door 62 is installed in the fire barrier wall separating the Unit 1 AFW pump room (Fire Area

32) from the Unit 2 AFW pump room (Fire Area 31). The door is a UL listed sliding tin clad Class A (3-hour fire resistive rating) fire door. The door was maintained in the open position and closed automatically upon thermal activation and release of a fusible link located the leading edge of the coor. This configuration did not conform to the code. Specifically, the code specifies multiple automatic releasing devices (e.g., fusible links) and that these devices be located near the ceiling. The automatic closir.g of this door would have been significantly delayed.

For a fire in Fire Area 32, Unit 1 AFW Pump Room, the closest readily accessible hose station was located along column line G in the Unit 2 Turbine Building. The hose station located on the Unit 1 Turbme Building side was located between the main feedwater pumps and the hose was not of sufficient length to reach Fire Area 32, and would have required additional hose to be found and added if the fire brigade chose to use it. Thus, if the fire brigade used the Unit 2 Turbine Building fire hose station to fight a fire in Fire Area 32, it would have to enter Fire Area

~ 31 and breach the fire barrier wall by opening the sliding fire door separating the two pump rooms. The resulting heat and smoke could expose redundant post-fire safe shutdown equipment to a common fire and smoke environment, and impede post-fire safe shutdown operator actions in Fire Area 31. This was considered a manual fire fighting weakness.

D. Assessment -

A qualitative screening assessment of the potential risk significance of the fire protection DID

l August Spector - FPE_ INS.wpd p:gy 47]

Pre-decisional Draft (Work in progress) principles weaknesses was performed.

If the fire in Fire Area 32 caused fire damage such that the preferred and SSD analysis designated AFW system could not be recovered, the operators would follow the EOPs for a loss

. of feedwater event and implement a bleed and feed operation. Since a recovery methodology (using the pressurizer PORVs and the charging pumps in the bleed and feed mode) is immediately available and unaffected by the fire, the degradation in SSD effectiveness is considered to be medium.

The installation of the fire door / closer between the two AFW pump rooms does not meet the code.' Therefore its ability to close in a timely manner to fully protect the opening is not credited by this assessment. In addition, the 1-hour raceway fire barriers protection for the "B" train AFW cables in Fire Area 32 were removed and could not be credited. Therefore, the FB degradations for Fire Areas 31/ 32 are considered to be high.

The inspection did not evaluate the effectiveness of the sprinklers or the detection system provided for Fire Areas 31/ 32. Therefore, it is assumed that the these system will function as designed. However, it is assumed that degradations associated with the placement and location of detectors and obstructions to sprinklers may exist. Therefore, for this assessment, it was assumed that medium degradations affecting AD/AS effectiveness exist.

During the inspection an assessment of the fire brigade and its effectiveness were not performed. Therefore, it is assumed that the fire brigade performance would be average and exhibit medium degradations in D/MS.

The potential risk significance of these identified weaknesses in the fire protection DID principles can be expressed as follows:

PRS = IF + FB + SSD + AD/AS +D/MS Therefore, if a challenging fire were assumed to occur in the Fire Area 32, the following is an approximation of the potential risk significance of the fire protection DID findings:

IF = -2 FB = 0 i SSD = -2 AD/AS = -0.75 D/MS = -0.5 PRS = -5.25 (White)

Example 28 For example 28, we will assume that the fire door between fire areas 31/32 has a low

p- [ August Spector - FPE_ INS.wpd .

t Page 48]

Pre-decisional Draft (Work in progress) degradation. As a result, we will focus on the one hour barrier from now on. Fix the fire barrier in example 2A.

IF = -2 FB = -1 SSD = -2 due to feed and bleed AD/AS = -0.75 D/MS = -0.5 PRS (DRT) = -6.25 But the SRT must also be calculated since the barrier has a low degradation.

IF = -2 FB = .0 SSD = -4 due to feed and bleed and one train of AFW (which is assumed adequate)

AD/AS = -0.75 D/MS = -0.5 PRS (SRT) = -7.25 Therefore sum of SRT and DRT is -6.25 (Green)

Note that fixing automatic suppression instead of the fire barrier may not produce a Green since the cabling without a fire barrier may not be fully protected by successful automatic suppression.

l l

l t

State Reoresentatives/ Public Interest Grouos/Reaional Administrators Thomas Ortciger, Director Illinois Department of Nuclear Safety 1035 Outer Park Drive Springfield, Illinois 62704 Jill Lipoti, Ph.D., Assistant Director Radiation Protection Programs j Division of Environmental Safety, Health and j Analytical Programs i Department of Environmental Protection P.O. Box 415 Trenton, New Jersey 08625-0415 Mr. David Lochbaum Union of Concerned Scientists 1616P Street NW Suite 310

Washington, D.C. 20036-1495 Regional Administrators (Code 9U) i i

i l

l l

I l \

I' b _

j