ML20096F314

Evaluation of Thermo-Lag Fire Barriers. W/Three Oversize Drawings

ML20096F314
Person / Time
Site:	Oyster Creek
Issue date:	12/13/1995
From:	Barbieri F GENERAL PUBLIC UTILITIES CORP.
To:
Shared Package
ML20096F319	List: ML20096F314 ML20265E729 ML20265E734 ML20265E744
References
102, NUDOCS 9601230325
Download: ML20096F314 (257)

Text

{{#Wiki_filter:, P M Nuclear-Oyster Creek Evaluation of Thermo-Lag Fire Barriers Report 102 REV. O 1 Project Not/TDR No.: y 3285V8 Authors: F. P. Barbieri / Y)if y ^ Date Approvals: CDW- -= n./,1, s-- SECTION MANJp3ER DATE 0," : A' /3 */J - 95 DEPARTpNT $ REC'I6R / DATE fh Iz d Wr f DIREDTOR,~TtC8NICAL FUNCTIONS DAT8 9601230325 951229 PDR ADOCK 05000219 p PDR

e Trpical Riport 102 Rev. O Page 2 of 22 ABSTRACT The purpose of this report is to provide the methodology for establishing the fire endurance rating of installed Thermo-Lag fire barrier raceway systems. This report summarizes the results of the evaluations which establish the aforementioned fire endurance ratings and identifies those Thermo-Lag fire barrier raceway systems which meet the requirements of Appendix R, Section IIIC, those barriers which do'not meet Appendix R and will be modified or upgraded to neemt Appendix R, and those barriers which do not meet Appendix R and for which an evaluation will be performed to justify the fire endurance rating in an exemption request. This report also provides the methodology used to evaluate the hazards in each fire area or fire zone where Thermo-Lag fire barrier raceway systems are installed. These hazard evaluations will be documented in exemption requests and will serve as the basis for supporting such exemptions where the fire endurance rating of the Thermo-Lag fire barrier raceway system does not meet the r'equirements of Appendix R, Section IIIG.

. _~ Tcpical RIport 102 s Rev. O Page 3 of 22 TABLE OF CONTENTS Eigt 1.0 FURFOSE 4 2.0 NETHODOLOGY 4 2.1 Establishing Actual Fire Rating 4 2.2 Establishing Cable Qualification Rating 7 2.3 Evaluating Fire Essards 10 3.0 SUNNARY OF RESULTS 14 3.1 Dffico "A" 480 V Switchgear Room (Fire Bone OS-FE-6A) 14 3.2 office "B" 480 V Switchgear Room (Fire Bone OB-FE-65) 15 3.3 Reactor Building 51' Elevation (Fire Bone RB-FS-1D) 16 3.4 Reactor Building 23' Elevation (Fire Bone RB-FS-1E) 18 3.5 Reactor Building -19' Elevation (Fire Bone RB-FE-1F2) 19 3.6 .Turbins Building Basement, South End (FIRE SONE TB-FS-11D) 20 3.7 Turbine Building, Switchgsar Room, West End, Messanine (Fire Ione TB-FS-11C) 21

4.0 REFERENCES

22 1 i 013/026

l Topical R: port 102 Rev. O Page 4 of 22 1.0 FURPOSE The purpose of this report is to provide the methodology for establishing the fire endurance rating (equivalent rating by test comparison) of installed Thermo-lag fire barrier raceway systems at oyster Creek. Fire endurance rating is established by identifying the " Actual Fire Rating" or the rating consistent with the fire endurance test acceptance criteria as defined in NRC Generic Letter 86-10 Supplement 1, " Fire Endurance Test Acceptance Criteria for Fire Barrier Systems Used to Separate Redundant Safa shutdown Trains Within the Same Fire Area". Fire endurance rating is also established by identifying the " Cable Qualification Rating" or the rating which.'s based upon establishing the maximum temperature inside a fire barrier envelop that is considered acceptable to demonstrate cable functionality. Use of the " Cable Qualification Rating" is a deviation from the GL 86-10 acceptance criteria but is acceptable based upon an engineering evaluation as described herein. Results of these evaluations are reported in Section 3. This report also provides the methodology used to evaluate the hazards in each fire area or fire zone. The hazard evaluation will serve as the basis for dupporting exemptions from Appendix R, Section IIIG. 2.O METEODOLOGY 2.1 Establishing Actual Fire Rating To assess material performance and provide a basis for evaluation of installed Thermo-Lag fire barriers, an industry fire endurance test program was conducted by the Nuclear Energy Institute (NEI). To address issues with the fire endurance capability of installed barrier configurations, the industry test program: Assessed current industry configurations through the use of survey data, Conducted tests to establish performance of various baseline and upgraded fire barrier system assemblies, and Developed a guideline to assist utilities in evaluating installed barrier configurations. The guidelir.e developed by NEI is known as the "NEI Application Guide for Evaluation of Thermo-Lag 330 Fire Barrier Systems" (Report no. 0784-00001-TR-02 Revision 1) or the " Application Guide". The Application Guide provides a process and data for evaluation of installed Thermo-Lag fire barrier configurations using information obtained from NEI and utility fire endurance test programs. GPU Nuclear has used this process to Establish the extent that installed barrier configurations can be bounded by previous tests, Determine the fire endurance capability (or " Actual Fire Rating") that installed barrier configurations, which are bounded by test, can be reasonably expected to provide, and Propose upgrades to installed barrier configurations where deemed necessary to achieve an acceptacle fire rating. 013/026

-.... - - - -.~ d Topical Raport 102 Rev. O Page 5 of 22 In order to evaluate the extent that installed barrier conditions at Oyster Creek can be bounded by test configurations and data in the Application Guide, GPU Nuclear performed the followings A walkdown of the fire areas / sones was conducted to document the inst' lled a barrier configurations with digitized computer images. The parameters identified by NEI during the industry fire endurance test program which pertain to fire endurance capability as identified in the Application Guide were included by GPU Nucisar in an electronic database. Each fire barrier system 'was separated into individual segments or elements for evaluation purposes. Individual elements are constituted by i one or more of the following distinguishing characteristics:

1) change in barrier construction technique; i

2) significant change in protected raceway or contents; 1

3) variation from applicable barrier installation requirements;

4) change in type of barrier material; or

5) change in orientation of protected raceway or change which necessitates a change in barrier construction technique.

Data collected during the walkdown and collected by a review of the I original fire barrier construction details were entered into the database to permit detailed comparisons of relevant parameters from the NEI, Temas j Utilities (TU), and TVA programs. Each of the test assemblies in the industry test programs was separated into individual segments or elements for evaluation purposes as were the l installed fire barrier systems and entered into the data base in order to l permit the detailed comparisons of relevant parameters with the installed fire barrier systems. The quality of the barrier installation was originally verified by Quality Control during the installation process by continuous in-process inspections using checklists for consistency. Final inspections performed by GPUN Quality Control were documented in the work control packages. The use of checklists standardised the attribute verification and allowed no significant deviations from the original design / installation requirements. Repairs are performed to-the same requirements as the initial installation. Repairs are performed either by or under the supervision of certified installers and are re-inspected by certified inspectors. Surveillance procedures ensure a refueling interval inspection to re-l verify the integrity of the installed fire barrier envelopes. Additionally, QC does make inspections of opportunity and has caused i repairs to be initiated. ) A Quality Control Program as defined in the Oyster Creek FHAR-Section 5 (and implemented by the GPUN Operational Quality Assurance Plan) was applied to the material, processes and installation and inspection personnel. GPUN did not contract this work out to a third party licensed i by TSI. GPUN contracted TSI to train and certify GPUN personnel for both i installation and inspection. i f GPUN's initial experience with TSI's material shipments resulted in returning large amounts of each shipment. All inspections were documented on receipt inspection reports. All line items of the contract were inspected. These inspections consisted of either 100% sampling using t engineering approved inspection and sampling plans (Military Standard 105D was the prescribed sampling method). Some of the problems experienced were as follows: 1 013/026 l

o T pical Report 102 Rev. O Page 6 of 22 THICKNESS GPUN receipt inspectors verified thickness by taking readings on each piece. Any piece not meeting the minimum thickness was rejected and J returned to TSI. VOIDS AND CRACKS (resulting from bending / fabrication) During receipt inspection, material which had voids or cracking beyond specified limits was returned. The material was subject to considerable field work (cut and fit). Any voids were filled with TSI trowel grade j material during installation. PHYSICAL DAMAGE I During receipt inspection, gouges, crushing and chipped edges could be repaired and accepted. Separation of the material from the stress skin j was a cause for rejection and return of the material. Edges were verified 'to be straight and square for proper alignment and fit-up during j installation. ) l To ensure receipt of consistent quality material from TSI, GPUN instituted QC checks by GPUN vendor surveillance at TSI's factory prior to release i for shipment. Thorough receiving inspection checks were performed to .j established inspection plans to insure they maintained engineering requirements. t DETAILED EXAMINATIONS GPU Nuclear performed detailed exams to confirm the accuracy of Quality I Assurance records for important parameters which are not visible by l walkdown. The initial exam consisted of dismantling a 3 hour conduit barrier. Additional exams as committed in GPUN letter C321-95-2277 dated 3 September 22, 1995 were performed on 1 hour and 3 hour conduit barriers, I hour box configurations and the HVAC duct barrier. Parameters such as thickness, panel rib orientation, stress skin orientation, buttering of i joints, joint gap width, unsupported spans and type of joint were confirmed and docuenented in Quality Verification Inspection Report PIR

1. 950065 dated October 25, 1995.

The results of these inspections confirmed conformance of installed Thermo-Lag with original installation ) and design requirements and provides a reasonable basis for reliance on ) Quality Assurance records and installation requirements for parameters l which are not visible by walkdown. In order to establish the actual fire rating of the installed fire barrier assemblies, the industry test data was evaluated. Actual fire rating is a term i used to designate the fire endurance rating of the barrier consistent with the acceptance criteria contained in NFPA 251 (ASTM E-119), " Standard Fire Tests of Building Construction and Materials". NRC Gereric Letter 86-10 Supplement 1, adapts the acceptance criteria of NFPA 251 to caale raceway fire barrier wraps. In 86-10, supp.1, the staf f bases acceptability of a fire endurance qualification test for fire barrier materials applied directly to a raceway or component to be successful if the " average" unexposed side temperature of the fire barrier system, as measured on the exterior surface of the raceway or component did not exceed 250 deg F above its initial temperature and a visual inspection of cables inside the raceway should show no signs of degraded conditions. Also, individual temperature readings should not exceed the 250 dog F temperature rise by more than 30 percent, or 325 dog F above the initial temperature. 013/026 1 l

Topical Rsport 102 Rev. 0 Page 7 of 22 To establish the barrier rating (ACTUAL RATING) of a test assembly, GPU Nuclear reviewed the temperature data for the test and identified that point in time when the first individual temperature. reading on the unexposed side of the fire barrier for the entire raceway in the test assembly, as measured on the exterior surface of the raceway or component, exceeded 325 dog F above the initial l temperature. Note that this method establishes a rating for all elements of a l particular raceway size based upon the weakest link in the raceway. While it is possible to establish individual ratings in a test involving straight conduit, radial bonds and condulets based upon thermocouple readings restricted to these elements in a test assembly, it is conservative to establish a common rating for ^ all elements of a raceway based upon the single high reading for the entire raceway. To establish the actual rating for an installed configuration or element, the installed configuratior.'s relevant parameters were compared with those of the industry tested configurations. If an acceptable match was found, it was selected and the installed configuration is considered bounded by an acceptable industry test configuration. The actual rating of the matching industry configuration becomes the actual rating of the installed configuration or element as documented by a detailed evaluation. The results of these detailed ) comparisons and evaluations will be listed in this submittal only. These.results will be retained in the electronic databare ( Doc. No. TLDB-oc-814-1) and i l digitized computer image library mentioned previously and will be available for NRC review or audit. 2.2 Establishing Cable Qualification Rating In addition to establishing the " Actual Fire Rating" as previously described, GPU Nuclear has utilized a combination of actual test data and theoretically derived data to document raceway temperatures inside the Thermo-Lag fire barrier envelopes which deviate from the acceptance criteria used for qualifying fire barrier configurations as outlined in Generic Letter 86-10 Supp.1. This section outlines the method that serves as the basis for establishing the fire endurance rating of the barrier when considering the cable qualification temperature or the maximum temperature inside the fire barrier envelope that is considered l acceptable to demonstrate cable functionality. The acceptable fire endurance rating is termed the " Cable Qualification Rating". GL 86-10 Supp. I permits an evaluation which demonstrates that cables would perform their intended function during and after a postulated fire exposure when the internal raceway temperatures exceed the GL 86-10 Supp.1 acceptance criteria. GPU Nuclear considers the cable Qualification Rating directly comparable to the requirement 1 of III.G for one hour or three hour fire barriers. In other words, a fire barrier having a cable qualification rating of at least 60 minutes is considered to be a one hour fire barrier as required by III.G. If less than 60 minutes, the cable qualification rating can be used to establish the basis for an exemption from the requirement for a one hour fire barrier. In order to compare the internal raceway temperatures to cable failure temperatures, GPU Nuclear performed the following 2.2a Identification of Protected Circuits The Oyster Creek Fire Hazards Analysis Report (FHAR) Document No. 990-1746 Revision 7 was compared with the Thermo-Lag Installation Drawings to identify i those circuits required for safe shutdown which are protected by the Thermo-Lag l fire barriers. 1 l 1 I l \\ 013/026 i l

..-._~ ~.- -.-..- - -.- - - - ~ e I [ Topical Raport 102 Rev. O Page 8 of 22 2.2b Identification of cable oualification T- =catures (Note this is not to be confused with environmental qualification temperatures) Generic Letter 86-10, Supplement 1, Attachment to Enclosure 1, " Acceptable Methods for Demonstrating Functionality of Cables Protected by Raceway Fire Barrier Systems During and After Fire Endurance Test Exposure" provides the means 'for establishing the maximum temperature inside the fire barrier envelope that is considered acceptable to demonstrate cable functionality. GPU Nuclear used the recommended analysis of Section VI in the aforementioned attachment entitled " Cable Thermal Exposure Threshold" to perform.this analysis. After identifying the circuits required for safe shutdown as described in "2.2a" i above, the cable manufacturer was identified and the thermal exposure threshold (TET) temperature limit or insulation failure threshold temperature limit was obtained from Sandia Test Report SAND 90-0696 May,1991 "An Investigation of the Effects of Thermal Aging on the Fire Damageability of Electrical Cables". Insulation damage leading to a short circuit per the tests conducted in the aforementioned investigation is defined as 15 milliamps (mA) leakage current with cable conductor temperatures at 90 dog C which means that the cable was tested at the equivalent of full load current. The insulation failure threshold temperature limit is the temperature beyond which the insulation is expected to degrade causing a short circuit. While not the same as the short circuit rating as defined in Generic Letter 86-10, Supplement 1, these tests conclusively demonstrate cable functionality at elevated temperatures since under fire conditions the endurance rating of the barrier is based upon the point in time when the threshold temperature is reached. The aforementioned Sandia tests on the other hand, subjected cables to the insulation failure threshold temperature for at least 80 minutes without failure. Fire endurance ratings do not assume sustained operation under these elevated temperature conditions. The rating is based upon the duration of the test and the point in the time of the test when 3 the insulation failure threshold temperature is reached. This method is conservative because, as stated above, the Sandia tests subjected cables to the j insulation failure threshold temperature for AT LEAST 80 MINUTES. 2.2c Raceway Temocratures Inside Thermo-Lao Fire Barrier Envelones In order to follow the guidance of Generic Letter 86-10, Supplement 1 for ' establishing maximum allowable raceway temperatures inside the fire barrier envelope, additional evaluation of industry test data and consideration of the operating cable temperatures within the fire barrier system at the onset of the fire exposure were performed, This is necessary in order to compare test results with the TET temperature limit and establish a " cable qualification rating" for ~ the fire barrier envelope. Industry test data which was used to establish the " actual" fire rating described previously was further evaluated to document internal raceway temperatures beyond NRC acceptance criteria as defined in GL 86-10. To be consistent, the evaluation was limited to the temperature readings on the unexposed side of the fire barrier as measured on the exterior surface of the raceway. The evaluation documented the temperatures and the duration of the fire test at the aforementioned locations in increments of approximately 100 dog F up to the point where the test was terminated. og. for 3/4 in. conduit in NEI test 2-1, the actual rating is based upon a conduit surface temperature of 387 dog F at 27 minutes. Since the test was continued beyond this point, the following data was utilized to compare conduit surface temperatures with the insulation f failure threshold temperature mentioned in 2.2b. e i 013/026

s. Topical R3 port 102 4 Rev. O Page 9 of 22 3/4" Conduit Outside surface Temp (Maximum) l TIME (MIN) TEMP (DEG F) 28 396 36 499 41 609 44 720 By reviewing the temperature data in this case it is apparent that the barrier f ailed to provide any protection after about 44 minutes. In the case of the 3/4"' conduit, Thermo-Lag material was consumed directly exposing the conduit to test j oven temperatures. According to NEI test report 2-1, no joint failures occurred. i If the test was terminated before a duration of 60 minutes, it was necessary to develop a multi-dimensional heat transfer model to extrapolate temperatures e inside the raceway out to 60 minutes for the 2", 4" and 6" conduit assemblies since no temperature data was avetlable. The model was verified against actual i industry test data to establish confidence in its validity. Comparison with test data on NEI " upgraded" conduit raceway configurations (NEI test 1-6) was performed. This is considered legitimate since the upgrades consisted of joint reinforcement which was accomplished by adding stress skin and trowel grade material at joints; however, the thickness of the barrier remained the same at i 1/2" except in the areas where the joints were reinforced. Burnthrough did not occur on the 1/2" Thermo-lag after 60 minutes. It is also assumed that the baseline joints of NEI test 2-1 would not have failed had the test run the full 60 minutes. As stated above, joint failure of the 3/4" conduit configurations did not occur. It is reasonable to assume that joint failure would not have occurred on the larger size raceways in NEI test 2-1 and that burnthrough would not have occured on 1/2" thick fire barriers on these raceways because another test (NEI Test 1-6) on the same nominal thickness for larger size raceways that went the full 60 minutes did not burnthrough. It is therefore considered reasonable to extrapolate internal raceway temperatures for NEI Test 2-1 out to 60 minutes for the purpose of comparison with cable qualification temperatures. The results of this model are documented in calculation C-9000-814-5310-002. This calculation is not included here but is available for NRC review or audit. It is necessary to account for the initial temperature of the cable within the fire barrier prior to the onset of the fire as compared to the test configurations. As such, the cable qualification rating is established based upon the time it takes for the internal raceway temperature to get within no less than 70 deg F of the insulation failure threshold temperature. This 70 dag F value provides the margin to account..for the. differences between actual room ambient temperatures plus cable temperature rise due to the insulating effects of the barrier and those ambient temperatures measured during industry testing. The industry tests were performed with initial ambient temperatures no less than 50 dog F while the maximum design room ambient temperatures are 104 deg F at oyster Creek. The additional temperature rise inside the fire barrier envelopes due to operating cables is 7 deg F maximum based upon test results documented in letter G/C/TMI-1CS/16503 dated September 15,

1988, J.

Brendien to J.W. Langenbach, "TSI Derating Check". While this test was conducted for installed cable tray barriers at TMI-1, it is reasonable to apply these.results to Oyster Creek since the barrier thickness at oyster Creek is the same as at TMI-1. The main factor impacting heat rise inside a raceway would be the thickness of the barrier and the number of circuits energized inside the raceway. The TMI-1 test was conducted in trays containing continuously energized power circuits which can be considered comparable to raceways at Oyster Creek. Both TMI-1 and Oyster Creek have barriers with a nominal thickness ranging from a minimu.a of 1/2 to 5/8 inch for one hour barriers and 1 to 1-1/8 inch for three hour barriers. Therefore I using THI-1 test results to factor into initial raceway temperatures at Oyster Creek is reasonable. The additional effect of the internal temperature of the raceway due to continuously energized power circuits is only about 10% of the factor being used. Adding the difference between the max room ambient 013/026

______.__.____..__._m._ Tcpical Rsport 102 Rev. O Page 10 of 22 ] temperature and the minimum ambient test temperature (104-50=54 dog F) to tEie temperature rise of 7 dog F yiolds a total factor of 61 dog F. The factor being l used is 70 dog F. This establishes that the cable qualification rating is consistent with maximum cable operating temperatures. This meets the guidance in GL 86-10 for an engineering analysis to demonstrate the functionality of cables inside fire barrier envelopee. 2.3 Evaluating Fire Essards i To evaluate the cable qualification ratings of the Thermo-Lag fire barriers which do not meet the requirements of Appendix R Section III.G (ie. less than 1 hour or 3 hour), two methods are employed. These methods consider the insitu and transient hazards in a fire area / zone. These are described as follows: 2.3a Method 1 The first method is typical of the " traditional approach". A comparison of the cable qualification rating with the c.v0rall fire loading is performed using 80,000 BTU /Ft2 as equivalent to a one hour f. ire to develop a fire load to rating factor hereir referred to as the rating factor. A combustible loading of 80,000 . BTU /Ft2 has been considered as equivalent to the heat release in an ASTM E-119 i Test oven for a one hour fire duration test. A source reference to support this assumption is Table 6-6a of the NFPA' Fire Protection Handbook, Seventeenth Edition. If the Thermo-Lag fire barrier is located in a fire area / zone which is provided with an area wide fire suppression system, the aforementioned factor is multiplied by an additional factor of 3 to account for the additional margin provided by the suppression system. This f actor is an assumption based upon the fact that Appendix R Section III.G.2.c reduces the requirement for a, hour c barrier in III.G.2.a to 1 hour with the presence of a suppression and detection system; hence the additional factor of 3. The rating factor is used for assessing the fire hazard with the actual rating of a raceway fire barrier that does not meet the requirements of Appendix R. No s strict acceptance criteria is established as the acceptability of a rating factor must consider all fire hazard and fire protection features where the raceway fire barrier is located. Acceptability is therefore based upon engineering judgement. The following reference point will be taken into consideration in judging the acceptability of a rating factor. The basis for this is drawn from the Oyster Creek FHAR, paragraph 1.3 " Delineation of Fire Areas /Eones". To summarize, this section of the FHAR established that reasonable assurance that a fire will not propagate from one fire zone to another is provided in the plant fire hazards analysis by passive and active fire protection features. Through this assurance, it can be justified that fire zones which may not consist of continuous wall to wall, floor to ceiling boundaries of fire rated construction can be analyzed by themselves. Fire zones are subdivisions of fire areas which take into consideration the physical boundaries which exist between one fire zone and another in the same fire area. Fire zone boundaries can consist of non-rated physical boundaries with penetrations sealed with materials installed in accordance with configurations similar to a one hour minumum fire barrier. The criteria for acceptability of such a configuration is that the combustible loading on either side of the boundary is less than 40,000 BTU /Ft2, or 1/2 the equivalent of a one hour fire duration as discussed above or a rating factor 1 of 2. I 013/026 l l

Topical R: port 102 Rev. O Page 11 of 22 The following example is provided to illustrate how the cable qualification rating factor (RF) is calculated for a 1" conduit having a cable qualification rating of 41 minutes in an area provided with automatic fire suppression having a fire loading of 20,223 BTU /FT2: FIRE LOAD (FL) CABLE QUAL. RATING (CQR) RESULT (RF)=CQR/FL 15 min. 123 min. RF=8.2 (20,223 BTU /Ft2) (= 41 min.x 3 w/ suppression) NOTE THAT THIS METHOD IS CONSERVATIVE BECAUSE IT PRESUMES COMPLETE COMBUSTION OF ALL COMBUSTIBLES IN THE FIRE AREA OR SONE OF CONCERN. 2.3b Method 2 The second method for evaluating the cable qualification ratings of the Thermo-Lag fire barriers with respect to the actual fire hazards is by the use of fire modelling. GPU Nuclear has used EPRI's enhanced FIVE Fire modeling to approximate exposure time-temperature history for Thermo-Lag configurations in five of seven fire zones. The results are contained in the Appendices to this submittal. The following summarizes the EPRI technique and the results: The source document for this discussion is EPRI Report Project 3385-05 " Methods for Evaluation of Cable Wrap Fire Barrier Performance". The fire hazard tool will calculate location dependent exposure time-temperature histories, as discussed above, resulting from defined fire scenarios. These scenarios are established by assessing the location and distribution of in-situ and transient hazards in the area under evaluation. Exposures can be determined at any selected location. Fire growth and propagation are based on actual fire test data. The tool j therefore provides estimates of fire exposures for installed raceway fire barrier envelopes including the time required for the fire to grow and propagate and the time to reach critical temperatures at the envelopes. The timing estimated by the tool provides a basis for quantitative assessment of important time dependent j measures such as fire brigade response. The fire modelling method is described in FIVE (EPRI's Fire Induced Vulnerability Evaluation, TR.100370), Section 10.4, April 1992. The model utilizes the FIVE plume / ceiling jet / hot gas layer correlations to determine pre-flashover conditions within the evaluated fire zone / area. Enhancements to the FIVE fire modelling method, documented in the EPRI " Fire Risk Analysis Implementation Guide" (TR 104030), Draft January,1994, are incorporated 1 into the fire hazard tool. These enhancements are based on conservative interpretation of actual fire test data and provide the capability of calculating fire growth / propagation in addition to providing improved definition of the j burning characteristics of combustible materials typically found in power plants. 1 The fire hazard tool also uses data published by NEI regarding the burning characteristics of Thermo-Lag (NUMARC Thermo-Lag 330-1 Combustibility Evaluation Methodology Plant Screening Guide, September 1993). 013/026 l 1

l Tcpical R: port 102 Rev. O Page 12 of 22 The fire hazard tool calculates temperatures at specified time steps in order to define temperature exposure profiles at raceway fire barrier envelope locations. l The temperatures at each time step, prior to fire zone / area compartment i flashover, are determined from the FIVE equations considering the enhancements incorporated in the Fire Risk Analysis Implementation Guide. The temperature results are not sensitive to the size of the time step selected. Compartment flashover is conservatively assumed to occur when the hot gas layer reaches 1000 dog F. The temperature profile at tha time is taken as the ASTM E-119 time-temperature history at all locations within the fire zone / area compartment. Field walkdowns are performed to identify and locate all potential fire ignition aources which could potentially impact Thermo-Lag raceway fire barrier envelopes. FIVE Table 1.2 is used as a guide in identifying the types of credible ignition sources. The approach described in EPRI's Fire Risk Analysis Implementation Guide (pages '4-18 through 4-27 and Appendix E) is used to screen those ignition sources that do not result in damaging fires and define scenarios with potential for damage. The field walkdowns use pre-calculated critical distance values which are developed using the FIVE fire modelling methods (FIVE Section 10.4). Ignition sources which are determined to have no potential to cause damage (other than to the ignition source itself) are eliminated from further consideration. Ignition sources which are determined to have the potential of igniting adjacent cable, Thern.o-Lag or other combustibles are identified for detailed fire modelling to be performed as described below. FIVE analyses are performed using computerized Excel spreadsheets similar in format to FIVE worksheets 1,2 and 3. Time-temperature histories are developed by inputting the heat release rate (HRR) and total heat content of the fuel (Q Tot) for the fire source into the FIVE equations at the appropriate time steps. Most material specific parameters pertinent to fire growth and propagation are extracted directly from FIVE. Physical configurations are conservatively determined from the walkdowns. There are conservatisms built into the model. One example of the conservatism of the fire model is the assumption that the fire begins to burn at its peak release rate at time =0. This assumption results in the shortest time to reach peak temperatures (in the plume, ceiling jet and hot gas layer). This is conservative as it maximizes the incident heat flux and temperatures at the target raceway fire barrier envelope. NOTE ALSO THAT THE MODEL TAKES NO CREDIT FOR THE PRESENCE OF AUTOMATIC SUPPRESSION OR ANY MANUAL FIRE FIGHTING ACTIVITIES. Implementation of the fire hazard tool is accomplished as follows: Equipment is selected for evaluation as s' fixed ignition source based on criteria , contained in the FIVE methodology guide. Equipment that does not meet the selection criteria may be included for evaluation as an ignition source if it is in close proximity to an important piece of equipment or associated cable. Fire modelling of fixed ignition sources is performed in a stepwise fashion using the FIVE methodology and Fire Risk Analysis Implementation (FPRA) methodologies developed by EPRI. The detailed steps involved in modelling are provided below. Not all steps of the fire modelling were necessary to identify and evaluate the effect of each ignition source on other cables and equipment. In most cases, the conclusion that the ignition source causes no adverse impact was reached upon completion of the deterministic steps 1 through 9. 1. Determine the physical characteristics of the fire zone. This requires floor area, ceiling height, room volume, maximum fire zone ambient temperature and identification of any features that could affect the analysis. This informa-tion is obtained from plant design documents. 013/026

Tep val R2 port 102 Rev. O Page 13 of 22 2. Identify all fixed ignition sources in the fire sone. This information is collected from plant databases and verified by walkdown of the zone. During the course of the walkdown,, any material storage locations are also identified. These locations are treated as fixed ignition sources. 3. Select an appropriate heat loss factor. The heat loss factor accounts for heat absorbed into the fire zone floor ceiling and walls. A figure of 0.94 i is utilized in modelling based on guidance in the EPRI Report Project 3385-05 " Methods for Evaluation of Cable Wrap Fire Barrier Performance". Longer duration scenarios (greater than 5 minutes) such as the formation of a damag-ing hot gas layer, have been experimentally determined to have heat loss factors in the range of 0.94 to 0.98. 4. Determine target damage and ignition temperatures. For Thermo-Lag, an igni-tion temperature of 1000 DEG F is used based on guidance in the Thermo-Lag Combustibility Evaluation Methodology plant Screening Guide. 5. Determine target damage radiant heat fluxes. For Thermo-Lag, an ignition heat flux of 2.2 BTU /s/FT2 is used based upon guidance in the Thermo-Lag Combustibilty Evaluation Methodology Plant Screen Guide. 6. Identify Heat Release Rates (HRRs) for the different categories o'f fixed ignition sources located within the fire sone. The FPRA manual provides HRRs for most of the standard equipment categories. In cases where HRRs are not explicitly stated, judgement is used to select an HRR. For fixed ignition sources containing oil, the HRR is calculated from the amount of oil contained in the source and the potential surface area of the oil spill. 7. Assign a location factor *for each fixed ignition source in the fire zone. The location factor accounts for the higher plume temperatures found for fires near walls and in corners. 8. Calculate in-plume damage heights and radiant damage ranges for each fixed ignition source located in the fire zone. This calculation uses equations contained in the FIVE manual. 9. Determine targets for each ignition source. This step utilizes the damage heights and ranges calculated in step 8. Any piece of equipment, cable, conduit or tray within the damage heights and ranges were noted. Targets that are outside of the radiant damage range and are not located in the plume are evaluated for potential damage from the ceiling jet. Fixed ignition sources without any potential targets are excluded from further analysis other than the loss of the fixed ignition source itself and impact of forming a damaging hot gas layer. .If a target is found to be within the ignition range of a fixed ignition source, the target is in turn treated as an ignition source and additional targets are identified. This models the propagation of a fire from a source to a series of combustible targets. Additionally, the heat content of the ignited target must be considered to determine if the formation of a damaging HGL is possible. Modeling of transient ignition sources and combustibles is similar to fixed ignition sources with some additional steps to account for the uncertainty in location and the amount of combustibles. These additional steps are as follows:

10. Determine what transient combustibles are likely to be present or traverse the fire zone. Due to the potential for large HRRs from oil fires, special emphasis was placed on identifying oil lubricated components as well as any such equipment that was likely to be reached by traversing the fire zone under analysis. The type of container in which oil is transported is also of importance.

Oil transported in 55 gallon drums or NFPA approved containers are assumed to be not exposed to potential ignition sources. 013/026

-. ~. - 8 Topical Rsport 102 Rev. O Page 14 of 22

11. Select an appropriate transient fuel package ' based on what types of transient materials are likely to be present in the fire zone.

The fuel packages and their associated HRRs are selected from Sandia National Laboratory (SNL) testing.

12. Determine target sets based on specific floor locations. Target sets that could be damaged by a fire involving the transient fuel package are identified.

Once the fire hazard tool has determined the time-temperature history for each scenario, the EPRI " Barrier Performance Tool" is then utilised to establish the exposure to the thermo-lag in each scenario equivalent to an ASTM E-119 exposure fire. Inputs to the tool are the calculated time temperature history as determined by the fire hazard tool. This tool is based upon the premise that two different exposure histories, which would cause two identical fire barrier raceway configurations to reach the same failure point would have equal areas under their incident heat flux curves. The " Barrier Performance Tool" therefore calculates the total heat load for the e'xposure history developed for each scenario by the fire hazard tool. The duration as expressed by the " Cable Qualification Rating" is then compared. with the calculated equivalent ASTM E-119 duration for each scenario to assess tne calculated severity of each scenario with an ASTM E-119 exposure. 3.0 stnetARY OF RESULTS The results of applying the above methodology towards establishing an actual fire rating and a cable qualification rating for Thermo-Lag fire barriers are as follows: 3.1 Office "A" 480 V Switchgear Room (Fire Bene OB-F5-6&) ENVELOPE TYPE NO. ACTUAL CABLE NEI TEST NO. ELEMENTS RTG. RTG. QUALIFICATION CNXA1125 2" Conduit 5 39 min. 21 min. 2-1 2" Radial Bend 5 39 min. 21 min. 2-1 Conduit 2" Box (Condulet) 2 39 min. 21 min. 2-1 2" Conduit (Penet.) 1 39 min. 21 min. 2-1 12"x6"x4" Box Penet. 1 HOLD The 2" conduit, 2" conduit radial band, 2" condulet and 2" penetration elements will be upgraded to provide an " actual" fire rating of one hour and will therefore meet the requirements of III.G.2.c. Materials which successfully demonstrate a 60 minute fire endurance rating for these conduits (either by themselves or in conjunction with the existing Thermo-Lag installations) will be installed. Such changes will not be submitted for NRC approval but handled under existing licensing conditions; le. 50.59 process. The commitment is to upgrade these envelopes to an " actual" fire rating of 60 minutes. All barriers in this fire zone currently will be upgraded to meet the requirements of Appendix R section III.G.2.c. Therefore, no exemptions will be necessary. 013/026

._ -~... - - + T pical R: port 102 Rev. O Page 15 of 22 3.2 office "B" 480 V switchgear Room (Fire Rose 05-F5-65) ENVELOPE TYPE NO. ACTUAL CABLE NEI. TEST NO. ELEMENTS RTG. RTG. QUALIFICATION CNCA1041 3" Conduit 1 39 min. 60 min. 2-1 3" Box (Condulet) 1 -39 min. 60 min. 2-1 3" Conduit i HOLD (Penetration) CNCA1043 3" Conduit 1 39 min. 60 min. 2-1 7 3" Box (Condulet) 1 39 min. 60 min. 2-1 l 3" Box Penetration 1 HOLD CNPA1042 1" Conduit 2 27 min. 41 min. 2-1 l 1" Box (condulet) 1 27 min. 41 min. 2-1 1" Box Penetration 1 HOLD 1" Conduit 1 HOLD (Penetration) ND. NUMBER 10'X5'X1' Box 1 HOLD j DUCTWORK From 40"X16" to NA NO RATING 30"x18" To summarize the above results, 3" conduit and 3"condulet elements meet the requirements of III.G.2.c because their " CABLE QUALIFICATION RATING" is at least one hour. The heat transfer model which projects temperatures inside these raceways out to 60 minutes was used to compare raceway temperatures at 60 minutes to the cable insulation failure threshold temperatures for required safe shutdown circuits inside these raceways. For these raceways, the projected raceway temperatures at 60 minutes are below the cable insulation failure threshold temperatures by more than 70 deg F; therefore these raceway barriers have a rating of 60 minutes. No additional justification is required for these barriers. ] The 1"

conduit, 1"

condulet and I" penetration elements will be upgraded to ) provide an " actual" fire rating of one hour and will therefore meet the 1 requirements of III.G.2.c. Materials which successfully demonstrate a 60 minute fire endurance rating for these conduits (either by themselves or in conjunction with the existing Thermo-Lag installations) will be installed. Such changes will not be submitted for NRC approval but handled under existing licensing conditions; ie. 50.59 process. The commitment. is to upgrade these envelopes to an " actual" fire rating of 60 minutes. The 10'X 5'X1' box (HOLD) Based upon existing information, a fire endurance rating cannot be established for the Thermo-Lag installed on the ductwork fire barriers. These barriers provide insulation for ductwork which is part of the ventilation system for the "A" Switchgear Room which is Fire Zone OB-FZ-6A. The insulated ductwork is routed through the "B" Switchgear Room or Fire zone OB-FE-65. It was provided with the Thermo-Lag for the ductwork in the event of a fire in the "B" switchgear Room to maintain ventilation in the "A" Switchgear Room. A " rated" configuration for these barriers is not critical to the ability of the ventilation system to maintain an adequate environment in the "A" Switchgear Room. In fact, acceptable l duct internal temperatures for a " rated" barrier (250 dog F above ambient) are not desirable from a practical standpoint even though a " rated" barrier may technically meet Appendix R. GPU Nuclear does not plan on establishing a fire endurance rating for the Thermo-Lag installed on ductwork. Corrective action will consist of insuring adequate ventilation in fire zone OB-FE-6A if a fire occurs in fire zone OB-FZ-68. Adequate ventilation may be insured by operating procedure changes if analysis of the existing ventilation system configuration 013/026 .-o u

T:pical R port 102 Rev. O Page 16 of 22 with Thermo-Lag serving as duct insulation provides an adequate basis for leaving the existing configuration as is. Preliminary results of our analysis indicates that loss of ventilation to the "A" Switchgear Room results in acceptable temperatures af ter one hour for equipment in this room required to operate in the event of a fire in the "B" Switchgear Room. If adequate ventilation cannot be proven analytically, corrective action may take the form of modifications to the "A" Switchgear Room ventilation system. All electrical raceway barriers in this fire zone currently meet the requirements of Appendix R, section III.G.2.c or will be upgraded to meet the aforementioned requirements. Therefore, no exemptions will be necessary. The Thermo-Lag installed on the ductwork in this zone does not require a fire endurance rating. Corrective action will be based upon completion of a - ventilation system analysis which considers Thermo-Lag as duct insulation, not a " rated" barrier as stated previously. By insuring adequate ventilation for the "A" Switchgear Room if a fire occurs in the "B" Switchgear Room, either procedurally or by modification, GPU Nuclear will meet the requirements of Section IIIG.l.a in that "one train of systems necessary to achieve and maintain hot shutdown from either the control room or emergency control station (s) is free of fire damage". Completion of the aforementioned ventilation system analysis will identify the scope of work required to insure adequate ventilation for the "A" Switchgear Room. 3.3 Reactor Building 51' Elevation (Fire Bone RS-FE-lD) ENVELOPE TYPE NO. ACTUAL CABLE NEI TEST NO. ELEMENTS RTG. RTG. QUALIFICATION CGCA1027 2" conduit 10 39 min. 60 min. 2-1 2" Radial Bend 8 39 min. 60 min. 2-1 Conduit 2* Box (Condulet) 2 39 min. 60 min. 2-1 2" Conduit 1 39 min. 60 min. 2-1 (Penetration) 2" Box Penetration 1 HOLD CGPA3026 2" conduit 10 39 min. 60 min. 2-1 2" Radial Bend 8 39 min. 60 min. 2-1 Conduit 2" Box (Condulet) 2 39 min. 60 min. 2-1 2" Conduit 1 39 min. 60 min. 2-1 (Penetration) 2" Box Penetration 1 HOLD CRCA1026 2" Conduit 4 39 min. 60 min. 2-1 2" Radial Bend 4 39 min. 60 min. 2-1 Conduit 2" Conduit 1 39 min. 60 min. 2-1 (Penetration) 2" Conduit 1 HOLD (Penetration) To summarize the above results, 2" conduit, 2" conduit radial bends, 2"condulet and 2" penetration elements meet the requirements of III.C.2.c because their " CABLE QUALIFICATION RATING" is at least one hour. The heat transfer model which projects temperatures inside these raceways out to 60 minutes was used to compare raceway temperatures at 60 minutes to the cable insulation failure threshold temperatures for required safe shutdown circuits inside these raceways. For 013/026

o Topical R: port 102 Rev. O Page 17 of 22 these raceways, the projected raceway temperatures at 60 minutes are below th'e cable insulation f ailure threshold temperatures by more than 70 dog F; therefore these raceway barriers have a rating of 60 minutes. No additional justification is required for these barriers. ENVELOPE TYPE NO. ACTUAL CABLE NEI TEST NO. ELEMENTS RTG. RTG. QUALIFICATION 62-153 2" Conduit 4 39 min. 56 min. 2-1 2" Radial Bond 4 39 min. 56 min. 2-1 conduit 2" Conduit 1 39 min. 56 min. 2-1 (Penetration) 2" Conduit 1 HOLD (Penetration) CGCR2086 la conduit 3 27 min. 36 min. 2-1 1" Radial Bend 3 27 min. 36 min. 2-1 Conduit la Box (Condulet) 1 27 min. 36 min. 2-1 1" conduit 2 27 min. 36 min. 2-1 (Penetration) CGCR3021 1" Conduit 9 27 min. 41 min. 2-1 1" Radial Bend 7 27 min. 41 min. 2-1 conduit 1" Box (Condulet) 3 27 min. 41 min. 2-1 1" Conduit 2 27 min. 41 min. 2-1 (Penetration) The above three envelopes (62-153, CGCR2086, CGCR3021) will be the subject of an exemption request from the requirement in Appendix R, Section III.G.2.c, for a ) one hour rated fire barrier in fire zona RB-FZ-1D. The heat transfer model which projects temperatures inside these raceways out to 60 minutes was compared with the cable insulation failure threshold temperatures for required safe shutdown circuits inside these raceways. For these raceways, the projected raceway j temperatures at the duration listed are below the cable insulation failure threshold temperatures by more than 70 dog F; thereferc thcas raceway barriers I have a cable qualification rating as listed absve. ENVELOPE TYPE NO. ACTUAL CABLE NEI TEST NO. ELEMENTS RTG. RTG. QUALIFICATION CGXA3028 1" Conduit 8 27 min. 14 min. 2-1 1" Radial Bend 13 27 min. 14 min. 2-1 Conduit 1" Box (Condulet) 2 27 min. 14 min. 2-1 i 1" conduit 1 27 min. 14 min. 2-1 (Penetration) 1" Box Penetration 1 HOLD The 1" conduit, 1 inch conduit radial bends,1 inch condulet, and 1 inch conduit penetration elements will be upgraded to provide an " actual" fire rating of one hour and will therefore meet the requirements of III.G.2.c. Materials which i successfully demonstrate a 60 minute fire endurance rating for these conduits (either by themselves or in conjunction with the existing Thermo-Lag installations) wil.1 be installed. Such changes will not be submitted for NRC approval but handled under existing licensing conditions; le. 50.59 process. The commitment is to upgrade these envelopes to an " actual" fire rating of 60 minutes. 013/026 i

.....~.. i T:pice.1 R; port 102 Rev. O Page 18 of 22 3.4 Reactor Building 23' Elevation (Fire Some RB-F5-15) ENVELOPE TYPE NO. ACTUAL CABLE NEI TEST NO. ELEMENTS RTG. RTG. QUALIFICATION CGPA2008 3" Conduit 2 39 min. 60 min. 2-1 3" Box (condulet) 3 39 min. 60 min. 2-1 3" conduit 1 39 min. 60 min. 2-1 (Penetration) 3" Conduit 1 HOLD (Penetration) ENVELOPE TYPE NO. ACTUAL CABLE NEI TEST NO. ELEMENTS RTG. RTG. QUALIFICATION CRCA1026 2" Conduit 1 39 min. 60 min. 2-1 2" Radial Bend 3 39 min. 60 min. 2-1 conduit 2" Box (condulet) 1 39 min. 60 min. 2-1 l 2" Conduit 2 HOLD ] (Penetration) To summarize the above results, 2"and 3" conduit, 2" conduit radial bends and 2"and 3" condulet elements meet the requirements of III.G.2.c because their " CABLE QUALIFICATION RATING" is at least one hour. The heat transfer model which projects temperatures inside these raceways out to 60 minutes was used to compare raceway temperatures at 60 minutes to the cable insulation failure threshold temperatures for required safe shutdown circuits inside these raceways. For these raceways, the projected raceway temperatures at 60 minutes are below the cable insulation failure threshold temperatures by more than 70 dog F; therefore these raceway barriers have a rating of 60 minutes. No additional justification is required for these barriers. ENVELOPE TYPE NO. ACTUAL CABLE NEI TEST NO. ELEMENTS RTG. RTG. QUALIFICATION 62-153 2" conduit 1 39 min. 56 min. 2-1 2" Radial Band 3 39 min. 56 min. 2-1 Conduit 2" Box (Condulet) 1 39 min. 56 min. 2-1 2" Conduit _2 HOLD (Penetration) i CGCA2010 1.5" Conduit 2 27 min. 36 min. 2-1 1.5" Box (condulet) 3 27 min. 36 min. 2-1 1.5" Conduit 1 27 min. 36 min. 2-1 1 (Penetration) i 1.5" Conduit 1 HOLD (Penetration) CGCR3021 1" conduit 10 27 min. 41 min. 2-1 I 1" Radial Bend 9 27 min. 41 min. 2-1 conduit-l 1" Box (Condulet) 3 27 min. 41 min. 2-1 1" Conduit 1 27 min. 41 min. 2-1 (Penetration) i Drywell Pen. Box 1 HOLD PEN Drywell Pen. Box 4 HO'E I l 013/026 J am e - + +, -- y.

i I Tepical R3 port 102 Rev. O i Page 19 of 22 j The above envelopes (62-153, CGCA2010, CGCR3021) will be the subject of an exemption request from the requirement in Appendix R, Section III.G.2.c, for a one hour rated fire barrier in fire zone RB-FZ-1E. The heat transfer model which projects temperatures inside these raceways out to 60 minutes was compared with a the cable insulation failure threshold temperatures for required safe shutdown j circuits inside these raceways. For these raceways, the projected raceway temperatures at the duration listed are below the cable insulation failure threshold temperatures by more than 70 dog F; therefore these raceway barriers have a cable qualification rating as listed above. j i ENVELOPE TYPE NO. ACTUAL CABLE NEI TEST i NO. ELEMENTS RTG. RTG. QUALIFICATION CGXR2051 1" Conduit 3 27 min. 23 min. 2-1 1" Radial Bond 3 27 min. 23 min. 2-1 Conduit 1" Box (Condulet) 1 27 min. 23 min. 2-1 l 1" Conduit 1 27 min. 23 min. 2-1 j (Penetration) ) Drywell Pen. Box 1 HOLD CRXR2050 1" Conduit 11 27 min. 23 min. 2-1 l 1" Radial Bond 10 27 min. 23 min. 2-1 Conduit 1" Box (Condulet) 3 27 min. 23 min. 2-1 1" Conduit 1 27 min. 23 min. 2-1 (Penetration) i Drywell Pen. Box 1 HOLD i Bend l The 1" conduit, 1 inch conduit radial bends,1 inch condulet, and 1 inch conduit j penetration elements will be upgraded to provide an " actual" fire rating of one hour and will therefore meet the requirements of III.G.2.c. Materials which l successfully demonstrate a 60 minute fire endurance rating for these conduits (either by themselves or in conjunction with the existing Thermo-Lag installations) will be installed. Such changes will not be submitted for NRC approval but handled under existing licensing conditions; le. 50.59 process. The i commitment is to upgrade these envelopes to an " actual" fire rating of 60 minutes. 3.5 Reactor Building -19' Elevation (Fire Bone RB-FE-1F2) ~ ENVELOPE TYPE NO. ACTUAL CABLE NEI TEST NO. ELEMENTS RTG. RTG. QUALIFICATION CGPA2OO2 1.5" Conduit 2 27 min. 41 min. 2-1 1.5" Radial Bend 1 27 min. 41 min. 2-1 i conduit l 1.5" Box (Condulet) 2 27 min. 41 min. 2-1 1.5" conduit 1 27 min. 41 min. 2-1 (Penetration) 1.5" Box Penetration 1 HOLD The above envelope (CGPA2OO2) will be the subject of an exemption request from the requirement in Appendix R, Section III.G.2.c, for a one hour rated fire barrier in fire zone RB-FZ-1F2. The heat transfer model which projects temperatures inside these raceways out to 60 minutes was compared with the cable insulation f ailure threshold temperatures for required safe shutdown circuits inside these raceways. For these raceways, the projected raceway temperatures at the duration listed are below the cable insulation failure threshold temperatures by more than 70 dog F; therefore these raceway barriers have a cable qualification rating as listed above. 013/026 I --e

t Tipical R port 102 Rev. O Page 20 of 22 3.6 Turbine Building Basement, south End (FIRE BONE TB-PS-11D) ENVELOPE TYPE NO. ACTUAL CABLE NEI TEST NO. ELEMENTS RTG. RTG. QUALIFICATION 86-71 4" conduit 11 50 min. 60 min. 2-1 4" Radial Bend 10 50 min. 60 min. 2-1 Conduit 4" Conduit 2 HOLD (Penetration) 4" Penetration 1 HOLD To summarize the above results, 4" conduit, and 4" conduit radial bends meet the requirements of III.G.2.c because their " CABLE QUALIFICATION RATING" is at least one hour. The heat transfer model which projects temperatures inside these raceways out to 60 minutes was used to compare raceway temperatures at 60 minutes to the cable insulation failure threshold temperatures for required safe shutdown circuits inside these raceways. For these raceways, the projected raceway temperatures at 60 minutes are below the cable insulation failure threshold temperatures by more than 70 dog F; therefore these raceway barriers have a r'ating of 60 minutes. No additional justification is required for these barriers ENVELOPE TYPE NO. ACTUAL CABLE NEI TEST NO. ELEMENTS RTG. RTG. QUALIFICATION 14-25 3.5" Conduit 2 39 min. 56 min. 2-1 3.5" Radial Bend 2 39 min. 56 min. 2-1 Conduit 3.5" Conduit 1 HOLD (Penetration) 3.5" Box Penetration 1 HOLD 14-28 2.5" Conduit 1 39 min. 56 min. 2-1 2.5" Radial Bend 2 39 min. 56 min. 2-1 conduit 2.5" Conduit 2 HOLD (Penetration) 62-93 1.5" Conduit 6 27 min. 36 min. 2-1 1.5" Radial Bend 5 27 min. 36 min. 2-1 Conduit 1.5" Box (condulet) 1 27 min. 36 min. 2-1 12"x12" Box 1 HOLD 86-71~ 4" Box 1 50 min. 50 min. 2-2 CGCTB017 1" conduit 11 27 min. 36 min. 2-1 1" Radial Bend 12 27 min. 36 min. 2-1 Conduit 1" Box (Condulet) 1 27 min. 36 min. 2-1 la Box 1 27 min. 36 min. 2-1 4"x4"x8" Sox 1 27 min. 36 min. 2-1 1" Conduit 1 HOLD (Penetration) 6"x6"x4" Junction Box 1 HOLD CGCTB071 1" Conduit 1 27 min. 36 min. '2-1 1" Radial Bend 1 27 min. 36 min. 2-1 Cond.ait 1" Box (Condulet) 1 27 min. 36 min. 2-1 8"x11"X2'-8" Box 1 HOLD 013/026

6 Topical R: port 102 Rev. O Page 21 of 22 ENVELOPE TYPE NO. ACTUAL CABLE NEI TEST .NO. ELEMENTS RTG. RTG. QUALIFICATION CGPTB020 1.5" conduit 12 27 min. 36 min. 2-1 1.5" Radial Bend 12 27 min. 36 min. 2-1 Conduit 1.5" Box (condulet) 3 27 min. 36 min. 2-1 1.5" Conduit 1 27 min. 36 min. 2-1 (Penetration) 1.5" Junction Box 2 HOLD CGPTB029 1" Conduit 9 27 min. 36 min. 2 1" Radial Bend 7 27 min. 36 min. 2-1 Conduit. 1" Box (condulet) 1 27 min. 36 min. 2-1 6"x6" Junction Box 1 HOLD CGPTB030 1.5" Conduit 11 27 min. 36 min. 2-1 1.5" Radial Bend 7 27 min. 36 min. 2-1 Conduit 1.5" Box (Condulet) 2 27 min. 36 min. 2-1 12*x12" Junction Box 1 HOLD The above envelopes (14-25, 14-28, 62-93, 86-71, OGCTB017, OGCTB071, CGPTB020, CGPTB029 and CGPTB030) will be the subject of an exemption request from the requirement in Appendix R, Section III.G.2.c, for a one hour rated fire barrier in fire zone TB-FZ-11D. The heat transfer model which projects temperatures inside these raceways out to 60 minutes was compared with the cable insulation failure threshold temperatures for required safe shutdown circuits inside these raceways. For these raceways, the projected raceway temperatures at the duration listed are below the cable insulation failure threshold temperatures by more than 70 deg F; therefore these raceway barriers have a cable qualification rating as listed above. ENVELOPE TYPE No. ACTUAL CABLE NEI TEST NO. ELEMENTS RTG. RTG. QUALIFICATION CGPTB020 1.5" Flex. 2 INDETERMINATE N/A Conduit The 1.5" flexible conduit will be upgraded to provide an " actual" fire rating of one hou'r and will therefore meet the requirements of III.G.2.c. GPU Nuclear is currently planning on designing and installing the upgrade for these elements to conform to Texas Utilities Scheme 11-2 for a 1-1/2" airdrop. These tests were successful in establishing a 60 minute fire endurance rating. This constitutes GPU Nuclear's current design detail for the aforementioned upgrades. If other types of upgrades which provide an " actual" fire rating of 60 minutes are identified due to more industry testing, an alternative configuration could be used. Such changes will not be submitted for NRC approval but handled under existing licensing conditions; ie. 50.59 process. The commitment is to upgrade these envelopes to an " actual" fire rating of 60 minutes. 3.7 Turbine Building, switchgear Room, West End, Messanine (Fire Ione TB-FE-11C) ENVELOPE TYPE NO. ACTUAL CABLE NEI TEST NO. ELEMENTS RTG. RTG. QUALIFICATION 86-71 4" conduit 2 91 min. 102 min. 2-3 4" Radial Bend 4 91 min. 102 min. 2-3 Conduit 4" Conduit 4 HOLD (Penetration) 013/026

( Topical R: port 102 s Rev. 0 Page 22 of 22 The above envelope (86-71) will be the subject of an exemption request from the requirement in Appendix R, Section III.G.2.a, for a three hour rated fire barrier in fire zone TB-FZ-11C. Request an exemption from the requirement for an area wide automatic suppression system. The temperature inside these raceways at 102 minutes, when the test was terminated, was compared with the cable insulation failure threshold temperatures for required safe shutdown circuits inside these raceways. For these raceways, the projected raceway temperatures at the duration listed are below the cable insulation failure threshold temperatures by more than 70 deg F; therefore these raceway barriers have a cable qualification rating as listed above.

4.0 REFERENCES

4.1 NRC Generic Letter 86-10, supplement 1, Enclosure 1, " FIRE ENDURANCE TEST ACCEPTANCE CRITERIA FOR FIRE BARRIER SYSTEMS USED TO SEPARATE REDUNDANT SAFE SHUTDOWN TRAINS WITHIN THE SAME FIRE AREA", dated March 25,.1994. 4.2 10 CFR Part 50 Appendix R, " FIRE PRCTECTION PROGRAM FOR NUCLEAR POWER FACILITIES OPERATING PRIOR TO JANUARY 1, 1979". i 4.3 NEI Report No. 0784-00001-TR-02, Revision 1, "NEI APPLICATION GUIDE FOR EVALUATION OF THERMO-LAG 330 FIRE BARRIER SYSTEMS". 4.4 GPU Nuclear Dyster Creek Nuclear Generating Station Fire Hazards Analysis Report (FHAR) No. 990-1746, Revision 7. 4.5 NFPA 251 (ASTM E-119), " STANDARD FIRE TESTS OF BUILDING CONSTRUCTION AND MATERIALS". 4.6 Sandia Test Report SAND 90-0696 May,1991 "AN INVESTIGATION OF THE EFFECTS OF THERMAL AGING ON THE FIRE DAMAGEABILITY OF ELECTRICAL CABLES". 4.7 Gilbert Commonwealth Letter G/C/TMI-1CS/16503 Sept. 15,1988, J. Brendlen to J.W.Langenbach, "TSI DERATING CHECK". 4.8 NFPA Fire Protection Handbook, Seventeenth Edition. 4.9 EPRI TR 100370, April, 1992, " FIRE INDUCED VULNERABILITY EVALUATION". 4.10 EPRI TR 104030, draft January, 1994, " FIRE RISK ANALYSIS IMPLEMENTATION CUIDE". 4.11 NUMARC, Sept 1993, "THERMO-LAG 330-1 COMBUSTIBILITY EVALUATION METHODOLOGY PLANT SCREENING GUIDE". 4.12 EPRI Project 3385-05, June 1995, " METHODS FOR EVALUATION OF CABLE WRAP FIRE BARRIER PERFORMANCE". APPENDICES A. USING FIVE FIRE MODELLING TO APPROXIMATE EXPOSURE TIME-HISTORY FOR THERMO-LAG CONFIGURATIONS IN OC FIRE ZONE RB-FZ-1D. B. USING FIVE FIRE MODELLING TO APPROXIMATE EXPOSURE TIME-HISTORY FOR THERMO-LAG CONFIGURATIONS IN OC FIRE ZONE RB-FZ-1E. C. USING FIVE FIRE MODELLING TO APPROXIMATE EXPOSURE TIME-HISTORY FOR THERMO-LAG CONFIGURATIONS IN OC FIRE ZONE RB-FZ-1F2. D. USING FIVE FIRE MODELLING TO APPROXIMATE EXPOSURE TIME-HISTORY FOR THERMO-LAG CONFIGURATIONS IN OC FIRE ZONE TB-FZ-11D. E. USING FIVE FIRE MODELLING TO APPROXIMATE EXPOSURE TIME-HISTORY FOR THERMO-LAG CONFIGURATIONS IN OC FIRE ZONE TB-FZ-11C. 013/026

TA 102 .(. AlPPEDDIXA Using FIVE Fire Modeling to Approximate Exposure Time-History for Thermo-Lag Configurations in OC Fire Zone RB-FZ-1 D i Letter Report september 27, ins f Prepared by: GFUN Risk Analysis Group Using FIVE Fire Modeling to Approximate Exposure Time-History for Thermo-Lag Configurations

_____________.m._. l \\ l 1. PURPOSE The purpose of this report is to document the detailed fire modeling performed for Are scenarios invoMng Thermo Lag wrapped Are barriers, as identified during plant walkdowns of OC fire zone RB-FZ-1D. Thermo-Lag exposure time-Nstory results were obtained for each T-L ' configuration determined to be impacted by a fire source / combustible. 2. APPROACH The approach followed for modeling of fire scenarios is based on methodology detailed in the EPRI " Methods for Evaluation of Cable Wrap Fire Barrier Performance" document (Ref.1) which was based on fire modeling techniques developed for FNE (Ref. 2). Fire modeling was accomplished in two steps: J l

1. Screening walkdowns to eliminate from further consideration those fire ignition sources and potential Are scenarios that cannot develop into damaging fires.

2. Fire modeling of specific scenarios that were not screened out in the first stop.

For each Thermo-Lag configuration analyzed, one or more fire scenarios (including bounding scenarios) were postulated based on the following: a. information gathered during plant walkdowns (i.e., fixed or transient ignition source and Thermo-Lag target location, type and quantity of combustibles present in zone); b. information obtained from reviewing applicable sections in Appendix R documents for OC (Ref. 3); c. Information obtained from reviewing OC procedure 120.5, Rev. 5, " Control of Combustibles." (Ref. 4) d. Information provided by GPUN engineers (specializing in fire protection, risk assessment, mechanical). Thermo-Lag FNE analyses were performed using computerized Excel spreadsheets which duplicate FNE Worksheets 1,2, or 3 and are linked to tables containing the same information as in FNE Reference Tables. For each Thermo-Lag fire modeling scenario, results are presented as exposure time-temperature profiles. In all scenarios evaluated, time dependent temperatures were derived by incrementally adding the total heat content of the fuel into the Are ceiTvei^ur.Gnt hot gas layer. 3. FIRE MODEUNG ASSUMPTIONS The following Thermo-Lag fire modeling assumptions were made: 1. Worst-case fixed fire source scenarios were modeled for each Thermo Lag sub-1

i 1 configuration. The determination of which Axed fire souros might cause the worst-case 3ermo-Lag Are scenario was made during the plant Are area walkdown and 4 was based on minimum damage threshold heights and distances raimilatad for typical fixed tre sources (e.g., electrical cabinets, electrical motors, tube oil in pumps) found at Oyster Creek. 2. Worst-case transient fire source scenarios observed during walkdowns were also modeled. Worst-case transient fire soonarios considered were similar to the type and quantities of transient combustibles allowed by OC station procedure (Ref. 4). 3. To ensure conservative results,. worst case fire plume (or ceiling jet) scenarios and/or radiant Aux scenario were first modeled For T-L sub-configurations not in. the plume or colling Jet, hot gas layer Sre scenarios were evaluated. 4. Electrical cable inside cabinets, motor control centers, switchgear and cable trays was assumed to be non-qualified (i.e., non IEEE-383 type). Dis assumption is conservative since electrical cable purchased and installed at Oyster Creek in the past decade was IEEE-383 type. 5. The Fire Hazard Tool developed for the EPRI Cable Wrap Fire Barrier Tallored Collaboration assumes that the fire wigui,-ni temperature will follow the ASTM E-119 time-temperature curve as soon as the sig,,.nl hot gas layer (HGL) reaches.10007. When the HGL temperature reaches 10007, the plume (or ceiling jet) temperatures are assumed to follow that of the E-119 (see diaa na*Iart in Appendix B). 6. The Thermo-Lag wrap is assumed to bum by itself (i.e., even after the initial fire source is out of fuel) unless otherwise stated. Per, the EPRI Method (Ref.1, Appendix l-D), credit may be taken for a time lag before the Thermo-Lag ignites based on exposure temperature. 7. The maximum plume and ceiling jet temperature is assumed to be 16007, which corresponds to flame temperature. This asswrpLn is valid until hot gas layer temperature exceeds 16007, at which point the plume temperature is assumed to be equal to hot gas layer temperature. 8. Fire Zone RB-FZ-1D specific data: total area of 9,100 ft2 (Ref. 3 - Oyster Creek Fire Hazard Analysis Report, Rev. No. 8); fire zone ceiling height of 24 feet ( Oyster Creek General arrangement Drawings); ambient temperature of 1007. I 4. FIRE MODELING RESULTS ] 4.1 Fire Zone RB-FZ-1D / Fire zone RB FZ-1D is the Reactor Building 51' Elevation. There are three areas in this fire zone that contain the T-L wrap. They are designated as location Nos. 6,7, and 8 on the genera! arrangement drawing shown in Appendix D. 2 m.

Location No. 6 Four scenarios were evaluated for this T-L configuration. The Arst two scenarios model auto- - Ignition of norsqualified cable trays and their impact on the T-L wrapped conduits above. The third scenario evaluates a transient ignition source / combustible that was noted during the walk downs in the area, and the last scenario models the HGL contribution due to a core spray booster pump oil fire. Scenario #1. There are two vertical T-L wrapped conduit runs that start from Location No. 9 at Elevation 33' below and go up to elevation 51' and then curve along the wall. An l&C cable runs by the two vertical T-L conduits in this area, however, this is a qualified cable and auto-Ignition is not a concem. There are, however, three very loosely-packed cable trays in this area that run about 7 ft. below The T-L wrap. This was modeled here as one fully packed cable tray self-igniting that could impact the T-L wrap. Two ft. of a 24'-wide cable tray was assumed to initially self-Ignite and the T-L target was modeled 'in-plume" with the fire source located against the wall. This Are is assumed to propagate horizontally along the exposed cable tray with a flame speed of 10 feet per hour (per Ref.1, Appendix l-C). The heat release rate considered for this scenario (before the T-L wrap is ignited) is as follows:

1. Cable tray - 23.4 BTU /sec-ft2 (Ref. 2, FIVE Table 1E, cable #5 adjusted to consider Equation 2 on FIVE manual, page 10.4-10); since 4 ft2 are initially buming, the HRR for this source alone is 93.6 BTU /sec; this HRR will increase with time, as fire propagates horizontally along the tray (i.e., 6 minutes into the fire, the HRR for this source will be 141 BTU /s, which includes the HRR from 2 ft2 of additional cable tray);

I The total available heat (Otot) from this cable tray fire source is:

1. Cable tray - 4,000,000 BTUs [it assumes 10,000 BTU /lbm of cable insulation (per Ref. 6, page E2-1) x 10 linear feet x 40 lbm/ linear foot (engineering judgment for 100% Alied,24 i

inch wide cable tray)]; Results The analysis as shown in Table 1 indicates that the "in-plume" temperatures exceed 1000'F at t=24 minutes into this scenario, so the T-L wrap will be ignited and bum. However, the HGL temperature is only 112*F. Once the cable-trays' Are ignites the T-L wepped conduits, they will start to bum. T-L wrap HRR is assumed to be 8.8 BTU /sec-ft2 (per Ref. 5); the amount of Thermo Lag that could be involved in this Are scenario is assumed to be 12 linear feet (6 linear feet per conduit). The total surface area is 15.7 ft2 (condults are 4-inch in diameter with / 1/2 inch thick T-L wrap); the HRR is thus 138 BTU /sec. This HRR will be considered only for calculating the HGL temperature rise. The total available heat (Qtotal) from this amount of T-L fire source is 705,600 BTUs (it assumes 58,800 BTU / linear foot, from Ref. 5). A sample model 1 worksheet is provided in Appendix A, Figure A-1. ) 3 l .__,___m

Table 1. Temperaturo-Time History for Fire Scenario 1, Location No. 6. Time HMM to HRH to UIUI Temp. Temp. l i Target HGL (min.) (Stu/s) (Blu/s) (Blu) Target - (F) HGL - (F) ) O 94 94 0 571 100 5 141 141 Sutou ou 102 12 155 155 115440 795 104 15 235 235 203040 909 107 l 24 252 420 354240 1015 112 30 329 457 naJov 1121 117 i 35 we 514 tu(400 1221 123 42 423 561 909350 1317 130 l 45 470 505 1125240 1411 135 54 517 555 1354040 1554 145 50 554 702 151stou 1593 155 ) Scenario #2. There are three T-L wrapped conduit runs along the wall in this area with two i 24'-wide cable trays (1 ft. apart on top of each other) about 8 ft directly undemeath the T-L conduits. The cable trays are 6 ft. off the floor and are perpendicular to the wall. At the location where this scenario is postulated, this area communicates with a higher elevation j through a large opening but there is a 3.5-4' lip on the ceiling and the T-L ends at the ceiling of this lip. Therefore, this area was modeled as a 20' by 20' room with a 24' ceiling for this i scenario. This fire scenario modeled a fixed ignition source / combustible, a 2 foot wide cable tray with non IEEE-383 cable,6 feet off the room's floor. Since the cable is non-qualified, the cable fire can self-ignite (per Ref. 2, Reference Table 1.2, " Fire ignition Sources and Frequencies by Applicable Plant Locations) and cable would start to burn. The fire will then propagate (almost immediately) to a second cable tray, located 1 foot above. In addition to the vertical propagation, the fire is also assumed to propagate horizontally along the exposed cable tray with a flame speed of 10 feet per hour (per Ref.1, Appendix l-C). The targets are three 3-inch diameter conduits wrapped in 0.5 inch thick Thermo-Lag. The T-L wrapped conduits pass perpendicularly over the cable trays,8 feet above them. Due to the close distance between the cable trays and T-L wrapped conduits, the T-L will be ignited by the cable tray fire. Once this happens, additional heat is released into this fire zone. / / The heat release rates considered for this scenario are as follows (by fire source): first cable tray - 23.4 BTU /sec-ft2 (Ref. 2, FIVE Table 1E, cable #5 adjusted to consider 4 -- -- O

4 Equation 2 on FIVE manual, page 10.4-10); since 4 ft2 are initially burning, the HRR for this source alone is 93.6 BTU /sec; this HRR will increase with time, as Are propagates horizontally along the tray (i.e.,6 minutes into the Are, the HRR for this source will be 140.4 BTU /s, which includes the HRR from 2 ft2 of additional cable tray); second cable tray - 23.4 BTU /sec-ft2 (Ref. 2, FIVE Table 1E, cable #5 adjusted to consider Equation 2 on FIVE manual, page 10.4-10); just as with the first cable tray, the HRR for this source alone is 140.4 BTU /sec (since 6 ft2 is initially assumed to be burning); this HRR will increase with time, as Are propagates,%,6ia.ily along the tray (i.e.,6 minutes into the Are, the HRR for this source will be 187.2 BTU /s, which includes the HRR from 2 ft2 of additional cable tray); T-L wrap HRR is assumed to be 8.8 BTU /sec-ft2 (per Ref. 5); the amount of Thermo-Lag i that could be involved in this Are scenario is assumed to be 7 linear feet (visual observation) for a total surface area of 7.3 ft2; the HRR is thus 64 BTU /sec; this HRR wiu be considered only for calculating the HGL temperature rise. The total available heat (Otot) from each Are source is: first cable tray - 4,000,000 BTUs [it assumes 10,000 BTU /lbm of cable insulation (per Ref. = 6, page E2-1) x 10 linear feet x 40 lbm/ linear foot (engineering judgment for 100% filled,24 inch wide cable tray)); second cable tray - 4,000,000 BTUs [it assumes 10,000 BTU /lbm of cable insulation (per = Ref. 6, page E2-1) x 10 linear feet x 40 lbm/ linear foot (engineering judgment for 100% filled,24 inch wide cable tray)]; Thermo-Lag wrap - 333,200 BTUs (it assumes 47,600 BTU / linear foot, from Ref. 5). t Results The results are presented in Table 2 below. As indicated by the temperature-time histories in exceeds 1000*F at t=12 minutes, so the T-L wrap wiu be ignited and bum. However, i the table, the 'in-plume' temperature at the T-L targets 8 feet above the buming cable trays temperature is only 263*F. As indicated by the temperature-time histories in Table 2, post-1 flashover conditions (i.e., above 1000*F HGL temperature) occur at t=36 minutes and the - HGL temperatures will follow those of the ASTM E-119 from then on. Appendix B contains the basis for this assumption. A sample model worksheet is provided in Appendix A, Figure A-2. 5

] Table 2. Temperature-Time History for Fire Scenario 2, Location No. 6. Time MMM to HMM to ulVI Temp. Temp. Target. HGL (min.) (Stu/s) (Stu/s) (Blu) Target - (F) MGL - (F) 0 234 234 0 wuic 100 5 325 325 115050 953 151 12 422 455 293040 1212 253 18 515 550 501540 1493 405 24 510 574 744450 1500 512 30 704 755 luicuw50 1500 905 35 798 562 1331250 1500 1329 42 592 955 10/0440 1500 lauw-48 985 1050 N ' tis 1500 1452" 54 1080 1144 2455250 1500 1525" 50 1174 1238 2910950 1500 1550" " Temperatures follow those of AslM 15-119. Scenario #3. A wooden ladder was noted stored on the floor during walkdowns of the same area described above for Scenario #2 for this fire zone. This is modeled here as a transient ignition source / combustible. There is a 24'-wide cable tray four feet above "in-plume" of P'- fire which will be ignited due to close proximity to the ladder fire source. It is assumed that 2 ft. of this cable tray will be initially ignited and this fire is assumed to propagate horizontally along the exposed cable tray with a flame speed of 10 feet por hour (per Ref.1, Appendix l-C). The HGL temperatures were calculated here for the impact on the T-L in the area 14 ft. above the floor. The heat release rates considered for this scenario are as follows (by fire source): wooden ladder - 12 BTU /sec-ft2 (Ref. 2, FNE Table 2E); it is assumed that a total surface area of 10 ft2 are initially burning, the HRR for this source alone is 120 BTU /sec. cable tray - 23.4 BTU /sec-ft2 (Ref. 2 FIVE Table 1E, cable #5 adjusted to consider i Equation 2 on FNE manual, page 10.410); the HRR for this source is 93.6 ETU/sec (since / 4 ft2 is initially assumed to be burning); this HRR will increase with time, as fire propagates horizontally along the tray (i.e., 6 minutes into the fire, the HRR for this source will be 140.4 BTU /s, which includes the HRR from 2 ft2 of additional cable tray); 6

The total available heat (Otot) from each fire source is: s 1 i wooden ladder - 400,000 BTUs [it assumes the ladder weighs 50 lbm x 8,000 BTU /lbm l (per Ref. 6, Page E2-3)]; i cable tray - 4,000,000 BTUs [it assumes 10,000 BTU /lbm of cable insulation (per Ref. 6, i e page E2-1) x 10 linear feet x 40 lbm/ linear foot (engineering Mwe a for 100% filled,24 j inch wide cable tray)); i i i i Results i l 1 1 Th.e analysis indicates that as a result of the ladder Igniting, the "in-plume" temperatures at the cable tray 4 ft. above reaches 937"F at t = 6 minutes. Two feet of this cable tray is assumed to ignite at this time and propagate horizontally along the exposed cable tray. The HGL temperatures calculated for the T-L wrapped conduits 14 ft. above the floor in this area are j shown in Table 3 below. As can be seen, the HGL temperatures exceed 1000'F at the end of i this scenario (t = 60 minutes). This fire soonario is ended at 1 hour since the T-L modeled is assumed to be a 1-hour wrap. It should also be noted that these HGL temperatures are l bounded by Scenario # 2 above. A sample model worksheet is provided in Appendix A, l Figure A-3. The target temperature reflected on this worksheet is the temperature seen by the i cable tray 4 feet above the ladder and The T-L wrap (14 ft above) sees the HGL temperatures in this scenario. l Table 3. Temperature-Time History for Firs Scenario 3, Location No. 6. i e A I i - Time HRR to HRH to GTOT Temp. Temp. Target HGL j (min.) (Blu/s) (Btu /s) (Stu) Target - (F) HGL - (F) j 0 120 120 0 100 100 l 6 214 214 /tv40 129 129 l 12 261 261 171000 156 156 j 15 305 305 261550 213 213 24 355 355 409550 zrz ziz g l 30 402 402 554400 345 345 36 449 449 715040 435 435 42 496 496 594600 545 545 / 45 543 543 1090050 553 553 54 590 590 1302450 553 553 j 60 637 537 1531500 1056 1056 7

A I Scenario #4. The fixed ignition source modeled here is the core spray booster pump P ' 2A. This pump contains up to 2 gaHons of tube oil (Ref. 7). Total Q available is 310080 BTU ' 2 (from Ref. 6, page E2-1, assumed to be the same at OC), heat release rate 110 BTUs/sec-ft

• (from Ref. 2, Table 2E). The potential oil spread area around the pump was estimated at 5 ft. The closest thermo-Lag wrap conduits run along the north war about 20 ft. sway longitudinany from the pump and at the highest elevation 4 ft. below the coHing, Source (i.e.,

pump) was modeled as being 'in-conter'. l Resutts i i Th,ermo-Lag target is not damaged (i.e., temperature does not reach 1000 T) by HGL contributions, see Table 4. There just isnt enough energy in 2 gabons of oil. The maximum temperature at the T-L wrap was calculated to be 105T. A sample model worksheet is provided in Appendix A. Figure A-4. i l Table 4. Temperature-Time History for Fire Scenario 4, Location No. 6. l l l l Time HRR to HRR to QTOT Temp. Temp. Target HGL (min.) (Blu/s) (Btu /s) (Btu) Target - (F) HGL - (F) 0 550 550 0 100 100 6 550 550 198000 103 103 r 10 550 550 330000 105 105 11 0 0 330000 105 105 20 0 0 330000 105 105 30 0 0 330000 105 105 36 0 0 330000 105 105 42 0 0 330000 105 105 48 0 0 330000 105 105 54 0 0 330000 105 105 i 60 0 0 330000 105 105 / 8

Location No. 7 Scenario aP1. The bounding scenario for this T-L configuration run was identified as the eisetrical panel v50F as the Axed ignition source during wakdowns. This panel is 6 feet off the floor and this was assumed as the virtual surface of the Are. The T-L conduit runs horizontally along the wall 52 inches above this panel. The T-L wrap is approximately 3 inches in diameter 0.5 inches thick, and was modeled W of Are "against the wall". Results As shown in Table 5, the energy contribution from the panel as modeled ;,T,,T=" ^ M/ raises the "in-plume" temperature to 1600 Y st the target and is assumed to ignite the T-L wrap. The heat release rate assumed for this electrical cabinet is 400 BTUs/sec (containing non-qualified cable, Ref.1, Appendix l-B). A fire duration of 15 minutes was assumed for this electrical 2 cabinet. It is assumed that 2 feet of T-L wrap with a surface area of 1.6 ft (3 inches diameter) begins to bum. From NEl study on Thermo-Lag 330-1 Combustibility (Ref. 5 - Thermo-Lag 330-1 Combustibility, Attachment 1 and Table 1 of (ttachment 3), heat release rates and total BTU for the T-L wrap were obtained (8.8 BTU /sec-ft and 47,000 BTU / linear foot, respectively). The T-L heat release rate of 14 BTU /see was added to that of the electrical cabinet (400 BTU /sec) for HGL calculations. The virtual surface of the fire is at electrical cabinet 6 feet off the floor. The cabinet fire is exhausted at time t=15 minutes. At this time, the T-L temperature (in plume) is 1600*F. With the T-L wrap burning, the HGL temperature is only 108T. Ref.1, Appendix l-D indicates that "once Thermo-Lag is ignited, there is no test data to demonstrate how and when it will stop buming after the fire source is removed. Therefore, the Hazard tool assumes that once the fire source is removed, the Thermo-Lag in the plume (or ceiling jet) could be exposed to temperatures ranging from HGL (if the Thermo-Lag stops buming completely) to peak plume temperature (if the Thermo-Lag continues to bum)". It was very conservatively assumed here that the Thermo-Lag will continue to bum at peak plume temperature for the remainder of the scenario. A sample model worksheet is provided as Appendix A Figure A 5. i 9

l .s l Table 5. Temperaturo-Time History for Fire Scenario 1, Location No. 7. Time HMM to HHH to UIUI Temp. Temp. Target HGL (min.) (Stu/s) (Stu/s) (Stu) Target - (F) HGL - (F) 0 400 400 0 1500 100 5 400 414 149040 1500 103 12 400 414 295050 1500 105 l 15 400 414 Jtzovu .1500 105 15 0 14 JfJ440 1500 105 25 0 14 351000 1500 105-35 0 14 390240 1500 105 42 0 14 3M 1500 109 45 0 14 4vuseu 1500 109 54 0 14 405350 1500 109 50 0 14 410400 1500 109 Location No. 8 Scenario #1. At this location, there is one vertical T-L conduit run that is by the equipment hatch. An open top cable tray (TGXR 2006) runs vertically right next to the T-L conduit 6 inches apart at its closest point. There are no other fixed Ignition sources nearby except the RBCCW pump (P 5-2) which is more than 20 feet away. Results j The analysis, shown in Figure A-R1, used FNE Worksheet #3 (Ref. 2) to model this scenario. It was assumed that 2 t}of 12-inch wide cable tray (TGXR 2006) will self-ignite. A heat release rate of 23.4 BTU /sec-ft ( Ref.1 FNE Table 1E, cable # 5 adjusted to consider Equation 2 on FNE manual, page 10.410) was assumed for the buming cable tray. The total HRR for this source is 47 BTU /sec. The results indicate that if the cable tray is WM/(. 0.8 f of the T-L wrap, the T-L wrap reaches a critical damage flux of 2.2 BTU /sec-ft this is the f minimum ignition radiant flux for T-L, per Ref. 5, page A18). This ar#.:. is extremely / 1 conservative because it uses a point source radiant host flux model (for the 2 feet of cable tray) and it assumes 40 percent of the total heat release rate of accidental fires is radiative, with the remainder released convectively. 10 --es-- mv+w r 6.--e'

1 5. FIRE MODEUNG

SUMMARY

Fire modeling was performed for three locations containing Thermo-Lag fire barrier in this Are zone. Fire scenarios involving fixed and transient sources were analyzed. The ar. die;e results are presented as temperature-time histories in Tables 1 through 5. These results are used in the following section to d;^;;'+ Total Heat Load and then convert to equivalent ASTM E-119 8xPosures. 6. APPROXIMATING EXPOSURE TIME-HISTORY OF THERMO-LAG SUB-CONFIGURATIONS - THE TOTAL HEAT LOAD CONCEPT Calculation of an exposure time-temperature profile for a speclSc T-L Sutx:onfiguration is based on the methodology presented in Reference 1. The time -temperature profiles are used to develop the total heat load (Reference 1, Appendix lil-A) for each Sre exposure. The total heat load concept holds that the area under the incidenthost Sux vs.. time curve, or total heat load, can be used as a measure of fire severity. Comparing the total heat load of a fire scenario exposure to the total heat load of a test exposure (ASTM E-119) abows one to predict the response of a Thermo-Lag barrier to the acenario exposure. The total heat load per unit area of fire barrier surface is equal to the area under the incident heat Sux-time curve, corrected for convection effects as described in Reference 1, Appendix til-A. 6.1 Total Heat Load Results Location Scenario Description Peak Total Heat Load No.(GA Temperature Equivalent E-

DWG, at Target 119 Exposure App. D)

(T) Duration (min.) 6 1 cable Tray 1593 <G 6 2 Cable Tray 1500 >50 6 3 Wooden Ladder 1066 <18 6 4 CS Booster Pump 105 <10 7 1 Electrical Panel 1500 > 50 8 1 Gable tray Radiant >0.8 Ft. 4 Exposure Appendix C contains tables with Total Heat Load values r*Wmf for lhermo-Lag scenarios. / 11 ,_=

1 4 7. REFERENCES 1. EPRI Report " Methods for Evaluation of Cable Wrap Fire Barrier Performance", July, 1995 2. EPRI TR-100370, " Fire-induced Vulnerability Evaluation," April 1992. .3. GPUN, " Fire Hazards Analysis Report OC," Doc. No. 990-1746, Rev. 8, May 24, 1995. 4. GPUN - OC Procedure No.120.5 (Revision 5), May 20,1993. 5. ~ Thermo-Lag 330-1 Combustibility Study, NEl,1994. 6.. GPUN - TMI-1 Procedure No.1035 (Revision 24), August 23,1994. 7. Personal communication with Malcom Gonzales of GPUN, in charge of plant oil program. 0 re-he mm / 12

Appendix A Scenarios FIVE Worksheets ^* I i ? o/ i 4 \\ l

1 Figure A-1 Temperature calculation at time min 6 i Maximum ambient temperature F 100 Floor Area ft2 9100 1 Location of Target Plume I 2 Height of Target above Fire Source R 7.00 3 Height fmm Fire Source to Collin0, H A 12.00 4 Ratio of Target Height / Ceiling Hei0ht 0.583 5 Lon0ltudinal Distance from Fire Source to .A NA Target. L 6 Lon0ltudinal Distance to Helpht Ratio, IJH NA ) 7 Enclosure Width, W R NA 8 Height to Width Ratio, H/W NA 9 Peak Fire intensity Stuts 141 10 Fire Location Factor 2 11 Effective Fire intensity Stuts 282 12 Plume Temperature Rise at Target F 571 13 Temp Rise Factorat Target 1.00 14 Temperature Rise at Target F 571 15 Temperature at Target F 672 16 Temperature in HGL F 102 17 Qtot to HGL Blu 50760 18 Estimated Heat Loss Frac, tion 0.94 19 Calculated Onet Stu 3046 20 Calculated Enclosure Volume, V R3 100200 21 Calculated OnetN Qnet/ft3 0.03 i 22 HGL Temperature increase F 2 / 4 ~ w,

Figure A-2 Temperature calculation at time min 6 Maximum ambient temperature F 100 FloorArea ft2 400 1 Location of Target Plume 2 Height of Target above Fire Source R 8.00 3 Het0ht from Fire Source to Ceiling, H R 18.00 4 Ratio of Target Height / Ceiling Height 0.444 5 Longitudinal Distance from Fire Source to ft NA Target. L 6 Lon0ltudinal Distance to Hei0ht Ratio, I./H NA 7 Enciosure Width, W ft NA 8 Height to Width Ratio, H/W NA 9 Peak Fire intensity Stu/s 328 10 Fire Location Factor 2 11 Effective Fire intensity Blu/s 656 12 Plume Temperature Rise at Target F 802 13 Temp Rise Factor at Target 1.00 14 Temperature Rise at Target F 802 15 Temperature at Target F 983 16 Temperature in HGL F 181 17 Qtot to HGL Btu 118080 18 Estimated Heat Loss Fraction 0.94 19 Calculated Qnet Btu 7085 20 Calculated Enclosure Volume, V ft3 7200 21 Calculated OnetN Onet/ft3 0.98 22 HGL Temperature increase F 61 l / G . - - ~ - +

..-..- =.-. - -- - . _....- _ -.-. _._.~ _ - Figure A-3 Temperature calculation at time min 6 Maximum ambient temocrature F 100 Floor Area A2 400 1 Location of Target Plume 2 Height of Target above Fire Source A 4.00 3 Height from Fire Source to Ceiling. H A 24.00 4 Ratio of Target Height / Ceiling Hei0ht 0.167 5 Longitudinal Distance from Fire Source to A NA Target L 6 Longitudinal Distance to Hei0ht Ratio, LJH NA 7 Enclosure Width, W A NA j 8 Height to Width Ratio, H/W NA 9 Peak Fire intensity Stuts. 214 10 Fire Location Factor 1 11 Effective Fire intensity Stu/s 214 12 Plume Temperature Rise at Target F 1207 13 Temp Rise Factor at Target 1.00 \\ 14 Temperature Rise at Target F 1207 15 Temperature at Target F 1338 16 Temperature in HGL F 129 17 Otot to HGL Blu 77040 ) 18 Estimated Heat Loss Fraction 0.94 19 Calculated Onet Stu 4622 20 Calculated Enclosure Volume, V ft3 9600 21 Calculated OnetN Onet/ft3 0.48 22 HGL Temperature increase F 29 i l 1 i

---

r--

w, r .s

Figure A-4 Yemperature "hMM st tune min 6 ' Maximum amtnent temperature F 100 Floor Area 22 9100 1 Location of Target Hot Gas Layer 2 Height of Target above Fire Souros R 20.00 3 Height from Fire Souros to Caging, H R 24.00 4 Ratio of Target HeighWCeWng Height 0.833 5 Longitudinal Distance from Fire Source to R NA Target, L 6 Longitudinal Distanos to Height Ratio. L H NA 7 Enclosure Walth, W R NA 8 Height to Width Ratio, HMf NA 3 Peak Fire intensky StWs 550 10 Fire Location Factor 1 11 Effective Fire intensky StWs $50 I2 Plume Temperature Rise at Target F 155 13 Temp Rise Factorat Target 0.00 14 Temperature Rise at Target F 0 15 Temperature at Target F 103 16 Temperature in HGL F 103 17 Otot to HGL Stu 196000 16 Estimated Heat Loss FractiM 0.94 19 Calculated Onet Stu 11880 20 Calculated Enclosure Volume, V R3 218400 21 Calculated OnetN Onet/ft3 0.05 22 HGL Temperature increase F 3 I

... ~ _. -...... - -... ..-,.-. -.---...-. ~ F10ure A-5 Temperature calcuishon at time min 6 Maximum emuent temperature F 100 FloorArea A2 0100 1 Lo5 tion of Target Plume 2 F/ sight of Target stove Fire Source A 4.33 3 Height from Fire Source to Collin0, H A 18.00 4 'Astic of Target Height / Ceiling Height 0.241 5 Lon04tudinal Distance from Fire Source to R NA Target, L 6 Longitudinal Distance to Height Ratio, IJH NA 7 Enclosure Width, W R NA 8 Height to Width Ratio, HNV NA 9 Peak Fire intensity Bluts 400 10 Fire Location Factor 2 11 Effective Fire intensity Stuts 800 12 Plume Temperature Rise at Target F 1600 13 Temp Rise Factor at Target 1.00 14 Temperature Rise at Target F 1600 15 Temperature atTarget F 1600 16 Temperature in HGL F 103 17 Qtot to HGL Blu 140040 16 Estimated Heat Loss Fraction 0.94 19 Calculated Onet Blu 8042 7 20 Calculated Enclosure Volume, V R3 163800 21 Calculated OnetN Onet/R3 0.05 22 HGL Temperature increase F 3 /

?. Flours A-R1, scenario 1. Loc. No. 8 i WORKSHEET 3: RADIANT EXPOSURE SCENARIOS 1 Critical Radiant Flux to the Target StWs#t2 2.2 <-User input (Representative value = 1) 2 Peak Fire intensity StWs 47 <-User input l 3 Radiant Fraction of Heat Release 0.4 <-User input (Representathe value = 0.4) 4 Radiant Heat Release Rate StWs 18.8 l 1 5 Critical Radiant Flux Distance ft 0.8 1 ff Die exposure Are le located u4dn 9de detence $u$cated Irl beK 5) of to tuget, artioel sonehene een enour. omaise wie ange, armoel eenemens em not W ser se esenerlo neuter moreinerstart. 1 l l 1 i i 1 ) 1 O e /

i Appmunx s Basis for Following ASTM E-119 Time-Temperature Relationship When T(HOL) is 1000 Y i 1 l i /

r. j i TRANSm0N TO ASTM E-119 TIME-HISTORY IN THE HOT GAS LAYER ISSUE: The Fire Hazard Tool currently under development in the EPRI Cable Wrap Fire Barrier Tailored Collaboration assumes that the Are compartment temperature will follow the ASTM E-119 time-temperature relationship as soon as the compartment hot gas layer reaches 10007. BASIS: The Handbook (Bg[, pg 6-75) states that "Flashover of an enclosure is likely to occur if the temperature of the upper gas layer reaches -g-; verir?; 10007 (800'C). The EPRI assumption of 10007 is therefore conservative. As stated near the top of page 667 of the Handbook, the intensity of the Are will be somewhat lower when the walls and ceiling absorb significant amounts of energy rather-than when act as insulation or radiation barriers. Since power plant compartments are typically concrete barriers which absorb significant amounts of heat, the intensity of Are will tend to be less severe than Aros in compartments with insulated barriers. ASTM E 119 has been the accepted standard for testing Are barrier systems since the early 1900's. The standard, which was developed from actual fire tests, has been used to evaluate the fire endurance capabilities of fire barriers since it was developed. The time-temperature relationship represented by the curve represents a fully-involved Are i compartment. According to Fred Mowrer of the University of Maryland, plume and ceiling effects disappear in a fully involved fire compartment as the plume /colling jet gas are no longer more buoyant than surrounding gases. Use of the ASTM E 119 time-temperature relationship to represent a fully Irwolved power plant compartment Are is therefore consistent with industry practice. Assuming that that occurs when the hot gas layer reaches 100(TF is more conservative than the 11007 noted in the Handbook. 'j

Reference:

NFPA Fire Protection Handbook, Seventeenth Edition, Section 6/ Chapter 6.

L i Appendix C Total Heat Load Results 4 //

L Figure C-1 Total Heat Load Scenario 1, Loc.No.6 TarDat Hot Gas Layer ASTM E-119 Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min) (F) (Btu /s.sf) (1000 Btu /sf) (1000Btutsf) (1000Btutsf) 0 671 1.2 0.0 0.00 0.00 6 672 1.2 0.44 0.11 0.41 12 795 1.7 0.97 0.19 1.65 18 909 2.2 1.67 0.27 3.56 24 1018 2.8 2.56 0.36 5.93 30 1121 3.6 3.71 0.45 8.64 36 1221 4.4 5.14 0.54 11.59 42 1317 5.4 6.91 0.64 14.73 48 1411 6.5 9.07 0.74 18.03 54 1584 9.2 11.90 0.85 21.49 60 1593 0.3 15.23 0.97 25.15 O 4 e .O /

I4 j FGure C-2 Total Heat Load i Scenario 2, Loc.No.6 TarDat Hot Gas Layer ASTM E 119 3 l Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load l (min) (F) (Btu /s-sf) (1000 Btu /sf) (1000 Btu /sf) (1000 Btu /sf) 1 0 902 2.1 0.0 0.00 0.00 6 963 2.5 0.83 0.13 0.41 5 12 1212 4.3 2.06 0.27 1.35 18 1493 7.8 4.24 0.44 3.56 j 24 1600 9.5 7.34 0.72 5.91 30 1600 9.5 10.75 1.29 6.64 36 1600 9.5 14.16 2.67 11.59 l 42 1600 9.5 17.57 4.64 14.73 48 1600 9.5 20.97 7.31 18.03 i 54 1600 9.5 24.38 10.10 21.49 60 1600 9.5 27.79 13.12 25.15 i I e L.

c Qure C-3 i Total Heat Load Scenario 3, Loc.No.6 TarDat Hot Gas Layer ASTM E-119 Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min) (F) (Btu /s-sf) (1000 Btu /sf) (1000 Btu /sf) (1000Blu/sf) 0 100 0.4 0.0 0.00 0.00 6 129 0.3 0.12 0.12 0.41 12 168 0.4 0.23 0.23 1.65 18 213 0.3 0.35 0.35 3.56 24 272 0.5 0.49 0.49 5.93 30 345 0.5 0.66 0.66 8.64 36 435 0.6 0.87 0.87 11.59 42 545 0.9 1.14 1.14 14.73 48 683 1.3 1.52 1.52 18.03 54 853 1.9 2.10 2.10 21.49 60 1066 3.2 3.02 3.02 25.15 l i l l l l \\

C Figure C-4 Total Heat Load Scenario 4, Loc. No. 6 TarDet Hot Gas Layer ASTM E-119 Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min) (F) (Blu/s-sf) (1000 Btu /sf) (1000Blu/sf) (1000 Btu /sf) 0 100 0.4 0.0 0.00 0.00 6 103 0.2 0.11 0.11 0.41 10 105 0.2 0.16 0.16 1.24 11 105 0.2 0.18 0.18 1.52 20 105 0.2 0.30 0.30 4.57 30 105 0.2 0.44 0.44 8.93 ~ 36 105 0.2 0.53 0.53 11.88 42 105 0.2 0.61 0.61 15.02 48 105 0.2 0.70 0.70 18.32 54 105 0.2 0.78 0.78 21.77 60 105 0.2 0.87 0.87 25.44 i I I l /

s Fi9ure C-5 Total Heat Load Scenario

1. Loc. No.7 Target Hot Gas Layer ASTM E-119 Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min)

(F) (Blu/s-s0 (1000 Btu /s0 (10008tu/s0 (1000 Btu /s0 0 1600 9.5 0.0 0.00 0.00 6 1600 9.5 3.41 0.11 0.41 12 1600 9.5 6.82 0.19 1.65 15 1600 9.5 8.52 0.23 2.59 l 16 1600 9.5 9.09 0.25 2.94 25 1600 9.5 14.20 0.38 6.53 36 1600 9.5 20.46 0.54 11.73 42 1600 9.5 23.86 0.62 14.87 48 1600 9.5 27.27 0.71 18.17 54 1600 9.5 30.68 0.80 21.63 60 1600 9.5 34.09 0.88 25.29 1 /

i Appendix D General Arrangement Drawings O m i i l l o /

e, 7 l-o s 4 g 4 -F/KE zoWe.' 4 "pt.asrr gzF4Z-/d .,t.- ,e. - ,e.,- , e. G W r- " 7 ~3 m-7 / v- ]LOCAttoW M wr,, // i_.;- ll ,_*.m.-. - -E6 N=? aw. n.rm [t h ~ ~~' 3-h Mf'~ jt..g s wE ~ ~l,-. ~~--w- .,,1 < b.<, ai /. A 3 f<- jE, N% d k[$,',., ~' %\\ b'N s h 2 A, ndtbe, N o e -==. 3 tr'sW_9d G e s w4 .,Jy ?, %w u 9 g 1. ( \\ p - a I q =:.. -s E._\\ l4 4y i'FA b., 7_--- 1 A. ~" ,m s , e y._.,,,,._ 9 7 E-1 h- -p::=: t

a i

y g%. L ..- ~. .J." .~ _A - __ q u3 a -A -t. J ' w% % M. = :. =,. v.i.h.WE5%p~a~-/T[ 7 RMa M, f nn ..~ m. n z q' Q H i:arfJi=:di\\ fa' p_ N v an / -'. j A_ . / =W.:: A l : O i\\l w,=ic Y % c, =. " EhJ / N.NC - + p.). a-=""* J a CP rn (LocAfiobl /~M% // l'.\\ N/_ !!L " '- s. Q. 7 , I ( 5BR'ss t,-rtrv ~~ 1 pu s,- r O 0 W i E r i,, " w - - " &QW ' 4;1rpedili;'-O r .s g g*- M-j me.m =_. w% u \\ x j me, a

---

~*e

'M d.* 5 PE*" L-Yh (,. [~p3 O

== .a fu Eb m l1 tI g.- .c,, g A. g M SE* 3 ._Q> N.- JIEM AllEMe AulI4,1d4.4C m -1 4.: a= tt ry = $ 9 a I laatf ~ ptaw rtry '. ~ .D iS gi I 7 i I t i

3

)

TR lo2. APPENDIX 8 Using FIVE Fire Modeling to Approximate Exposure Time-History for Thermo-l.ag Configurations in OC Fire Zone RB- ~ FZ-1 E Letter Report september 27, sees Prepared by: GPUN Risk Analysis Group Using FIVE Fire Modeling to Approximate Exposure Time History for Thermo-Lag Configurations / i

1 1. PURPOSE The purpose of this report is to document the detailed fire modeling performed for Are scenarios involving Thermo-Lag wrapped Are barriers, as identined during plant walkdowns of OC Bro zone RB-FZ-1E. Thermo-Lag exposure time-history results were obtained for each T-L con 6guration determined to be impacted by a fire source / combustible. 2. APPROACH l The approach followed for modeling of fire scenarios is based on methodology detailed in the EPRI " Methods for Evaluation of Cable Wrap Fire Barrier Performance" document (Ref.1) which was based on Are modeling techniques developed for FIVE (Ref. 2). Fire modeling was accomplished in two steps:

1. Screening walkdowns to eliminate from further consideration those fire ignition j

sources and potential fire scenarios that cannot develop into damaging fires. l

2. Fire modeling of specific scenarios that were not screened out in the Srst step.

For each Thermo-Lag configuration analyzed, one or more fire scenarios (including bounding scenarios) were postulated based on the following: a. information gathered during plant walkdowns (i.e., Axed or transient ignition source and Thermo-Lag target location, type and quantity of combustibles present in zone); b. Information obtained from reviewing applicable sections in Appendix R documents for OC (Ref. 3); c. Information obtained from reviewing OC procedure 120.5, Rev. 5, " Control of Combustibles." (Ref. 4) d. Information provided by GPUN engineers (spec.;e;a;,,g in fire protection, risk assessment, mechanical). Thermo-Lag FIVE analyses were performed using computerized Excel spreadsheets which duplicate FIVE Worksheets 1,2, er 3 and are linked to tables containing the same information as in FIVE Reference Tables. For each Thermo-Lag Are modeling scenario, results are. presented as exposure time-temperature proflies. In all scenarios evaluated, time dependent temperatures were derived by incrementally adding the total heat content of the fuel into the fire coiiip.d,,,ent hot gas layer. 3. FIRE MODELING ASSUMPTIONS The following Thermo-Lag fire modeling assumptions were made: 1. Worst-case fixed fire source scenarios were modeled for each Thermo-Lag sub-configuration. The determination of which fixed fire source might cause the worst-case Thermo-Lag Are scenario was made during the plant Bra area walkdown and 1

was based on minimum damage threshold heights and distances calculated for typical fixed fire sources (e.g., electrical cabinets, electrical motors, lube oil in pumps) found at Oyster Creek. 2. Worst-case transient fire source scenarios observed during walkdowns were also modeled. Worst-case transient fire scenarios considered were similar to the type and quantities of transient combustibles allowed by OC station procedure (Ref. 4), 3. To ensure conservative results, worst-case fire plume (or ceiling jet) scenarios and/or radiant flux scenario were first modeled. For T-L sutx:onfigurations not in the plume or ceiling jet, hot gas layer fire scenarios were evaluated. 4. Electrical cable inside cabinets, motor control centers, switchgear and cable trays. was assumed to be non-qualified. (i.e., non IEEE-383 type). This ass' mption is u conservative since electrical cable purchased and installed at Oyster Creek in the past decade was IEEE-383 type. 5. The Fire Hazard Tool developed for the EPRI Cable Wrap Fire Barrier Tailored Collaboration assumes that the fire corrpartment temperature will follow the ASTM E-119 time-tempemture curve as soon as the compe,L,,Gnt hot gas layer (HGL) reaches 1000T. When the HGL temperature reaches 10007, the plume (or ceiling jet) temperatures are assumed to follow that of the E-119 (see discussion in Appendix B). 6. The Thermo-Lag wrap is assumed to bum by itself (i.e., even after the initial fire source is out of fuel) unless otherwise stated. Per, the EPRI Method.(Ref.1, Appendix l-D), credit may be taken for a time lag before the Thermo-Lag ignites. based on exposure temperature. 7. The maximum plume and ceiling jet temperature is assumed to be 16007, which corresponds to flame temperature. This assumption is valid until hot gas layer temperature exceeds 16007, at which point the plume temperature is assumed to be equal to hot gas layer temperature. 8. Fire Zone RB FZ-1E specific data: i total area of 12,140 ft2 (Ref. 3 - Oyster Creek Fire Hazard Analysis Report, Rev. No. 8); fire zone ceiling height of 25 feet ( Oyster Creek General arrangement Drawings); ambient temperature of 100'F. t 4. FIRE MODELING RESULTS 4.1 Fire Zone RB-FZ-1E Fire zone RB FZ 1E is the Reactor Building 23' Elevation. Fire modeling was performed for three areas on Elevation 23'-6" and two areas on Elevation 33'-5" in this fire zone that contain the T-L wrap. They are designated as location Nos. 10,11,12,9, and 15 on the general arrangement drawings shown in Appendix D. Location No.10 Two scenarios were modeled for this T-L configuration. There are two vertical runs of T-L 1 2 I

wrapped conduits about 4 fL long at elevation 12 fL above the floor at this location. One conduit is 11/2" with 1/2" thick T-L wrap and the other is 3" with 1/2" thick T-L wrap. The first. scenario models a transient ignition source / combustible and the other evaluates the impact of scenario #1 for location #11 on the T-L configuration here. Scenario #1. This fire scenario modeled a transient ignition souros/ combustible, a 5 gallon container wist.;i,ir,g lube oil. Per Oyster Creek Procedure

• Control of Combustibles,"

Procedure No.120.5, Rev. 4 (Ref. 4), combustible liquids could be transported throughout the plant in an approved closed metal container (but not noosssarily a safety-can). For this fire scenario, it was postulated that 1 quart of lube oil is spilled out of a can on the1loor and is not cleaned up. The oil is later Ignited by a welder working nearby or another transient ignition source and begins to bum. The oil fire could potentially impact a T-L wrap 12. feet above. The heat release rate for the lube oil is 110 Btu /s-ft 2 (from Ref. 2, Table 2E). The oil could spread unconfined over an area of 13.5 ft 3 (54 ft.8/ gal from Ref. 2, Table 3E). De heat release rate assumed for this fire scenario is 1,485 Btu /s (110 Btu /s-ft. x 13.5 ft.a). The heat content of one quart of oil is (17,111 Btu /lb.

• 60 lb/1 ft 3.
• 1 ft.3/30 quarts) 34,222 Btu (from Ref. 2, Table 2E). The oil fire duration will be less than one minute, but one minute will be used for conservatism. The virtual surface of the fire is at the floor level. De location factor for this source is one.

Results The short oil fire does not cause an exposure "in-plume" temperature high enough to ignite the Thermo-Lag. The fire is of a short duration, with very slight hot gas layer effects. A sample model worksheet is provided as Appendix A, Figure A-1. Table 1. Temperature Time History for Fire Scenario 1, Location No.10. Time HRR to HRR to QTOT Temp. Temp. Target HGL (min.) (Blu/s) (Btu /s) (Btu) Target - (F) HGL - (F) 0 1485 1485 0 804 100 1 1485 1485 89100 805 101 2 0 0 89100 101 1 01 10 0 0 89100 101 101 15 0 0 89100 101 101 25 0 0 89100 101 101 30 0 0 89100 101 101 35 0 0 89100 101 101 40 0 0 89100 101 101 50 0 0 89100 101 101 60 0 0 89100 101 1 01 3 a ,,,s

0 '8 1 Scenario #2. This scenario evaluates the impact of HGL produced by the self ignition of stack of four cable trays described in scenario #1 for location #11 on the T-L wrap in this location # 10. Betidt This fire zone is pretty open and location nos.10 and 11 freely communicate with each other, therefore, the HGL temperatures calculated for scenario #1 for location #11 is also assumed to apply here. It should be noted that the bottom cable tray, the auto-ignition source / combustible for that scenario, is at 12 ft. below the ceiling and the T-L wrap here is at about 9 ft. below the ceiling (top of the T-L wrapped conduits), therefore, the target is in the HGL region of scenario #1 for location #11. The results are presented in Table 2. j Table 2. Temperature-Time History for Fire Scenario'2, Location No.10. Time HHH to HMM to UlUI Temp. Temp. Target HGL (min.) (Btu /s) (Stu/s) (Btu) Target - (F) HGL - (F) 0 1047 1047 0 100 100 5 1234 1234 444240 111 111 12 1421 1421 955500 124 124 15 1505 1505 1534550 135 135 24 1795 1795 2150500 155 155 30 1952 1952 2594400 174 174 36 2159 2159 30/s240 195 195 42 2355 2355 4 m 00 221 221 45 2543 2543 5435550 245 245 54 2730 xtau 6421550 xtv utw 50 2917 2917 7471500 313 313 / 4

Location No,11 ) Three scenarios were medeled for this T-L configuration. Three T-L wrapped 1"-conduits in this area run along the ceiling (CGCR3021, CGPR3019, & CNXR2050, see B&R Drawing Nos. 1303 & 1317). There are four 24"-wide cable trays stacked on top of each other that run direc0y underneath this T-L configuration. In the north view, there is a 24"-wide cable tray that runs about 4-inches apart from the T-L wrapped penetration box on the dryweg at about 8 fL below the ceiling.1his was found to be the most limiting case from au the four T-L wrapped penetration boxes in this area (Penetration Nos. 8,9,18, & 19). There is also a core spray booster pump on the floor at this elevation that is 15 ft. away longitudinaHy from the T-L wrapped conduits on the ceiling. l l Scenario #1. The fixed ignition source considered here, is the self-ignition of stack of four cable trays (Nos. 23,25,24, & 26). They are stacked one on top of each other about 1 ft. apart with the closest one (cable tray # 23) being about 9 ft. directly below the T-L wrap conduit runs on the ceiling. All the cable trays are 24 inches wide and cable trays # 24 and # 26 (the bottom two trays) have light loads. This scenario models self-ignition of 2 ft. of cable tray # 26 initiapy which win then ignits 4 ft. of cable tray # 24,6 ft. of cable tray # 25, and 8 ft of cable tray # 23. It is conservatively assumed that all the cable trays ignition occur instantaneously for r= Won simpitfication. In addition to the vertical propagation, the fire is also assumed to propagate horizontally along the exposed cable tray with a flame speed of 10 feet per hour (per Ref.1, Appendix l-C). The total amount of cable tray available for this fire scenario is 10 feet per tray (for the first hour of fire scenario). The T-L target was assumed "in-plume" with the fire ignition source modeled as "in-center". The heat release rates considered for this scenario before the T-L wrap is ignited are as follow (by fire source):

1. first cable tray (# 26) - 23.4 BTU /sec-ft2 (Ref. 2, FIVE Table 1E, cable #5 adjusted to consider Equation 2 on FIVE manual, page 10.4-10); since 4 ft2 are initially burning, the propagates horizontally along the tray (i.e.,6 minutes into the fire, the HR HRR for this source alone is 93.6 BTU /sec; this HRR win increase with time, as fire will be 140.4 BTU /s, which includes the HRR from 2 ft2 of additional cable tray);

5

I

2. second cable tray (# 24) - 23.4 BTU /sec-ft2 (Ref. 2, FWE Table 1E, cable #5 adjusted to consider Equation 2 on FNE manual, page 10.4-10); just as with the first cable tray, the -

HRR for this source alone is 187 BTU /sec (since 8 ft2 are initiaby burning); this HRR will increase with time, as fire propagates horizontally along the tray (i.e., 6 minutes into the fire, the HRR for this source wul be 234 BTU /s, which includes the HRR from 2 ft2 of additional cable tray); Sc third cable tray (# 25) - 23.4 BTU /sec-ft2 (Ref. 2, FNE Table 1E, cable #5 adjusted to consider Equation 2 on FWE manual, page 10.410); just as with the second cable tray, the HRR for this source alone is 281 BTU /sec (since 12 ft2 are initially buming); this HRR will increase with time, as fire propagates tituw' "; along the tray (i.e.,6 minutes into the fire, the HRR for this source win be 328 BTU /s, which includes the HRR from 2 ft2 of additional cable tray);

4. forth cable tray (# 23) - 23.4 BTU /sec-ft2 (Ref. 2, FNE Table 1E, cable #5 adjusted to consider Equation 2 on FNE manual, page 10.4-10); just as with the second cable tray, the HRR for this source alone is 374 BTU /sec (since 16 ft2 are initiaNy buming); this HRR will increase with time, as fire propagates horizontaHy along the tray (i.e., 6 minutes into the fire, the HRR for this source win be 421 BTU /s, which includes the HRR from 2 ft2 of additional cable tray);

The total available heat (Otot) from each fire source is:

1. first cable tray (# 26) -2,000,000 BTUs [it assumes 10,000 BTU /lbm of cable lasulation (per Ref. 6, page E2-1) x 10 linear feet x 20 lbm/ linear foot (engineering judgment for 50%

filled,24 inch wide cable tray));

2. second cable tray (# 24) -2,000,000 BTUs [it assumes 10,000 STU/lbm of cable insulation (per Ref. 6, page E2-1) x 10 linear feet x 20 lbm/ linear foot (engineering judgment for 50%

filled,24 inch wide cable tray));

3. third cable tray (# 25) --4,000,000 BTUs [itassumes 10,000 BTU / ibm of cable insulation (per Ref.

6, page E2-1) x 10 linear feet x 40 lbm/ linear foot (engineering judgment for 100%

filled, 24 inch wide cable tray));

4. forth cable tray (# 23) -4,000,000 BTUs [It assumes 10,000 BTU /lbm of cable insulation

/ (per Ref. 6, page E2-1) x 10 linear foot x 40 lbm/ linear foot (engineering judgment for 100% filled,24 inch wide cable tray)); 6

1 e i 4 i Once the cable-trays' fire ignites the T-L wrapped conduits, they will start to bum. T-L wrap j HRR is assumed to be 8.8 BTU /sec-ft2 (per Ref. 5); the amount of Thermo-Lag that could be-l initially involved in this fire scenario is assumed to be 24 linear feet (8 linear feet per conduit l run that is directly above cable tray # 23). The total surface area is 12.8 ft2 (conduits are 1-j inch in diameter with 1/2 inch thick T-L wrap); the HRR is thus 111 BTU /sec. The impact of I horizontal propagation of cable tray fire on the T-L wrap conduits is ignored here due to j insignificant heat contribution from the T-L wrap as compared to the cable trays. The total j 'available heat (Qtotal) from this amount of T-L lire source is 571,200 BTUs (it assumes 23,800 BTU / linear foot, from Ref. 5). i l Resutts i i The results are presented in Table 3 below. As irAasted by the temperature-time histories in the table, the 'in-plume

• temperature at the T-L targets 9 feet above the burning cable trays reaches 1000*F immediately, so the T-L wrap will be ignited and bum. However, the HGL i

temperature is only 313*F after one hour into this scenario. A sample model worksheet is j. provided in Appendix A, Figure A-2. j 1 i i l l Table 3. Temperature-Time History for Fire Scenario 1, Location No.11. { Time HRR to HMM to Ulul Temp. Temp. Target HGL (min.) (Blu/s) (5tu/s) (Blu) Target - (F) HGL - (F) l O 935 1047 0 1043 100 } 5 1123 1234 444240 1054 111 12 1310 1421 955500 1159 124 l 15 1497 1505 1534550 1251 135 d 24 1554 1795 2150550 1391 155 30 1871 1952 2594400 1500 174 2 35 2055 2159 3070240 1500 195 l 42 2245 2355 4K N 1500 221 1 45 2432 2543 5435550 1500 245 / 54 2519 Z(Ju 6421550 1500 xtw / 50 2005 2917 7471500 1500 313 3 l 2 ~ 7 i w w-

Scenario #2. This scenario evaluates the radiant exposure from self-ignition of the 24*-wide cable tray running about 4 inches apart from the T-L wrapped penetration box on the drywell. Results ) The analysis, shown in Appendix A, Figure A-R1, uses FNE Worksheet #3 (Ref. 2). It was assumed thft 2 ft of 24-inch wide cable tray will self-ignite. A heat release rate of 23.4 BTU /sec-ft (per Ref. 2, FNE Table 1E, cable #5 adjusted to consider Equation 2 on FNE manual, page 10.4-10) was assumed for the burning cable tray. The total HRR for this source is 93.6 BTU /sec. The results indicate that if the cable tray is locate (at 1.2 ft or less of the T-L i wrap, the T-L wrap reaches a critical damage flux of 2.2 BTU /sec-ft (this is the minimum ignition radiant flux for T-L, per Ref. 5, page A1-8). This analysis is extremely conservative because it uses a point source radiant heat flux model (for the 2 feet of cable tray) and.it assumes 40 percent of the total heat release rate of accidental fires is radiative, with the remainder released convectively. l Scenario #3. The fixed ignition source modeled here is the core spray booster pump P 28. This pump contains up to 2 gallons of lube oil (Ref. 7). Total Q available is 310000 BTU 2 (from Ref. 6, page E2-1, assumod to be the same at OC), heat release rate 110 BTUs/sec-ft 2 (from Ref. 2, Table 2E). The potential oil spread area around the pump was estimated at 4 ft during the plant walkdown. The Thermo-Lag wrap conduits are located on the ceiling 15 feet away longitudinally from the pump and in the " Celling Jet" sublayer. Source (i.e., pump) was modeled as being 'in-center'. Ambient temperature assumed was 100 T. Results / Thermo-Lag target is not damaged (i.e., temperature does not reach 1000 T) by Colling Jet effects or Ceiling Jet /HGL contribution.There just isn't enough energy in 2 gallons of all. The maximum temperature at the T-L wrap was calculated to be 135T. Temperatwo-time histories are shown in Table 4. A sample model worksheet is provided in Appendix A, Fi0ure A-3. 8

4 6 Table 4. Temperaturo-Time History for Fire Scenario 3, Location No.11. Time HHH to HMM to UlUl Temp. Temp. Target HGL (min.) (Blu/s) (Blu/s) (Btu) Target - (F) HGL - (F) 0 440 440 0 131 100 5 440 440 laavuu 132 102 10 440 44D 254000 134 103 12 440 440 31ocuv 135 1D4 13 0 0 31ocuu 104 104 15 0 0 31ouuu 104 104 20 0 0 315500 104 104 30 0 0 315800 104 104 40 0 0 310c00 104 104 50 0 0 31ouuu 104 104 60 0 0 315800 104 104 Location No.12 There is a 1-inch T-L wrapped conduit (CNXR2051) and===MaM T-L wrapped penetration box (penetration 57) in this area (see B&R drawing Nos. E1303 & E1317). About five feet of a half full 24"- cable tray runs directly undemeath the penetration box, vertically on the drywell, placing the box right in the plume of such cable fire. Auto-ignition of this cable was modeled as one scenario for this area. Also, this area is very open and freely communicates with other areas in this fire zone, therefore, the HGL temperatures calculated for scenario #1 for location

1. 11 is also applicable here. Finally, a transient ignition source / combustible was considered for this location to be placed on the floor against the drywell undemeath the T-L wrapped conduit.

Scenario #1. This scenario models auto-ignition of 5 fL of a half full 24" wide cable tray against the wall directly undemeath the T-L wrapped penetration box. The penetration box was estimated to be 17 ft. above the floor. A heat release rate of 23.4 BTU /sec-ft" (Ref. 2, FIVE Table 1E, cable #5 adjusted to consider Equation 2 on FIVE manual, page 10.4-10); 2 was used for this cable tray. Since 5 ft are initially assumed to be buming for the "in-plume" calculations, the HRR for this source is 117 BTU /sec. at the beginning of the scenarlo. The total available heat from this source is 4,000,000 BTUs [it assumes 10,000 BTU /lbm of cable tray (per Ref. 6, page E2-1) x 20 linear feet x 20 lbm/ linear foot (engineering judgment for 50% filled, 24 inch wide cable tray)]. The target was assumed W" for this scenario. I 9 -e-e

Bitidif The er.0,C as shown in Table 5 below indicate that the "in-plume" energy contribution is able to immediately ignite the T-L wrap and the temperatures reach 1600'F at the target. Once the cable-tray's fire ignites the T-L wrapped penetration box, it will start to bum. T-L wrap HRR is assumed to be 8.8 BTU /sec-ft2 with a heat load of 36,750 BTU /ft2 for 1/2" thick T-L wrap (per Ref. 5). The total surface area of the T-L wrap that would be in the plume of such fire is approximated at 2.5 ft2.1he HRR is thus 22 BTU /sec., and the T-L wrap should last for 70 minutes ( for a total heat load of 91875 BTUs). The "HGL" temperatures were calculated for this scenario considering that the HRR for the buming cable tray increases with time, as fire propagates horizontally along the tray (i.e., 6 minutes into the fire, the HRR for this source will be 140.4 BTU /s, which includes the HRR from 1 ft2 of additional cable tray). A sample model worksheet is provided as Appendix A, Figure A 4. Table 5. Temperature-Time History for Fire Scenarlo 1, Location No.12. Time HHH to HMM to WIUI Temp. Temp. Target HGL (min.) (Blu/s) (Blu/s) (Blu) Target - (F) HUL - (F) 0 117 139 0 1500 100 ) 5 117 152 55320 1500 102 12 117 155 123 eau 1500 105 18 117 209 200520 1500 107 24 117 233 254400 1500 110 30 117 255 stoS50 1500 114 36 117 ztw 4truuu 1500 115 42 117 303 555050 1500 122 48 117 325 703440 1500 125 54 117 350 529440 1500 131 50 117 J/a 95arzu 1500 135 9 l 10

Scenario #2. This scenario evaluates the impact of HGL produced by self ignition of stack of four cable trays described in scenario #1 for location #11 on the T-L wrap in this location # 12. Results This fire zone is pretty open and location nos.11 and 12 freely communicate with each other, therefore, the HGL temperatures calculated for scenario #1 for location #11 is also assumed to apply here. It should be noted that the bottom cable tray, the auto 4gnition source / combustible for that scenario, is at 12 ft. below the ceiling and the T-L wrap here is at about 8 ft. below the ceiling, therefore, the target is in the HGL region of scenario #1 for location #11. The results are presented in Table 6 below. Table 6. Temperature-Time History for Fire Scenarlo 2, Location No.12. '^ Time HRR to HMM to ului lemp. Temp. Target HGL (min.) (Stu/s) (Btu /s) (Stu) Target - (F) MGL - (F) 0 1047 1047 0 100 100 ~ 5 1234 1234 444240 111 111 12 1421 1421 955500 124 124 15 1605 1505 1534550 135 135 24-1795 1795 2150550 155 155 30 1952 1952 2594400 174 174 35 2159 2159 30/n240 195 195 42 2355 2355 4"'"JO 221 221 45 2543 2543 5435550 245 245 54 ztso ztau 5421550 ztw ztw 50 2917 2917 7471500 313 313 I / Scenario #3. The ignition source / combustible considered in this scenario is maintenanos refuse being left in a plastic container on the floor against the drywell, placing the T-L wrap directly "in-plume" above. It was assumed that 50 lbm of maintenance trash (rags, paper, etc.) were left in the plastic container, for a total Q avaliable of 800,000 BTUs (400,000 BTUs for 11

4 3 l 1 I the trash, and 400,000 BTUs for the plastic can, Ref. 6, page E2-1). The heat release rate is i 138 BTUs/s (Ref.1. Appendix l-E). i i L j Results As shown in Table 7, the "in-plume" temperatures (ir.cluding the HGL energy contribution) for the T-L wrap target do not exceed 284*F by the time this scenario for the transient fuel source is ended. Therefore, the T-L wrap is not damaged. A sample model worksheet is provided in i Appendix A, Figure A-5. i' j Table 7. Temperature-Time History for Fire Scenarlo 3, Location No.12. i j Time HMM to HMM to ulul Temp. Temp. l Target HGL ~ (min.) (Blu/s) (Btu /s) (Btu) Target - (F) HGL - (F) l 0 138 138 0 zu 100 j 6 138 138 49ecu z(e 101 i 12 138 138 99380 ztw 101 j i 18 138 138 149040 ztw 102 ~ ~ 24 138 138 195tzu 250 103 30 138 138-248400 281 103 36 138 138 298080 281 104 42 138 138 347780 282 105 i 48 138 138 397440 282 105 54 138 138 447120 283 108 80 138 138 498800 284 107 ) i i o Location No. 9 ~ There are two vertical T-L wrapped conduits by the stairwell at this location. During the walkdowns, no fixed combustibles were noted within a 15'-radius. Possibility of HGL entrapment was also discarded during walkdowns. There is an extension cord within 2-3 ft. of these T-L conduit runs but its impact is deemed insignificant. There are no pumps nearby and it is very unlikely that any combt:stibles will be stored in this area. No scenarios were developed for this location. i 12

l Location No.15 This area "the TIP room" was not accessible during the walkdowns due to high radiation. The information gathered here was obtained from Plant personnel and drawings. There are two T-L wrapped penetrations at this location (Penetration Nos. 44 & 54, see drawing no. B&R E1317). The dimensions for this room were estimated at 20 lW4 fL by 20 ft. by 15 ft. height (from general arrangement drawings). Since there are no maintenance actMties being done in this area (High Radiation area) during power operation, only one transient ignition source was modeled for this room. Scenario #1. A wooden ladder was noted stored in this area from the pictures obtained prior to the application of the thermo-lag fire barrier for this location (pictures from Susan Hopson from OC Plant Rad. Engineering). This is modeled here as a transient ignition source / combustible. The heat release rate considered for this source (wooden ladder) is 12 BTU /sec-ft2 (Ref. 2, FIVE Table 2E); it is assumed that a total surface area of 10 ft2 is burning, the total HRR for this source is, therefore,120 BTU /sec. The total available heat (Otot) is 400,000 BTUs [it assumes the ladder weighs 50 lbm x 8,000 BTU /lbm (per Ref. 6, j Page E2-3)]. The ignition source was modeled conservatively "against the wall". The T-L wrapped penetrations were assumed "in-plume" located about half-way below the ceiling at 7.5 ft. above the fire source. Re:u3s The analysis as shown Table 8 indicates that as a result of the ladder igniting, the 'in-plume" temperatures in the room reach 863T at one hour into the scenario. Therefore, the T-L wrap 4 is not damaged. A sample model worksheet is provided in Appendix A, Figure A 6. j / 13 1

Table 8. Temperature-Time History for Fire Scenario 1, Location No.15. Tlrne H MM to MHH to UTUI Temp. Temp. Target MGL (min.) (Blu/s) (Blu/s) (Blu) Target - (F) HGL - (F) 0 120 120 0 557 100 5 120 120 4dicuu 5tEi! 125 12 120 120 55400 508-151 18 120 120 liswouu oso 175 24 120 120 Inspuu 554 acur 30 120 120 215000 593 235 36 120 120 Mum 724 acet s 42 120 120 302400 707 300 (,_48 120 120 345500 791 334 54 120 120 525 359 60 120 12c 432000 553 405 i 5. FIRE MODELING

SUMMARY

Fire modeling was performed for three locations containing Thermo-lag wrap in Oyster Creek fire zone RB-FZ-1E. The analysis results are presented as temperature-time history profiles in Tables 1 through 8. These results are used in the following section to develop Total Hoet Load and then convert to equivalent ASTM E-119 exposures. l l l I 6. APPROXIMATING EXPOSURE TIME-HISTORY OF THERMO-LAG SUS. CONFIGURATIONS - THE TOTAL HEAT LOAD CONCEPT Calculation of an exposure time-temperature profile for a specific T-L Sub oonfiguration is based on the methodology presented in Reference 1. The time -temperature proflies are used to develop the total heat load (Reference 1, Appendix lil-A) for each fire exposure. The total heat load concept holds that the area under the incident heat flux vs. timi curve, or total f heat load, can be used as a measure of fire severity. Co. Tips ing the total heat load of a fire 14 I l 1

acenario exposure to the total heat load of a test exposure (ASTM E-119) allows one to predict the response of a Thermo-Lag barrier to the scenario exposure. The total heat load. por unit area of fire barriw surface is equal to the area under the incident heat flux-time curve, corrected for convection effects as described in Reference 1, Appendix lil A. 6.1 Total Heat Load Results LC-: ^'--. Scenario C::::- 9:. Peak Total Heat Load No.(GA Tamperature Equivalent E-

DWG, at Target 119 Exposure App. D)

(V) Duration (mirL) 10 1 011 Transient 505 <10 10 2 Gable Tray 313 <12 11 1 Gable tray 1500 >50 11 2 Gable Tray Radiant > 1.2 FL EWL"o 11 3 CS Booster Pump 135 <10 12 1 Gable Tray 1500 >50 12 2 Gable Tray 313 <12 12 3 Maintenance Refuse 284 <18 9 None 15 1 Wooden Ladder 553 <24 Appendix C contains tables with Total Heat Load values rehd=* art forlhermo-Lag scenarios. 7. REFERENCES 1. EPRI Report " Methods for Evaluation of Cable Wrap Fire Barrier Performance", July, 1995 2. EPRI TR 100370, " Fire-induced Vulnerability tvaluation," April 1992. 3. GPUN,

• Fire Hazards Analysis Report - 00,* Doc. No. 990-1746, Rev. 8, May 24, 1995.

4. GPUN - OC Procedure No.120.5 (Revision 5), May M,1993. 5. Thermo-Lag 3301 Combustibility Study, NEl,1994. 6. GPUN - TMI-1 Procedure No.1035 (Revision 24), August 23,1994. 7. Personal communication with Malcom Gonzales of GPUN, in charge of plant oil program. s l 15 il.

i 1 l I l l l i Omo Appendix A Sample Scenario Worksheets 1 i e 1 4 e

== e-e4A 4,m 1 -.- ~ l l O I i i l l l o

Figure A-1 l +, Temocrature calculation at time min 1 Maximum ambient temperature F 100 Floor Area ft2 12140 l 1 Location of Target Plume 2 Height of Target above Fire Source A 12.00 3 Heigte from Fire Souros to Colling, H A 25.00 4 Ratio of Target Height / Ceiling Height 0.480 5 Longitudinal Distance from Fire Source to A NA TarDat. L 6 Longitudinal Distanos to Height Ratio, L/H NA 7 Enclosure Width, W R NA 8 Height to Width Ratio, HNV NA 9 Peak Fire intensity Blu/s 1485 10 Fire Location Factor 1 11 Effective Fire intensity Stu/s 1485 12 Plume Temperature Rise at Target F 704 13 Temp Rise Factor at Target 1.00 14 Temperature Rise at Target F 704 15 Temperature at Target F SOS f 16 Temperature in HOL F 1 01 17 Qtot to HGL Blu 89100 18 Estimated Heat Loss Fraction 0.94 19 Calculated Onet Stu 5346 l 20 Calculated Enclosure Volume, V ft3 303500 21 Calculated OnetN Onet/ft3 0.02 j 22 HGL Temperature increase F q o / 4 G

Figure A-2 Temperature calcunstson at time min 6 Maximum ambient temperature F 100 Floor Area ft2 12140 1 Location of Target Plume 2 Height of Target above Fire source ft 9.00 3 Height from Fire Source to Ceiling, H ft 12.00 4 Ratio of Target Hei0ht/ Ceiling Hei0ht 0.750 5 Longitudinal Distance from Fire Source to ft NA TarDat.L 6 Longitudinal Distance to Hei0ht Ratio, L/H NA 7 Enclosure Wxith, W ft NA 8 Height to Width Ratio, H/W NA 9 Peak Fire intensity Stu/s 1123 10 Fire Locat6on Factor 1 11 Effective Fire intensity Stuts 1123 12 Plume Temperature Rise at Target F 943 13 Temp Rise Factor at Target 1.00 14 Temperature Rise at Target F 943 15 Temperature at Target F 1964 16 Temperature in HGL F 111 i 17 Otot to HGL Blu ' 444240 18 Estimated Heat Loss Fraction 0.94 19 Calculated Qnet Stu 26654 20 Calculated Enclosure Volume, V ft3 145680 21 Calculated OnetN Onet/ft3 0.18 22 HGL Temperature increase F / 4 v

FQure A-R1, Scenario 2. Loc. No.11 WORKSHEET 3: RADIANT EXPOSURE SCENARIOS 1 Critical Radiant Flux to the Target Stu/s/ft2 2.2 <-User input (Representative value = 1) 2 Peak Fire intensity Stu/s 94 <-User input 3 Radiant Fram of Heat Release 0.4 <-User input (Rew..LL; value = 0.4) 4-Radiant Heat Release Rate Stu/s 37.6 5 Critical Radiant Flux Distance A 1.2 N the exposure are le loomeed ashh this dennes pussened m see s) et en serpet, erstmal conmone can ocaw. omakes shie ange, oracal eensuono are not insensed for the noenerio under consideremon. 9 // n w .=.

Figure A-3 Temperature calculation at time min 5 Maximum ambient temperature F 100 Floor Area ft2 12140 1 Location of Target Ceiling Jet 2 Height of Target above Fire Source ft 25.00 3 Height from Fire Source to Celling. H ft 25.00 t000 4 Ratio of Target Height / Ceiling Height 5 Longitudinal Distance from Fire Source to ft 15 Target, L 6 Longitudinal Distance to Height Ratio, tJH 0.60 7 Enclosure Wdth, W ft 25 8 Height to Wdth Ratio, H/W 1.00 9 Peak Fire intensity Stu/s 440 10 Fire Location Factor 1 11 Effective Fire intensity Stu/s 440 12 Plume Temperature Rise at Target F 92 13 Temp Rise Factor at Target 0.34 14 Temperature Rise at Target F 31 15 Temperature at Target F 132 16 Temperature in HGL F 102 17 Otof to HGL Btu 132000 18 Estimated Heat Loss Fraction 0.94 19 Calculated Onet Stu 7920 20 Calculated Enclosure Volume, V ft3 303500 21 Calculated OnetN Qnet/ft3 0.03 22 HGL Temperature increase F 2 /

I Figure A-4 Temperature calculation at time min 6 Maximum ambient temperature F 100 Floor Area ft2 12140 1 Location of Target Plume 2 Height of Target above Fire Source fr. 0.10 3 Height from Fire Source to Ceiling, H ft 8.00 4 Ratio of Target Height / Ceiling Height 0.013 5 Longitudinal Distance from Fire Source to 0 NA Target, L 6 Longitudinal Distance to Height Ratio, L/H NA 7 Enclosure Width, W ft NA 8 Height to Wxfth Ratio, H/W NA 9 Peak Fire Intensity Stu/s 117 10 Fire Location Factor 2 11 Effective Fire intensity Stu/s 234 12 Plume Temperature Rise at Target F 1600 13 Temp Rise Factor at Target 1.00 14 Temperature Rise at Target F 1600 15 Temperature at Target F 1600 16 Temperature in HGL F 102 17 Qtot to HGL Blu 58320 18 Estimated Heat Loss Fraction 0.94 19 Calculated Qnet Stu 3499 1 20 Calculated Enclosure Volume, V ft3 97120 21 Calculated OnetN Onet/fl3 0.04 22 HGL Temperature Increase F 2 /

.. _ _. _.. _ - ~. _.. _ Figure A-5 A Temperature calculation et time min 6 Maximum ambient temperature F 100 Floor Area ft2 12140 1 Location of Target Plume 2 HeiOM of Target above Fire Source A 14.00 3 HeiOM from Fire Source to Colling, H A 22.00 4 Ratio of Target Height / Ceiling Hei0ht 0.636 5 Lon0ltudinal Distance from Fire Source to ft NA Target, L 6 Lon08tudinal Distance to Hei0ht Ratio, UH NA l 7 Enclosure Width, W A NA 8 Hei0ht to Width Ratio, H/W NA 9 Peak Fire intensity 'Blu/s 138 ) 10 Fire Location Factor 2 11 Effective Fire intensity Stu/s 276 12 Plume Temperature Rise at Target F 177 13 Temp Rise Factor at Target 1.00 14 Temperature Rise at Target F 177 15 Temperature at Target F 278 16 Temperature in HOL F 101 17 Otot to HGL Btu 49680 18 Estimated Heat Loss Fraction 0.94 19 Calculated Onet Stu 2981 20 Calculated Enclosure Volume, V R3 267080 21 Calculated Onet/V Qnet/ft3 0.01 22 HGL Temperature increase F 3 m- -

FGute A4 1 Temperature calculation at time min 6 Maximum ambient temperature F 100 Floor Area - R2 415 1 Location of Target Plume 2 Height of Target above Fire Source R 7.50 3 Height from Fire Source to Ceiling, H ft 15.00 4 Ratio of Target Hei0ht/ Ceiling Height 0.500 5 Longitudinal Distance from Fire Souste to R NA Target L 6 Lon04tudinal Distance to Height Ratio, L/H NA l 7 Enclosure Width, W R NA 8 Height to Width Ratio, H/W NA 9 Peak Fire intensity Stu/s 120 10 Fire Location Factor 2 11 Effective Fire intensity Stuts 240 12 Plume Temperature Rise at Target F 457 13 Temp Rise Factor at Target 1.00 14 Temperature Rise at Target F 457 15 Temperature at Target F 882 16 Temperature in HGL F 125 17 Qtot to HGL Blu 43200 18 Estimated Heat Loss Fraction 0.94 19 Calculated Qnet Stu 2592 20 Calculated Enclosure Volume, V ft3 6225 21 Calculated OnetN Qnot/ft3 0.42 22 HGL Temperature increase F 25 o /

A Appendix B Basis for Following ASTM E-119 Time-Temperature Relationship When T(HOL) is 1000 7 i l / I l

TRANSITION TO ASTM E-119 TIME-HISTORY IN THE HOT GAS 1.AYER ISSUE: 'The Fire Hazard Tool currently under development in the EPRI Cable Wrap Fire Estrier Tailored Collaboration assumes that the fire ceir.pe i...ent temperature will follow the ASTM E-119 time-temperature relationship as soon as the compartment hot gas layer reaches 10007. ] BASIS: 1 1 The Handbook (BaL, pg. 6-75) states that "Flashover of an enclosure is likely to occur if the i temperature of the upper gas layer reaches approximately 10007 (600*C). The EPRI assumption of 10007 is therefore conservative. As stated near the top of page 667 of the Handbook, the intensity of the fire will be somewhat lower when the walls and ceiling absorb significant amounts of energy rather than when act as insulation or radiation barriers. I Since power plant compartments are typically concrete barriers which absorb significant amounts of heat, the intensity of fire will tend to be less severs than fires in compartments with insulated barriers. l ASTM E-119 has been the accepted standard for testing fire barrist r/ stems since the early 1900's. The standard, which was developed from actual fire tests, has been used to evaluate the fire endurance capabilities of fire barriers since it was developed. The time-temperature relationship represented by the curve represents a fully-involved fire compartment. According to Fred Mowrer of the University of Maryland, plume and ceiling effects disappear in a fully involved fire compartment as the plume / ceiling jet gas are no longer more buoyant than surrounding gases. Use of the ASTM E-119 time-temperature relationship to represent a fully involved power plant compartment fire is therefore consistent with industry practice. Assuming that that occurs when the hot gas layer reaches 10007 is more conservative than the 11007 noted in the Handbook. /

Reference:

NFPA Fire Protection Handbook, Seventeenth Edition, Section 8/ Chapter 6.

W Appendix C Total Heat Load Results 9 l

Figure C-1 a 7otal Heat Load kenario 1 Loc.No.10 TarDat Hot Gas Layer ASTM E-119 1 Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (mno (F) (Blu/s sf) (1000 Btu /sf) (1000 Btu /sf) (1000Blu/sf) 0 804 1.6 0.0 0.00 0.00 1 805 1.7 0.10 0.02 0.00 2 101 0.2 0.16 0.03 0.02 10 101 0.2 0.26 0.14 1.23 15 101 0.2 0.33 0.21 2.80 20 101 0.2 0.40 0.28 4.66 25 101 0.2 0.47 0.35 6.75 30 101 0.2 0.54 0.41 9.04 40 101 0.2 0.67 0.55 14.10 50 101 0.2 0.81 0.69 19.73 60 101 0.2 0.95 0.82 25.84 / .n

F2creC-2 A Total Heat Load Scenario

1. Loc. No.11 Target Hot Gas Layer ASTM E-119 Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min)

(F) (Blu/s-s0 (1000Blu/s0 (1000 Btu /s0 (1000 Btu /s0 0 1043 3.0 0.0 0.00 0.00 6 1054 3.1 1.10 0.11 0.41 12 1169 4.0 2.37 0.20 1.65 18 1281 5.1 4.01 0.30 3.56 24 1391 6.4 6.07 0.41 5.93 30 1500 7.9 8.63 0.54 8.64 36 1600 9.5 11.76 0.68 11.59 42 1600 9.5 15.17 0.82 14.73 48 1600 9.5 18.57 0.95 18.03 54 1600 9.5 21.98 1.11 21.49 60 1600 9.5 25.39 1.27 25.15 i i 1 / \\

FGure C-3 1 } Total Heat Load Scenario

3. Loc. No.11 Target Hot Gas Layer ASTM E 119 Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min)

(F) (Blu/s-sf) (1000Blu/sf) (1000Blu/sf) (1000 Btu /sf) 0 131 0.3 0.0 0.00 0.00 5 132 0.3 0.08 0.00 0.34 10 134 0.3 0.17 0.16 1.37 12 135 0.3 0.20 0.19 1.93 l 13 104 0.2 0.22 0.20 2.21 15 104 0.2 0.25 0.23 2.84 j 20 104 0.2 0.32 0.30 4.71 i 30 104 0.2 0.46 0.44 9.07 40 104 0.2 0.60 0.58 14.13 50 104 0.2 0.73 0.71 19.76 60 104 0.2 0.87 0.85 25.87 a i 4 4 h ll e a er

1 i Fc' ure C-4 Total Heat Load l Scenario 1, Loc.No.12 J Target Hot Gas Layer ASTM E-119 Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min) (F) (Stu/s.sf) (1000Blu/sf) (1000 Btu /sf) (1000Blu/sf) i 0 1600 9.5 0.0 0.00 0.00 1 6 1600 9.5 3.41 0.11 0.41 1 12 1600 9.5 6.82 0.19 1.65 18 1600 9.5 10.23 0.28 3.56 24 1600 9.5 13.64 0.36 5.93 30 1600 9.5 17.05 0.45 8.64 36 1600 9.5 20.46 0.54 11.59 1 42 1600 9.5 23.86 0.63 14.73 j 48 1600 9.5 27.27 0.73 18.03 54 1600 9.5 30.68 0.83 21.49 i 60 1600 9.5 34.09 0.93 25.15 i 4 J i I i l l i 4 l ) i i i i

)

4 i / i 4

FEure C-5 b. Total Heat Load Scenario

3. Loc. No.12 Target Hot Gas Layer ASTM E-119 Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min)

(F) (Stu/s-sf) (10008tu/sf) (10008tu/sf) (1000 Btu /sf) 0 277 0.5 0.0 0.00 0.00 6 278 0.5 0.17 0.11 0.41 12 279 0.5 0.34 0.19 1.65 18 279 0.5 0.51 0.27 3.56 24 280 0.5 0.68 0.35 5.93 30 281 0.5 0.85 0.44 8.64 36 281 0.5 1.02 0.52 11.59 42 282 0.5 1.19 0.60 14.73 48 282 0.5 1.37 0.69 18.03 54 283 0.5 1.54 0.77 21.49 1 60 284 0.5 1.71 0.86 25.15 I e l i l i ] /

F%re C4 t L Total Heat Load Scenario 1, Loc. No.15 TarDet Hot Gas Layer ASTM E-119 Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min) (F) (Btuts-sf) (1000 Btu /sf) (1000Blu/sf) (1000Blu/sf) 0 557 0.9 0.0 0.00 0.00 6 582 1.0 0.34 0.11 0.41 12 608 1.0 0.70 0.22 1.65 18 635 1.1 1.07 0.35 3.56 ~ 24 664 1.2 1.48 0.47 5.93 30 693 1.3 1.93 0.59 8.64 36 724 1.4 2.42 0.74 11.59 42 757 1.5 2.93 0.92 14.73 48 791 1.7 3.51 1.10 18.03 54 826 1.8 4.13 1.29 21.49 60 863 2.0 4.80 1.49 25.15 l /

d Appendix D General Arrangement Drawings i /

8 i I t 1 A _-. PLA8tf 8 ~ Y .i.. - }iii-.- a a - a mm ew w:::: _ b ) y ..J L l'4 g T E-E':' '( h 'L{ O jE

8

. i l B s l h Il $ t-{N Q)l] S I; $i' ss 4 fgr i i 0 G l.hr..l N;, l J, ;.3,,' li c e ' k r p p gt 1 Log,grion e(" i i

=._t f

p t f , i g --p Q 10 s[,, l.e P%7..._a,,, yu f,[ d $ - F .., 9,., . ~ T'k j a 7 g p- { ~'g: p "!.b.23 Q= o y .y _J 1 a- _n

i m, gywoma

.n. s l ama.as g) bk : hN I J _$_ [r; ;@ .**-1 p,, I k b l! %.i

• l 4-ECL Ir'm.L

,ne, \\ i /.b.i _ 'm M:: ju 4L,,,, (qg -r. u= z.:. g -~ 4 -- p E=--:

I 3:3

'55 ::

=-

!.s r.ar 4 ii /', &a t Ei )j m

~ '

g ._ t !E h[_"NDk1 . \\\\ N (k k~ g s 't LocAttoi N% F-S M- .h a 11 x.4 L fy,ts-2 C'U =w-w s =.. E ; -.- 3 yac}. p, r g ~ l c Zi-w N pd -,. t

77M, i s

( n x S_ ri m .u m-Mdx M.=

• • lfS[. =z,.=

l \\ \\ l l d I ~..'.F..,.'"".* z t y. 7J Ai l, h WE (C31.)RM. 70 g 3 r <

m'

m xt u_

iw s m d

=

4 \\ M._. 'ir t ~~ 1 =:. M ,i

==e, .30' PL AN ELEV 2 9'. E' P /AE 2.c W E EIT-82. -N- .a. E I I l t e

WUN 6 _.;;.n.. g --~j m s.- e:r - ~.//, = mnf* .-,_a, ~.m W /..> - K l --I. 1een'h 9 ...a <, -d p-- -- eTp fNE h2 ~$1T - s t w n., n. ag .-9 y; J,. __,p - - s d 1 . M A s 7 .t=:,c;:. ai -= - f , y f))"* w!_.__ ML.\\ wT~hv w edn~ i.# x - -. sw.- _+ 1

+r-

~ - 4 m#... %4'i-9.-J C~%,lEkM. R 4 4_p-f'T c, 2=. b dr/

1. Er

'J ' s,. _ -$ WWW -== 1 0 --

.

11 /) - ? A- = 9. _, s ~ zu m 4 9 a .l ,'l_ i : 7,,,,., :. h 3 g._ _,.,, _- u I g' p'4stD _ y i. NT 5 ,[', I [ -4 4 T. ) J 1')l.,.M b-w%.. b e 6.sme.e.es r. ca.** _E = l ei s, ' ; _. ~Lg. ] is=i \\,TEL [ ym'"*** r8 .= g~__ .x- , 3 e-- ';' ~~ , s._ __. 3 H \\ v u, e p ) bC W_- fu

==== w e 5 1 y ,q ~ J" ~ =~ h p\\; g ,wn '~~ F,, J / N g. { ys=; %,Y.n_ a w, a EP m--- p ae '.h.// . \\ &m u _ ~^ EEE. (**D e I 213': --. PL AN ELEV 51-3, O O 6 i- _ _rp . - ( _- _. " '"- @. T.. - ,) j V7 e p'. e* V ._ n g i g p-%g v poi = (i (g -j. i r_. LOCAT10Ne~:,Dt'& (,,~ m L A, g

=

a .q kLoCAtt0N .en[y6g' = n 5. ~ ..- y,'<anad,8. 2. g::5 I I 1 S.' N R. S4 ( 5'"""" ._ k 15 i.-" Aia a= = * -~~- T' _ __ _ d _ Q 4,..:. p:. e.,

r. r v j

ni n.83. a.a= rav 2. esasa AA g his lid i 9 i i t

.... _. ~. _ _.. _ _ a T/R/02. ~ APPEA@tX C i Using FIVE Fire Modeling to Approximate Exposure Time-History for Thermo-Lag Configurations in OC Fire Zone RB-FZ-1 F2 Letter Report September 26,1995 Prepared by: . GPUN Risk Analysis Group Using FIVE Fire Modeling to Approximate Exposure Time-History for Thermo-Lag Configurations j / el

a T i ) 1. PURPOSE The purpose of this report is to document the detailed tre modeling performed for fire j scenarios invoMng Thermo-Lag wrapped cable conduits as identified during plant walkdowns j { of OC fire zone RB-FZ-1F2. Thermo-Lag exposure time history results were obtained for each i l T-L configuration determined to be impacted by a fire source / combustible. i l 2. APPROACH i l The approach followed for modeling of fire soonarios is based on methodology detailed in the EPRI

• Methods for Evaluation of Cable Wrap Fire Barrier Performance" document (Ref.1) which was based on Are modeling techniques developed for FNE (Ref. 2). Fire modeling was

) j accomplished in two steps: s l

1. Screening walkdowns to eliminate from further consideration those fire ignition l

sources and potential fire scenarios that cannot develop into damaging fires. j i j

2. Fire modeling of specific scenarios that were not screened out in the first step.

For each Thermo-Lag configuration analyzed, one or more fire scenarios (including bounding j scenarios) were postulated based on the following: s a. Information gathered during plant wakdowns (i.e., itxed or transient ignition j source and Thermo-Lag target location, type and quantity of combustibles present in zone); b. Information obtained from reviewing applicable sections in Appendix R documents for OC (Ref. 3); j c. information obtained from reviewing OC procedure 120.5, Rev. 5, " Control of j Combustibles." (Ref. 4) d. Information provided by GPUN engineers (specializing in fire protection, risk l assessment, mechanical). i l Thermo-Lag FNE analyses were performed using computertzed Excel spreadsheets which duplicate FNE Worksheets 1,2, or 3 and are linked to tables containing the same information i as in FNE Reference Tables. For each Thermo-Lag fire modeling scenario, results are presented i as exposure time-temperature profiles. In all scenarios evaluated, time-dependent temperatures j were derived by incrementally adding the total heat content of the fuel into the fire ceiT,p.6,i,Ent l hot gas layer. l I i i 3. FIRE MODEUNG ASSUMPTIONS i The following Thermo-Lag fire modeling assumptions were made: i 1. Worst-case fixed fire source scenarios were mowed for each inermo-Lag sub-configuration. The determination of which itxed fire source might cause the worst-4 1 n . ~,... n

i l + was based on minimum damage threshold heights and distances calculated for typical fixed fire sources (e.g., electrical cabinets, electrical motors, lube oil in pumps) found at Oyster Creek. 2. Worst-case transient fire source scenarios observed during walkdowns were also modeled. Worst-case transient fire scenarios considered were similar to the type and quantities of transient combustibles allowed by OC station procedure (Ref. 4). 3. To ensure conservative results, worst-case fire plume (or ceiling jet) scenarios and/or radiant flux scenario were first modeled. For T-L sub-configurations not in the plume or ceiling jet, hot gas layer fire scenarios were evaluated. 4. Electrical cable inside cabinets, motor control centers, switchgear and cable trays was assumed to be norqualified (i.e., non IEEE-383 type). This assumption is conservative since electrical cable purchased and installed at Oyster Creek in the past decade was IEEE-383 type. 5. The Fire Hazard Tool developed for the EPRI Cable Wrap Fire Barrier Tallored Collaboration assumes that the fire cciTviu u n-i,1 temperature will follow the ASTM ^ E-119 time-temperature curve as soon as the compartment hot gas layer (HGL) reaches 10007. When the HGL temperature reaches 10007, the plume (or ceiling jet) temperatures are assumed to follow that of the E-119 (see discussion in Appendix B). 6. The Thermo-Lag wrap is assumed to bum by itself (i.e., even after the initial fire source is out of fuel) unless otherwise stated. Per, the EPRI Method (Ref.1, Appendix l-D), credit may be taken for a time lag before the Thermo-Lag ignites based on exposure temperature. 7. The maximum plume and ceiling jet temperature is assumed to be 16007, which corresponds to flame temperature. This assumption is valid until hot gas layer temperature exceeds 16007, at which point the plume temperature is assumed to be equal to hot gas layer temperature. 4. FIRE MODEUNG RESULTS 4.1 Fire Zone RB-FZ-1F2 Fire zone RB-FZ-1F2 is the Reactor Building South-West Comer Room at Elevation -19'6". This fire zone houses two core spray pumps, the reactor building equipment drain tank and, the reactor building equipment drain pump. The T-L wrap conduit run in this fire zone starts at Elevation 2 ft. (about 21 ft. off the floor) and runs along the west wall and then curves and exits through the ceiling. The general location of the T-L wrap in this fire zone is shown as location # 2 on the general artangement drawing shown in Appendix D.This area was not accessible during / the walkdowns due to high radiation and the information used here was obtained from previously / taken pictures, drawings, and communication with plant fire protection personnel. 2

I a i Scenario No.1 I The fixed fire ignition source considered here is the Reactor Building Equipment Drain Pump (P-l 22001). This pump contains up to 2.5 gallons of tube oil in the motor and the gearbox (Ref. 6) and sits at 4* Elevation (General Arrangement Drawings).This source was chosen because it is i more bounding than the core spray pumps. Each of core spray pumps contain up to 2 gallons of oil (Ref. 6) and they are at Elevation -19 ft. The total Q available for this source is 3g7,600 BTU j (Ref. 5, Page E2-1, assumed to be the same at OC), heat release rate 110 BTUs/sec-ft (frg,m Ref. 2, Table 2E). The potential oil spread area around the pump was estimated at 5 ft The Thermo-Lag wrap is located 8 ft above the fire source at its closest proximity and is outside of i a j fire plume. The floor area for this Are zone is 560 ft (Ref. 3). The ceiling height is 40 ft. (general i arrangement drawings). Ambient temperature assumed was 100 T. Results The time temperature profile scenario calculated by the model are presented in Table 1. The i model results indicate that hot gas layer effects in the zone due to this postulated fire cause the temperature to rise to approximately 194T. A sample model worksheet is provided in Appendix A, Figure A-1. l Table 1. Temperature-Time History for Fire Scenario 1. j Time HRH to HMM to UTU1 Temp. Temp. j Target HGL (min.) (Blu/s) (Btu /s) (Btu) Target - (F) HGL - (F) O 550 550 0 100 100 l 5 550 550 155000 135 135 i 10 550 550 330000 179 179 { 11.7 550 550 355100 1M 1M i 12 0 0 355100 1M 1H 15 0 0 355100 1H IM l 20 0 0 355100 1M IM 30 0 0 355100 1M IM 40 0 0 355100 1M IM 3 i 50 0 0 355100 1M 1H SO O 0 355100 1M IM / J = 3

Scenario No. 2 This fire scenario modeled a transient ignition source / combustible, a 5 gallon container containing lube oil. Per Oyster Creek Procedure " Control of Combustibles," Procedure No. 120.5, Rev. 4 (Ref. 4), combustible liquids could be transported throughout the plant in.an approved closed metal container (but not necessarily a safety-can). For this fire scenario, it was postulated that 1 quart of lube oil is spilled out of a can on the floor and is not cleaned up. The oil is later ignited by a welder working nearby or another transient ignition source and begins to burn. The oil fire could potentially impact a T-L wrap 20 feet above the floor. The heat release rate for the iube oil is 110 Btu /s ft.2 (from Ref. 2, Table 2E). The oil could spread unconfined over an area of 13.5 ft 3 (54 ft.2/ gal from Ref. 2, Table 3E). The heat release rate ass'umed for this fire scenario is 1,485 Stu/s (110 Btu /s-ft.a x 13.5 ft.2). The heat content of one quart of oil is (17.111 Btu /lb.

• 60 lb/1 ft 3.
• 1 ft.3/30 quarts) 34,222 Btu (from Ref. 2, Table 2E). The oil fire duration will be less than one minute, but one minute will be used for conservatism. The virtual surface of the fire is at the floor level. The location factor for this source is one.

Results The short oil fire does not cause an exposure temperature high enough to ignite the Thermo-L.ag. The fire is of a short duration, with very slight hot gas layer effects. A sample model worksheet is provided as Appendix A, Figure A-2. - 4 Table 2. Temperature-Time History for Fire Scenario 2. Time HRR to HRR to QTOT Temp. Temp. Target HGL (min.) (Blu/s) (Blu/s) (Btu) Target - (F) HGL - (F) 0 1485 1485 0 400 100 1 1485 1485 89100 414 114 2 0 0 89100 114 114 10 0 0 89100 114 114 15 0 0 89100 114 114 25 0 0 89100 114 114 30 0 0 89100 114 114 35 0 0 89100 114 114 40 0 0 89100 114 114 50 0 0 89100 114 114 / 60 0 0 89100 114 114 l / 4 r

5. FIRE MODELING

SUMMARY

Fire modeling was performed for one fixed and one transient fire Source-Thermo-lag wrap configuration in Oyster Creek Fire Zone RB-FZ-1F2. The analysis results are presented as time-temperature profiles. These results are used in the following section to develop total heat load and then convert to equivdent ASTM E-119 exposures. 6. APPROXIMATING EXPOSURE TIME-HISTORY OF THERMO-LAG SUB-CONFIGURATIONS - THE TOTAL HEAT LOAD CONCEPT Calculation of an exposure time-temperature profile for a specific T-L Sub configuration is based on the methodology presented in Reference 1. The time 4emperature profiles are used to develop the total heat load (Reference 1 Appendix Ill-A) for each fire exposure. The total heat load concept holds that the area under the incident heat flux vs. time curve, or total heat load, can be used as a measure of fire severity. CeT.p.i'.g the total heat load of a fire n scenario exposure to the total heat load of a test exposure (ASTM E-119) allows one to predict the response of a Thermo-Lag barrier to the scenario exposure. The total heat load per unit area of fire barrier surface is equal to the area under the incident heat flux-time curve, corrected for convection effects as described in Reference 1, Appendix lil-A. 6.1 Total Heat Load Results Location Scenario Descript!on Peak Total Heat Load No.(GA Temperature at Equivalent E-119

DWG, Target ("F) 8&=e Duration App.D)

(min.) 2 1 Oil fire at R5F.D 194 <12 Pump (P-22-001) 2 2 oil Transient 414 <10 4 Appendix C contains tables with Total Heat Load values calculated for Thermo-Lag scenarios 1, and 2. 7. REFERENCES / 1. EPRI Report " Methods for Evaluation of Cable Wrap Fire Barrier Performance", July, 1995 2. EPRI TR-100370, " Fire-induced Vulnerability Evaluation," April 1992. 5

O 3. GP , =Re hds Analysis Report - OC," Doc. No, 9901746, Rev. 8, May 24, 4. GPUN - OC Procedure No.120.5 (Revision 5), May 20,1993. 5. GPUN - TMI-1 Procedure No.1035 (Revision 24), %e, m 6. P communication with Malcolm Gonzales of GPUN, in charge of plant og Wedumb e i e / 6 ep -^

O e i Appendk A Sample Scenario Worksheets 1 = / \\ { l i

Figure A-1 4 Temperature calculabon at time min 5 Maximum ambient temperature F 100 FloorArea ft2 560 1 Location of Target Hot Gas Layer 2 Height of Target above Fire Source R 8.00 3 Hei0ht from Five Source to Ceiling, H A 28.00 4 Ratio of Target Height /Colling Height 0.286 5 LonDiludinal TAstance from Fire Source to A NA Tornet,L_ 6 Longitudinal Distance to Height Ratio, UH NA 7 Enclosure Wufth, W A NA 8 Height to Width Ratio, H/W NA 9 Peak Fire intensity Stuts 550 10 Fire Location Factor 1 11 Effective Fire intensity Stuts 550 12 Plume Temperature Rise at Target F 713 13 Temp Rise Famor at Target 0.00 14 Temperature Rise at Target F 0 15 Temperature at Target"~~ F 138 16 Temperature in HGL F 138 17 Otot to HGL Stu 165000 18 Estimated Heat Loss Fraction 0.94 19 Calculated Onet Stu 9900 20 Calculated Enclosure Volume, V A3 15680 21 Calculated OnetN Onet/ft3 0.63 22 HGL Temperature increase F 38 l ~ / i l

e Figure A-2 J Temperature calc utation at time min 1 Maximum ambient temperature F 100 Floor Area A2 560 1 Location of Target Piume 2 Height of Target above Fire Source ft 20.00 3 Height from Fire Source to Ceiling H ft 40.00 4 Ratio of Target Height / Ceiling Height 0.500 5 Longitudinal Distance from Fire Source to ft NA TarDat, L 6 Longitudinal Distance to Height Ratio, UH NA 7 Enclosure Width, W ft NA 6 Height to Width Ratio, H/W NA 9 Peak Fire intensity Stu/s 1465 10 Fire Location Factor 1 11 Effective Fire intensity Stu/s 1485 12 Plume Temperature Rise at Target F 300 13 Temp Rise Factor at Target 1.00 14 Temperature Rise at Target F 300 15 Temperature at Target F 414 16 Temperature in HGL F 114 17 Otot to HGL Btu 89100 18 Estimated Heat Loss Fraction 0.g4 19 Calculated Onet Stu 5340 20 Calculated Enclosure Volume, V ft3 22400 21 Calculated QnetN Qnet/ft3 0.24 22 HGL Temperature increase F $4 / 1 j

J Appemsxs Basis for Following ASTM E-119 Time-Temperature Relationship When T(HOL) is 1000 Y O t ~ as

- _ - - = - . - ~.. -. - s r a TRANSITION TO ASTM E-119 TIME-HISTORY IN THE HOT GAS LAYER ISSUE: The Fire Hr4M Tool currently under d=;: -;-Teat in the EPRI Cable Wrap Fire Barrier Tailored C % oration assumes that the fire ceTipeii,i,W.i temperature will follow the ASTM E-119 time-f aN4rature relationship as soon as the compartment hot gas layer reaches 1000T. - BASIS: The Handbook (BgL, pg. 675) states that Flashover of an enclosure is likely to occur if the temperature of the upper gas layer reaches approximately 1000'F (600*C). The EPRI assumption of 1000'F is therefore conservative. As stated near the top of page 6-67 of the Handbook, the intensity of the fire will be somewhat lower when the walls and ceiling absorb significant amounts of energy rather than when act as insulation or radiation barriers. Since power plant compartments are typically concrete barriers which absorb significant amounts of heat, the intensity of fire will tend to be less severe than fires in compartments with insulated barriers. ASTM E-119 has been the accepted standard for testing fire barrier systems since the earfy 1900's. The standard, which was developed from actual fire tests, has been used to evaluate the fire endurance capabilities of fire barriers since it was developed. The time-temperature relationship represented by the curve represents a fully-involved fire compartment. According to Fred Mowrer of the University of Maryland, plume and ceiling effects disappear in a fully involved fire compartment as the plume / ceiling jet gas are no longer more buoyant than surrounding gases. Use of the ASTM E-119 time-temperature relationship to represent a fully involved power plant compartment fire is therefore consistent with industry practice. Assuming that that occurs when the hot gas layer reaches 1000'F is more conservative than the 11007 noted in the Handbook. /

Reference:

NFPA Fire Protection Handbook, Seventeenth Edition, Section 6/ Chapter 6. 9

i 2 1 i i 1 l 1 4 l Appendix C l I Total Heat Load Results l s I l l l I W \\ l to

Floure C-1 Total Heat Load Scenario 1 TarDat Hot Gas Layer ASTM E-119 Time Temperatune inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min) (F; (Blu/s-sf) (1000 Btu /sf) (1000Blu/sf) (1000Blu/sf) 0 100 0.4 0.0 0.00 0.00 5 138 0.3 0.10 0.10 0.34 10 179 0.4 0.20 0.20 1.37 11.7 194 0.4 0.24 0.24 1.85 12 194 0.4 0.25 0.25 1.93 15 194 0.4 0.32 0.32 2.88 20 194 0.4 0.45 0.45 4.74 30 194 0.4 0.70 0.70 9.11 40 194 0.4 0.95 0.95 14.17 50 194 0.4 1.20 1.20 19.80 60 194 0.4 1.45 1.45 25.91 I e e / )

~.. - Figure C 2 ,I e Total Heat Load Scenario 2 Target Hot Gas Layer ASTM E 119 T6me Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min) (F) (Stu/s-sf) (1000Blu/sf) (1000 Btu /sf) (10005tu/sf) 0 400 0.7 0.0 0.00 0.00 1 414 0.6 0.04 0.02 0.G0 2 114 0.2 0.06 0.03 0.02 10 114 0.2 0.18 0.15 1.23 15 114 0.2 0.26 0.23 2.80 20 114 0.2 0.33 0.30 4.66 25 114 0.2 0.41 0.38 6.75 30 114 0.2 0.48 0.45 9.04 40 114 0.2 0.63 0.60 14.10 50 114 0.2 0.78 0.75 19.73 60 114 0.2 0.93 0.90 25.84 j i 0 6 ) i l / 4 9

A 3 Appendix D General Arrangement Drawings O I i 1 i \\ \\ 9 e

a 'p, 7 0 s 4 a l l l 4 H 1_ PLANT T-800STM - g g Q h h

a...

g,... _ 9, g = n : : :: < g.= l ;. .y = ~g-,f- ~ * * * * * .~...... 4 lQ{-. n. Q%M) 9/ ' ~ ' ' W

m. mMl

/, N ja? Leccio i s ,.= \\ / g, =. .'*tr p \\ /

ll3,,,

?;. .f, ( %, ~~ '~ 'E eWeT*

• 5

.w / \\ .}$W Sff '-==- M L i _,_,# N' \\Ms% M ""',7 3 'N' es-l-MQ ".. ] y" i na W 1 s e / !l!E g !,..,.,,i K x I Il','E','! n- / 4 h. y ' $m., Tj [N-a"'" i i R (_?',

1. ,'~$f

"'*m l..r l$~'" ?Y ' Y.E. o '%nt ?

f-

,/W I il ).f *W W - I'd e hang i ,. g V.7 .*' / l 1H L l=- 3.su#.. m m-.. \\ L'llll7I- ,,,,,,,,,,_/ "' f8'% -e ".sdr b,,$ M:='- * ..%N. a-a - *= D loa Eill'NI' N c PLAN EtEV 8 819'- G* g ilEEliE @=F#. r ::::: A T - F@jL7== e @ = a=. =. _c --x i 1 r k LOCATION ~~" 42 J%l- _-- 1 emames. i 1 w.-7.,,._,'M .x/ /N., 4" m m. e-b. m _. f. =.a, w ii.::p =n- '.r .,g 57gy&_,pantini g 3 u 3 4': et an smaviat et as. ri rv.-ic-s' ELE v -t-er '5 .# DibQ(T s .n 1 ilDOWa -a I v n -' I i t

i M /BA f)Pfemix D i I Evaluation of Thermo-Lag Cable Wrap Fire Barriers in Oyster Creek Fire Zone TB-FZ-11D i l Report No. 01-0082-05-2965-700-1 i July 21,1995 j i 1 Prepared by: l Science Applications International Corporation,Inc. 9

~ _ _ _ _ _ _ _.. _. _ _ _ _ _ _ _ _ _ -. 1 SAIC Report No. 01-0082-05-2965-700-1 1. PURPOSE The purpose of this report is to document the detailed fire modeling performed for fire scenarios involving Thermo-Lag wrapped cable conduits, as identified during plant walkdowns of the Oyster Creek Turbine Building Basement Floor, South End, Fire Zone TB-FZ-11D. This fire zone includes two separate Thermo-Lag (T-L) conduit configurations (based on plant walkdown notes of October 27,1994 and Oyster Creek i plant drawings of Appendix R Raceway-Turbine Building Basement, Ref.1). Fire exposure time-temperature profile results were obtained for each of these two T-L - configurations, as analyzed in seven fire modeling scenarios involving fixed and transient fire sources. l 2. APPROACH The approach followed for modeling of fire scenarios is based on methodology detailed in the EPRI " Methods for Evaluation of Cable Wrap Fire Barrier Performance" d'ocument (Ref. 2). which was based on fire modeling techniques developed for FIVE (Ref. 3). Fire modeling was accomplished in two steps:

1. Screening walkdowns to eliminate from further consideration those fire ignition sources and potential fire scenarios that can't develop into damaging fires.

2. Fire modeling of specific scenarios that were not screened out in the first step.

For each of the two Thermo-Lag configuration analyzed, one or more fire scenarios (including bounding scenarios) were postulated based on the following: information gathered during plant walkdowns (i.e., fixed or transient a. ignition source and Thermo-Lag target location, type and quantity of combustibles presentin zone); b. information obtained from reviewing applicable sections in Appendix R documents for Oyster Creek Unit (Ref. 4); c. information obtained from reviewing Oyster Creek procedure 120.5, Rev. 5, " Control of Combustibles" (Ref. 6) d. information provided by GPUN engineers (specializing in fire protection, risk assessment, mechanical). Thermo-Lag FIVE analyses were performed using computerized Excel spreadsheets which duplicate FIVE Worksheets 1,2, or 3 and are linked to tables containing the same information as in FIVE Reference Tables. For each Thermo-Lag fire modeling scenario,. results are presented as exposure time-temperature profiles. In all evaluated scenarios, ! time-dependent temperatures were derived by incrementally adding the total heat content of the fuel into the fire compartment hot gas layer. 4 1 y --y

SAIC Report No. 01-0082-05-2965-700-1 3. FIRE MODELING ASSUMPTIONS The following Thermo-Lag fire modeling assumptions were made: 1. Worst-case fixed fire source scenarios were modeled for each Thermo-Lag sub-configuration. The determination of which fixed fire source might cause the worst-case Thermo-Lag fire scenario was made during the plant fire area walkdown and was based on minimum damage threshold heights and distances calculated for typical fixed fire sources (e.g., electrical cabinets, electrical motors, lube oil in pumps) found at Oyster Creek. 2. Worst-case transient fire source scenarios observed during walkdowns were also modeled. Worst-case transient fire scenarios considered were identical to the types and quantities of transient combustibles identified specifically for TB-FZ-11D in walkdown. 3. To ensure conservative results, worst-case fire plume (or ceiling jet) scenarios and/or radiant flux scenario were first modeled. For T-L sub-configurations not in the plume or ceiling jet, hot gas layer fire scenarios were evaluated. 4. Electrical cable inside cabinets, motor control centers, switchgear and cable trays was assumed to be non-qualified (i.e., non IEEE-383 type). This assumption is conservative since electrical cable purchased and installed at Oyster Creek in the past decade was IEEE-383 type. 5. The Fire Hazard Tool developed for the EPRI Cable Wrap Fire Barrier Tailored Collaboration assumes that the fire compartment temperature will follow the ASTM E-119 time-temperature curve as soon as the compartment hot gas layer (HGL) reaches 1000*F. When the HGL temperature reaches 1000 F, the plume (or ceiling jet) temperatures are assumed to follow that of the E-119 (see discussion in Appendix B). 6. The Thermo-Lag wrap is assumed to burn by itself (i.e., even after the initial fire source is out of fuel) unless otherwise stated. Per, the EPRI Method (Ref. 2, Appendix I;D), credit may be taken for a time lag before the Thermo-Lag ignites based on exposure temperature. ) 7. The maximum plume and ceiling jet temperature is assumed to be 1600 F, which corresponds to flame temperature. This assumption is valid until hot gas layer temperature exceeds 1600 F, at which point the plume temperature is assumed to be equal to hot gas layer temperature. 2

SAIC Report No. 010082-05-2965 700-1 8.- ' Fire Zone'TB-FZ-11D specific data: total area of 9,668 ft' (Ref. 4 - Oyster Creek Fire Hazard Analysis e Report, Rev. No. 5); fire zone ceiling height of 20 feet (General Arrangement dwg.); e ambient temperature of 100'F. e 4. ' FIRE MODELING RESULTS 4.1 Thermo T.ma C'nnfiauratirm 1. Scenario 1 The ignition source / combustible considered in this scenario is a Protective Clothing I (PC) storage bin (a transient combustible) observed during the plant walkdown. It is assumed that the PC bin (3 ft tall) contains 50 lb. of PC's and the plastic container itself, for a total heat content (Q) of 800,000 Btu (from Ref. 5, page E2-1). The heat content assumes 400,000 Btus for the plastic container and 400,000 Btus (8000 Btu /lb.

• 50 lb.)

for the PCs. A heat release rate of 285 Btu /s was chosen for this fire source based on test data (Ref. 2 App. I-E NBS-Lee, clothing). The test report indicates that there were 10 lb. of clothing used for this test and that the heat release rate was 57 Btu /s. The heat release rate is assumed to increase in proportion to the mass of the PCs. This increase is conservative, because heat release rates are normally more dependent on surface area j than on combustible mass. At this heat release rate, the PC bin source will burn for 47 minutes (800,000 Bru/285 Btu /s). The targets in this scenario are two Thermo-Lag wrapped conduits, CGCTB019 and CGPTB020, each 1.5 inch in diameter, (Ref.1). The Th'ermo-Lag wrap is assumed to be 0.5 inch thick, with a 6 inch combined total diameter. The targets are located 5 feet above the room floor and 3.5 feet offset from the source / combustible (i.e.,in the hot gas layer). However, one T-L wrapped support is located directly in the plume of the fire and is assumed to ignite and burn for the entire scenario. The heat release rate for the T-L wrap is 8.8 Btu /s/ft.' (Ref. 2, Appendix I-D). The T-L heat release rate of 41.5 Btu /s was added to that of PC bin fire source (285 Btu /s) for the hot gas layer calculations. The virtual surface of the fire is 3 feet above floor level. The fire source is not against the room wall, so the location factor is one. Results The time temperature profile scenario calculated by the model are presented in Table 1. The time-temperature profile shows that the temperature at the target will be at flame temperature for the entire scenario. The target in this case is the Thermo-Lag wrapped support, not the protected conduits themselves. The conduit exposure is expected to be only hot gas layer because it is outside of the plume region. The model results indicate that hot gas layer effects in the zone due to this postulated fire are negligible. A sample model worksheet is provided as Appendix A, Figure A-1. The time-temperature profile for this scenario is shown in Table 1. 3 s ---r ~ v-

__._.m r SAIC Report No. 010082-05-2965-700-1 Table 1. Temperature-Time History for Fire Scenario Modeled in Configuration 1, Scenario 1. Time HRR to HRR to HGL QTOT Temperature Temperature Target (min.) (Stu/s) (Btu /s) (Blu) Target - (F) HGL - (F) 0 285 327 0 1600 100 5 285 327 98100 1600 102 10 285 327 196200 1600 104 15 285 327 294300 1600 106 20 285 327 392400 1600 108 25 285 327 490500 1600 111 30 285 327 588600 1600 113 40 285 327 784800 3600 117 47 285 327 922140 1600 120 48 0 42 924660 1600 120 . 60 0 42 954900 1600 121 4.2 Thermo-I na Confieuratinn 1. Scenario 2 Scenario two postulates a fire in a 3' wide electrical cabinet on the wall,4 feet off the ground. The location factor for the fire is two. The target for the scenario is an unwrapped 12" cable tray 7 feet directly above the panel,11 feet off the ground. A second 24" cable tray is located 1 foot above the first cable tray. In addition to the vertical propagation, the fire is also assumed to propagate horizontally along the exposed cable tray with a flame spread rate of 10 feet per hour. The T-L of concern for this scenario is configuration 1 which are 6 feet offset from the cable trays and three feet below the ceiling. They will see only hot gas layer exposure. The heat release rates used for this scenario are as follows:

1. electrical panel with non qualified cable - 400 Btu /s (Ref. 3); this HRR is applicable to scenario time t=0 to 15 minutes;

2. first cable tray - 23.4 Btu /s-ft.' (Ref. 2, FIVE Table IE, cable #5 adjusted to consider Equation 2 on FIVE manual, page 10.4-10); since 3 ft.' (3 linear feet,1 foot wide) are initially burning, the HRR for this source alone is 70 Btu /s. An additional 23.4 Btu /s are added every 6 minutes to account for fire propagation;

3. second cable tray - 23.4 Btu /s-ft.' (Ref. 2, FIVE Table IE, cable #5 adjusted to j

consider Equation 2 on FIVE manual, page 10.4-10); 10 ft' (5 linear feet,2 feet wide) are burning, therefore the HRR for this second cable tray is 234 Btu /s, and additional 47 Btu /s are added every 6 minutes to account for fire propagation. I l 4 l

- - -. -. -... ~ _. .-.-._.- -.-.~. -..- SAIC Report No. 01-0082-05-2965-700-1 The total available heat (Q,,) from each fire source is:

1. electrical panel - 360,000 Btus (assumes that the panel contains enough fuel to sustain a 15 minute fire);

2. first cable tray - 2,400,000 Btus [ assumes 10,000 Bru/lb. of cable tray (per Ref. 5, page E2-1) x 10 linear feet x 24 lb./ linear foot (engineering judgment for 100% filled, 12 inch wide cable tray)). The fuel willlast for the full one hour duration of the fire;

3. second cable tray - 4,000,000 Btus [ assumes 10,000 Btu /lb. of cable tray (per Ref. 5, page E2-1) x 10 linear feet x 40 lb./ linear foot (engineering judgment for 100% filled, 24 inch wide cable tray)). The fuel willlast for the full one hour duration of the fire.

Results The results are presented in Table 2. The model results indicate that hot gas layer effects in the zone due to this postulated fire cause the temperature to rise to approximately 322*F. A sample model worksheet is provided in Appendix A, Figure A-2. Table 2. Temperature-Time History for Fire Scenario Modeled in Configuration 1, Scenario 2. Time HRR to HRR to HGL QTOT Temperature Temperature Target (min.) (Blu/s) (Stu/s) (Blu) Target - (F) HGL - (F) 0 704 704 0 100 100 5 774 774 232260 111 111 10 844 844 485580 123 123 15 915 915 759960 136 136 ~ 16 985 985 819048 138 138 25 1055 1055 1388748 167 167 30 1125 1125 1726308 184 184 35 1195 1195 2084928 203 203 40 1266 1266 2464608 224 224 50 1336 1336 3266088 270 270 60 1406 1406 4109688 322 322 4.3 Thermo-Lag Confieuration 1. Scenario 3 This fire scenario modeled a fire in an oil-filled transformer near 460V unit sub-station I-A-1. The transformer contains 190 gallons of mineral oil (from the plant walkdown). Each gallon of oil is equivalent to 125,656 Btus (from Ref. 3, Table 2E), for a total of about 23.9 million Brus for the entire volume of oilin the transformer. The heat release rate for the transformer oil is 135 Btu /s-ft' (from Ref. 2, Table 2E). The oil could spread outside the transformer over an area of 98.5 ft'(measured during the plant walkdown). The heat release rate assumed for this fire scenario is 13,300 Btus/s (135 Bru/s-ft' x 98.5 ft'). The fire duration is then (23,9M Btu /13,300 But/s) 30 minutes. The nearest T-L target is 1 foot off the room ceiling and 15 feet offset from the edge of the postulated s

SAIC Report No. 01-0082-05-2965-700-1 target is 1 foot off the room ceiling and 15 feet offset from the edge of the postulated transformer oil pool fire. The T-L target in this scenario is configuration 1. It consists of two conduits, CGCTB019 and CGPTB020, each 1.5 inch in diameter, wrapped in Thermo-Lag (Ref.1), with a 6 inch combined total diameter. They are in the ceiling jet region of the fire. When the exposure temperature at the targets reaches 1000'F, the Thermo-Lag is assumed to ignite. A twenty foot length of Thermo-Lag with a surface area of (20'

• 0.5's) 31.4 ft.'is assumed to start burning with a heat release rate of (8.8 i

Btu /s/ft.'

• 31.4 ft.') 276 Btu /s, which contributes to the hot gas layer. Because the Thermo-Lag is burning when the oil fire goes out, the exposure at the Thermo-Lag is conservatively assumed to remain the same for the remaining fire duration.

Results The time temperature profile for this model shows that the temperature at the target will reach approximately 1278'F due to a fire of this nature. This temperature is above the ignition temperature of Thermo-Lag. The model results indicate that hot gas layer effects in the zone due to this postulated fire will not reach flashover in the zone. A sample model worksheet is provided in Appendix A, Figure A-3. Table 3. Temocrature-Time Hietarv for Fire Scenario Modeled in Confieuration 1. Scenario 3. Time HRR to HRR to HGL QTOT Temperature Temperature Target (min.) (Btu /s) (Btu /s) (Btu) Target - (F) HGL - (F) 0 13300 13300 0 613 100 5 13300 13300 3990000 690 177 10 13300 13300 7980000 778 266 15 13300 13300 11970000 879 366 20 13300 13300 15960000 993 480 22 13300 13576 17589120 1044 532 30 13300 13576 24105600 1278 766 31 0 276 24122160 1278 766 40 0 276 24271200 1278 772 50 0 276 24436800 1278 779 60 0 276 24602400 1278 785 4.4 Thermo-Lae Confieuratinn 2. Scenario 1 ) This fire scenario models a fixed ignition source / combustible. The source is a 2 foot wide cable tray with non IEEE-383 cable,16 feet off the room's floor. Since the cable is non-qualified, the cable fire can self-ignite (per Ref. 2, Appendix I-A) and start to burn. The fire will then propagate immediately to a second cable tray, located 1 foot above it. In addition to the vertical propagation, the fire is also assumed to propagate horizontally along the exposed cable tray with a flame spread rate of 10 feet per hour (Ref. 2 Appendix I-C). The total amount of cable tray assumed to be available for this 6 ,m

~. SAIC Report No. 01 0082-05-2965-700-1 ~ i fire scenario is 10 feet for a 1 hour fire duration. The cable trays are only about 50% filled (visual observation during plant walkdown). The targets are two 2-inch diameter conduits wrapped in 0.5 inch thick Thermo-Lag. The T-L wrapped conduits pass perpendicularly over the cable trays,3 feet above them. Due to the close distance between the cable trays and T-L wrapped conduits, the T-L will be ignited by the cable tray fire (assumed conservatively to happen at time t=0 minutes). The heat release rates considered for this scenario are as follows (by fire source):

1. first cable tray - 23.4 Btu /s-ft.' (Ref. 2, FIVE Table IE, cable #5 adjusted to consider Equation 2 on FIVE manual, page 10.4-10); since 4 ft' (2 linear ft.) are assumed to be initially burning, the HRR for this source alone is'94 Btu /s; this HRR will increase with time, as fire propagates horizontally along the tray (i.e.,6 minutes into the fire, the HRR for this source will be 140 Btu /s, which includes the HRR from 2 ft' of additional cable tray);

2. 'second cable tray - 23.4 Btu /s-ft.' (Ref. 2, FIVE Table IE, cable #5 adjusted to consider Egitation 2 on FIVE manual, page 10.4-10); the HRR for this source alone is 187 Bru/s since 8 ft' (4 linear ft.) are assumed initially burning; this HRR will increase with time, as fire propagates horizontally along the tray (i.e.,6 minutes into the fire, the HRR for this source will be 234 Btu /s, which includes the HRR from 2 ft.' of additional cable tray);

3. T-L wrap HRR is assumed to be 8.8 Btu /s-ft.'(per Ref. 2 Appendix I-D); the amount of Thermo-Lag that could be involved in this fire scenario is assumed to be 10 linear feet (visual observation) with a circumference of (2 *.25'

• x) 1.6 ft. for a total surface area of 16 ft'; the HRR is thus 141 Btu /s; this HRR will be considered only for calculating the HGL temperature rise.

3 The total available heat (Q,,) from each fire source is:

1. first cable tray - 2,000,000 Btus [ assumes 10,000 Btu /lb. of cable tray (per Ref. 5, page E2-1) x 10 linear feet x 20 lb./ linear foot (engineering judgment for 50% filled, 24 inch wide cable tray)); This source has enough combustible material to burn for one hour.

2. second cable tray - 2,000,000 Btus [ assumes 10,000 Btu /lb. of cable tray (per Ref. 5, page E2-1) x 10 linear feet x 20 lb./ linear foot (engineering judgment for 50% filled,.

i 24 inch wide cable tray)]; This source has enough combustible material to burn for one hour. I

3. Thermo-Lag wrap - 588,000 Brus (assumes 58,800 Btu / linear foot, from Ref. 7).

Results The time temperature profile shows that the temperature at the target reaches flame temperature immediately, and remains at that temperature for the remainder of the fire (1 hour). This temperature is above the ignition temperature of Thermo-Lag. The 7 1

l SAIC Report No. 01 0082-05-2965-700-1 e model results indicate that hot gas layer temperature will reach 702*F after an hour. A sample model worksheet is provided in Appendix A, Figure A-4. i Table 4. Temperature-Time History for Fire Scenario Modeled in Configuration 2, Scenario 1. i Time HRR to HRR to HGL QTOT Temperature Temperature Target (min) (BTU /s) (BTU /s) (BTU) Tar 9et -(F) HGL - (F) 0 281 422 0 1600 100 6 375 516 185616 1600 123 12 468 609 404928 1600 151 18 562 703 657936 1600 186 24 655 796 944640 1600 227 30 749 890 1265040 1600 276 36 843 984 1619136 1600 335 42 936 1077 2006928 1600 405 48 1030 1171 2428416 1600 487 54 1123 1264 2883600 1600 586 60 1217 1358 3372480 1600 702 4.5 Thermo-Lac Configuratinn 2. Scenario 2 ' This fire scenario fixed' ignition source / combustible is a 2 foot wide electrical cabinet with exposed vertical cable igniting a 24 inch wide cable tray,14 feet above the room's floor. The scenario postulates that the fire propagates horizontally along the cable tray at a rate of 10 feet per hour (since cable is not IEEE-383 qualified), but away from the nearest Thermo-Lag targets. The nearest T-L targets are located approximately 5 feet offset from the cable tray and at about the same height (i.e./14 feet off floor). They are below the ceiling jet and will experience hot gas layer exposure only. The Thermo-Lag targets considered in this scenario are configuration 2. The heat release rates considered for this scenario are as follows (by fire source):

1. electrical panel with non qualified cable - 400 Bru/s (Ref. 3); this HRR is applicable to scenario time t=0-15 minutes;

2. cable tray -- 23.4 Btu /s-ft (Ref. 2, FIVE Table IE, cable #5 adjusted to consider Equation 2 on FIVE manual, page 10.4-10); since 4 ft' are initially burning, the HRR 1

for this source alone is 94 Btu /s; this HRR will increase with time, as fire propagates horizontally along the tray (i.e.,6 minutes into the fire, the HRR for this source will be 140 Btu /s, which includes the HRR from 2 ft' of additional cable tray). i 8

SAIC Report No. 010082-05-2965-700-1 The tot' l available heat'(Q,,) from each of the two fire sour' es is: a c

1. electrical panel - 360,000 Btus (it assumes that the panel contains enough fuel to sustain a 15 minute fire);

2. cable tray - 2,000,000 Btus [it assumes 10,000 Btu /lb. of cable tray (per Ref. 5, page E2-1) x 10 linear feet x 20 lb./ linear foot (engmeering judgment for 50% filled, 24 inch wide cable tray)]. This is enough combustible material for one hour.

Results The results are presented in Table 5. The time temperature history shows that the temperature at the target will not increase appreciably, reaching only 202*F. A sample model worksheet is provided as Appendix A, Figure A-5. Table 5. Temperature-Time History Fire Scenario Modeled in Configuration 2, Scenario 2. Time HRR to HRR to HGL QTOT Temperature Temperature Target (min.) (Stu/s) (Stu/s) (Blu) Target - (F) HGL - (F) 1 0 494 494 0 100 100 6 540 540 194544 112 112 12 587 587 405936 125 125 18 234 234 490176 131 131 24 281 281 591264 137 137 30 328 328 709200 145 145 36 374 374 843984 154 154 42 421 421 995616 164 164 48 468 468 1164096 175 175 54 515 515 1349424 188 188 60 562 562 1551600 202 202 ) . 4.6 Thermo-Lac Confieuration 2. Scenario 3 This fire scenario modeled a fixed ignition sourcc/ combustible, a 6 inch wide cable tray with non IEEE-383 cable,18 feet off the room's floor. Since the cable is non-qualified, - the cable fire can self-ignite (per Ref. 2, Appendix I-A) and start to burn. The fire will then propagate horizontally along the exposed cable tray with a flame spread rate of 10 feet per hour (per Ref.11). The total amount of cable tray assumed to be available for this fire scenario is 10 feet (for a 1 hour fire duration). The cable tray is only about 25% filled (visual observation during plant walkdown). The targets are two 2-inch diameter conduits wrapped in 0.5 inch thick Thermo-Lag. The T-L wrapped conduits are routed in the same direction as the cable tray,8 inches above it. Due to the close distance between the cable tray and T-L wrapped conduits, the T-L will be ignited by the cable tray fire (assumed conservatively to happen at time t=0 minutes). Once this happens, additional heat is released into this Turbine Building Basement fire zone. 9 j w ...m 3-

-~ SAIC Report No. 010082 05 2965-700-1 The heat release rates considered for this scenario are as follows (by. fire source):

1. cable tray - 23.4 Btu /s-ft (Ref. 3, FIVE Table IE, cable #5 adjusted to consider Equation 2 on FIVE manual, page 10.4-10). Two linear feet of the 6" wide tray is assumed to burn, so a heat release rate of 23 Btu /s is used. This HRR will increase with time, as fire propagates horizontally along the tray (i.e.,6 minutes into the fire, the HRR for this source will be 35 Btu /s, which includes the HRR from 0.5 ft' of additional cable tray);

2. T-L wrap HRR is 8.8 Btu /s-ft (per Ref. 2, Appendix I-D). The amount of Thermo-Lag assumed to burn is that portion of the Thermo-Lag directly above the burning cable tray. The circumference of'the Thermo-Lag wrapped conduit is 1.6 feet, hence, the HRR for the Thermo-Lag is 1.6 ft.

• I ft.
• 8.8 Btu /s/ft.2 =

14 Btu / linear foot. Two feet will be burning initially plus an additional foot every 6 minutes This HRR will be considered only for calculating the HGL temperature rise .The total available heat (Q,,,) from each fire source is:

1. first cable tray - 270,000 Brus [ assumes 10,000 Btu /lb. of cable tray (per Ref. 5, page E2-1) x 12 linear feet x 3 lb./ linear foot (engineering judgment for 25%

filled,6-inch wide cable tray)

• 70% combustion);

i

2. Thermo-Lag wrap - 706,000 Btus (assumes 58,800 Btu / linear foot

• 12 feet, from Ref. 7).

i Results I Table 6. Temperature-Time History Fire Scenario Modeled in Configuration 2, Scenario 3. Time HRR to HRR to HGL QTOT. Temperature Temperature Target (min) (BTU /s) (BTU /s) (BTU) Target - (F) HGL - (F) 0 23 52 0 1600 100 5 35 77 23202 1600 104 10 47-103 54138 1600 110 15 59 129 92808 1600 117 l 16 70 155 102089 1600 119 25 82 180 199537 1600 137 30 94 206 261409 1600 150 / 35 105 232 331015 1600 164 40 117 258 408355 1600 179 50 129 284 578503 1600 216 60 140 309 764119 1600 258 10 l

=. - ' j SAIC Report No. 010082-05-2965-700-1 . The results are presented in Table 6. The T-L wrapped target is at flame temperature for the entire scenario duration. The hot gas layer will reach approximately 234'F. A ' t sample model worksheet is provided as Appendix A, Figure A-6. l 4.7 Thermo-I ae Confieuratirm 2 Scenario 4 This fire scenario modeled a transient ignition source / combustible, a 5 gallon container i containing lube oil. Per Oyster Creek Procedure " Control of Combustibles," Procedure No.120.5, Rev. 4 (Ref. 6), combustible liquids could be transported throughout the { plant in an approved closed metal container (but not necessarily a safety-can). For this fire scenario,it was postulated that 1 quart oflube oil is spilled out of a can on the floor and is not cleaned up. The oil is later ignited by a welder working nearby or another transient ignition source and begins to burn. The oil ~ fire could potentially impact a T-L l wrap 14. feet above the ground. The heat release rate for the lube oil is 110 Btu /s-ft' (from Ref. 3, Table 2E). The oil cohld spread unconfined over an area of 13.5 ft' (54 ft.*/ gal from Ref. 3, Table 3E). The heat release rate assumed for this fire scenario is 1,485 Btu /s (110 Btu /s-ft' x 13.5 ft'). The heat content of one quart of oil is (17,111 Btu /lb.

• 60 lb./1 ft'.
• 1 ft.*/30 quarts) 34,222 Btu (from Ref. 3, Table 2E). The oil fire duration will be less than one minute, but one minute will be used for conservatism. The virtual surface of the fire is at the floor level. The location factor for this source is one.

Results The short oil fire does not cause an exposure temperature high enough to ignite the i Thermo-Lag. The fire is of a short duration, with almost no hot gas layer effects. A sample model worksheet is provided as Appendix A, Figure A-7. Table 7. Temperature-Time History for Fire Scenario Modeled in Configuration 2, Scenario 4. Time HRR to HRR to HGL QTOT Temperature Temperature Target (min.) (Btu /s) (Blu/s) (Blu) Target - (F) HGL - (F) 0 1485 1485 0 964 100 1 1485 1485 89100 965 102 2 0 0 89100 102 102 10 0 0 89100 102 102 15 0 0 89100 102 102 25 0 0 89100 102 102 30 0 0 89100 102 102 35 0 0 89100 102 102 t 40 0 0 89100 102 102 50 0 0 89100 102 102 60 0 0 89100 102 102 11 y

SAIC Report No. 01-0082-05-2965-700-1 5. FIRE MODELING

SUMMARY

Seven scenarios were modeled for the two T-L configurations in Oyster Creek Fire Zone TB-FZ-11D.. The analysis results are presented as time-temperature profiles. These results are used in the following section to develop total heat load and then convert to equivalent ASTM E-119 exposures. '6. APPROXIMATING EXPOSURE TIME-HISTORY OF THERMO-LAG SUB-CONFIGURATIONS - THE TOTAL HEAT LOAD CONCEPT Calculation of an exposure time-temperature profile for a specific T-L Sub-configuration is based on a the methodology presented in Reference 2. The time - temperature profiles are used to develop the total heat load (Reference 2, Appendix E-A) for each fire exposure. The total heat load concept holds that the area under the incident heat flux vs. time curve, or total heat load, can be used as a measure of fire severity'. Comparing the total heat load of a fire scenario exposure to the total heat load of a test exposure (ASTM E-119) allows one to predict the response of a Thermo-Lag barrier to the scenario exposure. The total heat load per unit area of fire barrier surface is equal to the area under the incident heat flux-time curve, corrected for convection effects as described in Reference 2, Appendix E-A. 6.1 Total Heat Load Results Raceway Scenario Description Peak Temperature Total Heat Load configurat at Target (*F) Equivalent E-119 ion Exposure Duration (min.) 1 1 Protective Clothing 1600 >60 1 2 Electrical Cabinet and 322 10 Cable Tray 1 3 Transformer 1278 40 2 1 Cable Trav 1600 >60 2 2 Electrical Cabinet and 202 10 Cable Tray 2 3 Cable Tray 1600 >60 2 4 Oil transient 965 10 Appendix C contains tables with Total Heat Load values calculated for Thermo-Lag sub-configurations 1, and 2. 7. REFERENCES 1. Oyster Creek Plant Drawing of Turbine Building Basement. 2. EPRI Report " Methods for Evaluation of Cable Wrap Fire Barrier Performance", July,1995 12

.. ~...-- SAIC Report No. 01@82 05-2965-700-1 3. EPRI TR-100370, "-Fire-Induced Vulnerability Evaluation," April 1992. 4. GPUN, " Fire Hazard Analysis Report - Oyster Creek," Doc. Non. 990-1746 Rev. 8. 5. GPUN - TMI-1 Procedure No.1035 (Revision 24), August 23,1994. 6. GPUN-Oyster Creek Procedure No.120.5, Rev. 5, " Control Of Combustibles," May 10,1993. 7. Thermo-Lag 330-1 Combustibility Study, NEI,1994. j = 8. " Methods for Evaluating Cable Wrap Fire Barrier Performance," EPRI draft report developed under the Tailored Collaboration Project, October 1994. i 9.- EPRI TR-100443, " Methods of Quantitative Fire Hazard Analysis," EPRI final report, May 1992. 10. EPRI NP-7332, " Design Guide for Fire Protection of Grouped Electrical Cables," EPRI Final Report, May 1991. g 4 I + e f 9 13 ~

SAIC Report No. 01-0082-05-2965-700-1 Appendix A Sample Scenario Worksheets m l

. -. - - -... ~ -..... _ -... - - - -. -. - -. - _ ~.. Figure A-1 SAIC Report No. 01008245-2965-700-1 4 I Temperature calculation at time min 5 Maximum ambient temperature F 100 Floor Area ft2 9668 1 Location of Target Plume 2 Height of Target above Fire Source ft 2.00 3 Height from Fire Source to Ceiling, H ft 17.00 4 4 Ratio of Target Height / Ceiling Height 0.118 5 Longitudinal Distance from Fire Source to ft NA ,i Tarnet. L 6 Longitudinal Distance to Height Ratio, L/H NA 7 Enclosure Width, W ft NA 8 Height to Width Ratio, H/W NA 9 Peak Fire intensity Btu /s 285 10 Fire Location Factor 1 9 11 Effective Fire intensity Stu/s 285 12 Plume Temperature Rise at Target F 1600 13 Temp Rise Factor at Target 1.00 14 Temperature Rise at Target F -~ 1600 16 Temperature at Target F 1800 16 Temperature in HGL F 102 17 Otot to HGL Btu 98100 18 Estimated Heat Loss Fraction 0.94 19 Calculated Onet Stu 5886 20 Calculated Enclosure Volume, V ft3 - 164356 21 Calculated OnetN Onet/ft3 0.04 22 HGL Temperature increase F 2 9 e een-- w

~ _. _ _. _ ~... _ _ _ Figura A 2 SAIC Report No. 010082-05-2965 700-1 Ternperature calculation at time rnin 5 Maximum ambient temperature F-100 Floor Area ft2 9668 1 Location of Target Hot Gas Layer 2 Height of Target above Fire Source ft 5.00 3 Height from Fire Source to Ceiling, H ft 8.00 4 Ratio of Target Height / Ceiling Height 0.625 5 Longitudinal Distance from Fire Source to ft NA Taraet. L 6 Longitudinal Distance to Height Ratio, L/H NA 7 Enclosse Width, W ft NA 8 Height 1o Width Ratio, H/W NA ~ 9 Peak Fire intensity Stu/s 774 10 Fire Location Factor 2 11 Effectivo Fire intensity Stu/s 1548 12 Plume Ternperature Rise at Target F 1600 13 Temp Riae Factorat Target 0.00 14 Temperature Rise at Target F 0 15 Temperature at Target 'F 111 16 Temperature in HGL F 111 17 Otot to HGL Btu 232260 18 Estimated Heat Loss Fraction 0.94 19 Calculated Onet . Blu 13936 20 Calculated linclosure Volume, V ft3 77344 21 Calculated OnetN Onst/ft3 0.18 22 HGL Tempe ature increase F e

Figuro A 3 SAIC Report No. 01-0082 2965-700-1 e Temperature calculation at time min 5 Maximum amtnent temperature F 100_ Floor Area ft2 9668 l 1 Location of Target Ceiling Jet 2 Height of Target above Fire Source ft 19.00 3 Height from Fire Source to CeBng, H ft 20.00 4 Ratio of Target Height /Colling Height 0.950 5 Longitudinal Distance from Fire Source to ft 15 Taroot. L 6 Longitudinal Datance to Height Ratio, UH 0.75 7 Enclosure Width, W ft 112 8 Height to Width Ratio, H/W 0.18 9 Peak Fire intensity Stu/s 13300 10 Fire Location Factor 1 11 Effective Fire intensity Stu/s 13300 12 F;Jme Temperature Rise at Target F. 1411 13 Temp Rise Factor.at Target 0.36 14 Temperature Rise at Target F_ 513 15 Temperature at Target F. _ 800 16 Temperature in HGL F 177 _ 17 Otot to HGL Btu 3990000 18 Estimated Heat Loss Fraction 0.94 19 . Calculated Onet Stu 239400 20 Calculated Enclosure Volume, V ft3 193360 21 Calculated Onet/V Onet/ft3 1.24 22 HGL Temperature increase F p 1 4 )

Rgure A-4 SAIC Report No. 09-0082-05-2965-700-1 Temperature calculaton at time min 6 Maximum ambient temperature F 100 Floor Area ft2 9668 1 Location of Target Plume 2 Height of Target above Fire Source ft 2.90 3 Height from Fire Source to Colling, H ft 3.00 4 Ratio of Target Height / Ceiling Height 0.967 5 Longitudinal Distance from Fire Source to ft NA Tarnet. L 6 Longitudinal Distance to Height Ratio, IJH. NA 7 Enclosure Width, W ft NA 8 Height to Width Ratio, H/W NA 9 Peak Fire intensity Sturs - 375 10 Fire Location Factor 1 11 Effective Fire intensity Stu/s 375 12 Plume Temperature Rise at Target F 1600 13 Temp Rise Factorat Target 1.00 14 Temperature Rise at Target F 1600 15 Temperature at Target F 1800 ~ 16 Temperature in HGL F 123 17 Otot to HGL Blu 185616 18 Estimated Heat Loss Fraction 0.94 19 Calculated Onet Stu 11137 20 Calculated Enclosure Volume, V ft3 29004 21 Calculated OneW Onet/ft3 0.38 22 HGL Temperature increase F 23 I l 4 ~ e O

i Rgure A-5 SAIC Report No. 01-0082-05 2965-7001 Temperature calculation at time min 6 Maximum ambient temperature F 100 Floor Area ft2 9668 i 1 Location of Target Hot Gas Layer 2 Height of Target above Fire Source ft 0.00 3 Height from Fire Source to Colling, H ft 6.00 4 Ratio of Target Height /Colling Height 0.000 5 Longitudinal Distance from Fire Source to ft NA Tarnet. L 6 Longitudinal Distance to Height Ratio, L/H NA 7 Enclosure Width, W ft NA 8 Height to Width Ratio, H/W NA ~ 9 Peak Fire intensity Stu/s 540 10 Fire Location Factor 1 11 Effective Fire Intensity Stu/s 540 12 Plume Temperature Rise at Target F 0 13 Temp Rise Factorat Target 0.00 14 Temperature Rise at Target F 0 15 Temperature at Target F 112 16 Temperature in HGL F 112 17 Otot to HGL Blu 194544 18 Estimated Heat Loss Fraction 0.94 19 Calculated Onet Stu 11673 20 Calculated Enclosure Volume, V ft3 58008 21 Calculated Onet/V Onet#t3 0.20 22 HGL Temperature increase F 12 l d G

i Figure A 6 SAIC Report No. 01-0082-05-2965-7001 Temperature Calculaton at time min 5 Maximum ambient temperature F 100 Floor Area ft2 9668 l 1 Location of Target Plume. ] 2 Height of Target above Rre Source ft 0.66 3 Height from Fire Source to Colling, H ft 2.00 ) 4 Ratio of Target Height /Celhng Height 0.330 5 L ongitudinal Distance from Rre Source to ft NA Timet. L j 6 Longitudmal Datance to Height Ratio L/H - NA 7 Enclosure Width, W ft NA 8 Heignt to Width Ratio, H/W NA 9 Peak Fire intensity Blu/s 35 10 Fire Location Factor 1 i i 11 Effective Fire intensity Stu/s 35 12 Plume Temperature Rise at Target F 1600 13 Temp Rise Factor at Target 1.00 14 Temperature Rise at Target F 1600 15 Temperature at Target F 1600 16 Temperature in HGL F 104 17 Otot to HGL Btu 23202 18 Estimated Heat Loss Fraction 0.94 19 Calculated Onet Blu 1392 20 Calculated Enclosure Volume, V ft3 19336 21 Calculated Onet/V Onet/ft3 0.07 22 HGL Temperature increase F 4 l 4 ---..v. ---v..

.. _~ - ~. - -.....-...... j Figure A-7 SAIC Report No. 01-0082 05 2965 700-1 i Temperature Calculaten at time rnin 1 l Maximum ambient temperature F 100 j Floor Area ft2 9668 l 1 Location of Target Plume 2 Height of Target above Fire Source ft 14.00 k 3 Height from Fire Source to Colling H ft 20.00 i 4 Ratio of Target Height /Colling Height 0.700 5 Longitudinal Distance from Fire Source to ft NA Tarnet. L 6 Longitudinal Datance to Height Ratio, L/H. NA 7 Enclosure width, W ft NA 8 Height to Width Ratio, H/W NA l 9 Peak Fire intensity Stu/s 1485 10 Fire Location Factor 2 i 11 Effective Fire intensay Btu /s 2970 12 Plume Temperature Rise at Target F 864 13 Temp Rise Factor at Target 1.00 14 Temperature Rise at Target F 864 l 15 Temperature at Target F 986 16 Temperature in HGL F 102 17 Otot to HGL Blu 89100 18 Estimated Heat Loss Fraction 0.94 19 Calculated Onet Blu 5346 20 Calculated Enclosure Volume V ft3 193360 21 Calculated Onet/V Onet/ft3 0.03 22 HGL Temperature increase F 2 e +e.. n

i SAIC Report No. 01-0082-05-2965-7001 Appendix B Basis for Following ASTM E-119 Time-Temperature Relationship When T(HGL) is 1000 *F 4 2 I 1 O e

SAIC Report No. 01-0082-05-2965-700-1 TRANSITION TO ASTM E-119 TIME-HISTORY IN THE HOT GAS LAYER ISSUE: The Fire Hazard Tool currently under development in the EPRI Cable Wrap Fire i Barrier Tailored Collaboration assumes that the fire compartment temperature will follow the ASTM E-119 time-temperature relationship as soon as the compartment hot gas layer reaches 1000*F. ) BASIS: The Handbook (Bel, pg. 6-75) states that "Flashover of an enclosure is likely to ' occur if the temperature of the upper gas layer reaches approximately 1000*F (600 C). The EPRI assumption of 1000*F is therefore conservative. As stated near the top of page 6-67 of the Handbook, the intensity of the fire will be somewhat lower when the walls and ceiling absorb significant amounts of energy rather than when act as insulation or radiation barriers. Since power plant compartments are typically concrete barriers which absorb significant amounts of heat, the intensity of fire will tend to be less severe than fires in compartments with insulated barriers. ASTM E-119 has been the accepted standard for testing fire barrier systems since the early 1900's. The standard, which was developed from actual fire tests, has been used to evaluate the fire endurance capabilities of fire barriers since it was developed. The time-temperature relationship represented by the curve represents a fully-involved fire compartment. According to Fred Mowrer of the University of Maryland, plume and ceiling effects disappear in a fully involved fire compartment as the plume / ceiling jet gas are no longer more buoyant than surrounding gases. Use of the ASTM E-119 time-temperature relationship to represent a fully involved power plant compartment fire is therefore consistent with industry practice. Assuming that that occurs when the hot gas layer reaches 1000'F is more conservative than the 1100'F noted in the Handbook.

Reference:

NFPA Fire Protection Handbook, Seventeenth Edition, Section 6/ Chapter 6. --. m

SAIC Report No. 01-0082-05-2965-700-1 Appendix C Total Heat Load Results S \\ i I e l

Figurs C-1 SAIC Report 01-0082 05-2965 700-1 Total Heat Load Scenario 1 Subconfio.1 Toroet Hot Gas Layer ASTM E-119 Time Temperature inc Heat Flux Total Heat Lood Total Heat Load Total Heat load (min) (F) (Btu /s sf) (1000 Btu /sf) (10008tu/sf) (1000 Btu /sf) 0 1600 9.5 0.0 0.00 0.00 5 1600 9.5 2.84 0.09 0.34 10 1600 9.5 5.68 0.16 1.37 15 1600 9.5 8.52 0.23 2.94 20 1600 9.5 11.36 0.30 4.81 25 1600 9.5 1420 0.37 6.90 30 1600 9.5 17.05 0.45 9.19 40 1600 9.5 22.73 0.60 --~~ 14.25 47 1600 9.5 26.71 0.70 18.10 48 1600 9.5 27.27 - 0.72 18.66 60 1600 9.5 34.09 0.91 25.84 O e O 4 a e 9 e

Figure C-2 SAIC Report 010082-05-2965 700-1 TotalHeat Load Scenario 2 Subconfia.1 Target Hot Gos Loyer ASTM E-119 Time Temperature inc Heat Fluy Total Heat Load TotalHeat Lood Total Heat Load (min) (F) (8tu/s-sf) (10008tu/sf) (1000 Btu /sf) (1000 Btu /sf) 0 100 0.4 0.0 0.00 0.00 5 111 0.2 0.09 0.09 0.34 10 123 0.3 0.17 0.17 1.37 15 136 0.3 0.25 0.25 2.94 16 138 0.3 0.27 0.27 3.30 25 167 0.4 0.44 0.44 6.88 184 0.4 0.55 0.55 9.17 30 __203 0.3 0.66 0.66 11.62 35 40 224 0.3 0.76 0.76 14.24 50 270 0.5 1.00 1.00 19.87 l 60 322 0.5 1.27 1.27 25.97 i T = ,rw-~\\, e

Figure C 3 SAIC Report 01-0082-05-2965-700-1 Total Heat Load Scenario 3 Subconfio.1 Totoet Hot Gas Loyer ASTM E 119 Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load i (min) (F) (Btu /s sf) (1000 Btu /sf) (1000 Btu /sf) (1000 Btu /sf) 0 613 1.0 0.0 0.00 0.00 5 690 1.3 0.35 0.11 0.34 1 10 778 1.6 0.79 0.23 1.37 15 879 2.1 1.34 0.38 2.94 20 993 2.7 2.06 0.58 4.81 22 1044 3.0 2.41 0.68 5.60 30 1278 5.1 4.34 1.25 9.10 31 1278 5.1 4.64. 1.34 9.57 40 12~78 5.1 7.38 2.19 14.13 50 1278 5.1 10.41 3.15 19.76 1 60 1278 5.1 13.44 4.14 25.87 \\ i 1 l i e f

• • -my.

r

Figure C-4 SAIC Report 01-0082 05-2965 7001 i Total Heat Load Scenario 1 Subconfio. 2 Toroet Hot Gas Layer ASTM E-119 Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min) (F) (Btu /s-st) (1000 Btu /sf) (1000 Btu /sf) (1000 Btu /sf) 0 1600 9.5 0.0 0.00 0.00 6 1600 9.5 3.41 0.11 0.48 12 1600 9.5 6.82 0.22 1.87 18 1600 9.5 10.23 0.35 3.92 24 1600 9.5 13.64 0.48 6 34 30 1600 9.5 17.05 0.63 9.05 36 1600 9.5 20.46 0.80 12.02 42 1600 9.5 23.86 0.99 15.21 48 1600 9.5 27.27 ~ 1.23 18.58 54 1600 9.5 30.68 1.55 22.14 60 1600 9.5 34.09 1.96 25.86 0 4 .-w

Figure C-5 SAIC Report 010082-05-2965 700-1 TotalHeat Load Scenario 2 Subconfic. 2 Torget Hot Gas Layer ASTM E-119 Tme Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min) (F) (Btu /s-sf) (1000 Btu /sf) (10008tu/sf) (1000 Btu /sf) 0 100 0.4 0.0 0.00 0.00 6 112 0.2 0.11 0.11 0.48 12 125 0.3 0.20 0.20 1.87 18 131-0.3 0.30 0.30 3.92 24 137 0.3 0.40 0.40 6.34 30 145 0.3 0.51 0.51 9.05 36 154 0.3 0.62 0.62 12.02 0.74 15.21 42 164 0.3 0.74 48 175 0.4 0.87 0.87 18.58 54 188 0.4 1.01 1.01 22.14 60 202 0.3 1.14 1.14 25.86 O e G 4 e e

Figure C 6 CAIC Report 010082-05-2965 700-1 TotalHeat Load Scenario 3 Subconfio. 2 Toroet Hot Gas Layer ASTM E-119 Time Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load (min) (F) (Stu/s-af) (1000Blu/sf) (1000 Btu /sf) (1000 Btu /sf) 0 1600 9.5 0.0 0.00 0.00 5 1600 9.5 2.84 0.09 0.34 10 1600 9.5 5.68 0.16 1.37 15 1@ 9.5 8.52 0.24 2.94 16 1600 9.5 9.00 0.25 3.30 25 16G) 9.5 14.20 0.40 6.88 30 1600 9.5 17.05 0.49 9.17 35 1600 9.5 19.89 0.59 11.62 1 40 1600: 9.5 22.73 0.70 14.24 50 1600 9.5 28.41 - 0.91 19.87 60 1600 9.5 34.09 1.14 25.97 9 0 f e 0 e a .m, p-

.i. Figure C 7 SAIC Report 01-0082 05-2965-700-1 TotalHeat Load Scenario 4 Subconfio. 2 Toroet Hot Gas Loyer ASTM E 119 4 Tme Temperature inc Heat Rux Total Heat Load Total Heat Load Total Heat Load j (min) (F) (Blu/s-sf) (1000Blu/sf) (10008tu/s0 (10008tu/s0 0 964 2.5 0.0 0.00 0.00 j 1 965 2.5 0.15 0.02 0.00 2 102 02 023 0.03 0.02 10 102 02 0.34 0.14 1.23 15 102 02 0.41 0.21 2.80 25 102 0.2 0.55 0.35 6.73 30 102 02 0.62 0.42 9.02 0.48 11.47 ) 35 102 02 0.69 40 102 02 0.76 0.55 14.09 50 102 02 0.89 0.69 19.72 l 60 102 0.2 1.03 0.83 25.83 4 1 i I i e + A f 4 4 a i l i, s e l J

e TA lo 2. APPEN91x E_ Using FIVE Fire Modeling to Approximate Exposure Time-History for Thermo-Lag Configurations in OC Fire Zone TB-FZ-11C Letter Report september as, sees 7 Prepared by: GPUN Risk Analysis Group Using FIVE Fire Modeling to Approximate Exposure Time-History for Thermo-Lag Configurations l I

\\ ~ i V 1. PURPOSE The purpose of this report is to document the detailed Are modeling performed for fire scenarios involving Thermo-Lag wrapped conduit and armored cable wrapped in Thermo Lag, as identified during plant walkdowns of OC fire zone TB-FZ-11C. Thermo-Lag exposure time-history results were obtained for each T-L configuration determined to be impacted by a fire source / combustible. j 2. APPROACH The approach followed for modeling of fire scenarios is based on methodology detailed in the EPRI " Methods for Evaluation of Cable Wrap Fire Barrier Performance" document (Ref.1) which was based on Are modeling techniques developed for FNE (Ref. 2). Fire modeling was accomplished in two steps:

1. Screening walkdowns to eliminate from further consideration those fire ignition sources and potential fire scenarios that cannot develop into damaging fires.

2. Fire modeling of specific scenarios that were not screened out in the first step.

For each Thermo-Lag configuration analyzed, one or more fire scenarios (including bounding scenarios) were postulated based on the following: Information gathered during plant walk' downs (i.e., fixed or transient ignition a. source and Thermo-Lag target location, type and quantity of combustibles present in zone); b. information obtained from reviewing applicable sections in Appendix R documents for OC (Ref. 3); c. Informati7n obtained from reviewing OC procedure 120.5, Rev. 5, " Control of Combustibles." (Ref. 4) d. Informati)n provided by GPUN engineers (specializing in Are protection, risk assessmet, mechanical). i Thermo-Lag FNE analyses were performed using computerized Excel spreadsheets which duplicate FNE Worksheets 1,2, or 3 and are' linked to tables containing the same information as in FNE Reference Tables. For each Thermo-Lag fire modeling scenario, results are presented as exposure time-temperature profiles. In all scenarios evaluated, time-dependent tempsiatures were derived by incrementally adding the total heat content of the fuel into the fire compartment hot gas layer. 3. FIRE MODEUNG ASSUMPTIONS / The following Thermo-Lag fire modeling assumptions were made: 1. Worst case fixed fire source scenarios were modeled for each Thermo-Lag sub-configuration. The determination of which fixed fire source might cause the worst-1 J <u- .,wm-g -im9 Y

. - - _~ -. - -.. -. _ -. _ - -- case Thermo-Lag fire scenario was made during the plant Are area walkdown and i was based on minimum damage threshold heights and distances calculated for j typical fixed fire sources (e.g., electrical cabinets, electrical motors, lube oil in pumps) found at Oyster Creek. 2. Due to the location of the T-L wrap in this fire zone (on top of switchgears 10 & 1D), no transient combustible scenarios were noted during walkdowns or postulated as plausible. 3. To ensure conservative results, worst-case Are plume (or ceiling jet) scenarios j and/or radiant flux scenario were first modeled. For T-L sub. configurations not in ) the plume or ceiling jet, hot gas layer Are scenarios were evaluated. 1 4. Electrical cable inside cabinets, motor control centers, switchgear and cable trays 1 was assumed to be norxtuallflod (i.e., non IEEE-383 type). This assumption is conservative since electrical cable purchased and installed at Oyster Creek in the j past decade was IEEE-383 type. 5. The Fire Hazard Tool developed for the EPRI Cable Wrap Fire Barrier Tallored Collaboration assumes that the fire compartment temperature will follow the ASTM E-119 time-temperature curve as soon as the compartment hot gas layer (HGL) reaches 1000*F. When the HGL temperature reaches 1000'F, the plume (or ceiling jet) temperatures are assumed to follow that of the E-119 (see discussion in Appendix B). 6. The Thermo-Lag wrap is assumed to bum by itself (i.e., even after the initial fire source is out of fuel) unless otherwise stated. Per, the EPRI Method.(Ref.1, Appendix l-D), credit may be taken for a time lag before the Thermo-Lag ignites based on exposure temperature. 7. The maximum plume and ceiling jet temperature is assumed to be 1600'F, which corresponds to flame temperature. This assumption is valid until hot gas layer temperature exceeds 1600"F, at which point the plume temperature is assumed to be equal to hot gas layer temperature, 4. FIRE MODELING RESULTS 2 4.1 Fire Zone TB-FZ-11C Fire zone TB-FZ-11 C is the Turbine Building Switchgear Room, West End of the Mezzanine Level. There is only one area in this fire zone that contains the 3-hr rated fire barrier T-L wrap condult. This T-L wrap conduit sits on top of switchgears 10 & 1D and it is designated as location No. 13 on the general arrangement drawing shown in Appendix D. This fire zone has. unprotected' openings through the floor to fire zone TB-FZ 11D and through the walls except the north wall. The north wall adjoins the fire areas TB-FA-3A and TB-FA-38 / which contain safety related 4160V switchgears 1C & 1D and are separated from this zone by / 3-hour fire rated roll-up doors. Also, Fire area TB-FA-26, Battery Room, is enveloped with three hour fire resistive rated barriers within this zone. Due to the location of the T-L wrap in this fire zone (on top of switchgears 10 & 1D), no transient combustible scenarios were noted during walkdowns or postulated as plausible. Based on fire i 2 + - -

modeling experience for other fire zones, even a quart of oil spill later ignited lasts less than one minute (see Fire Zone RB-FZ-1E, Location No.10, Scenario # 1). Ref. 5, Attachment 3, Table 13 indicates that even at maximum plume temperature of 1600*F, it takes about one minute for the Thermo-lag to reach the ignition temperature. Therefore, this transient combustible scenario will be insignificant. However, two fixed ignition sources were modeled for this zone. It should also be noted that even though, there is hydrogen seal oil in this fire zone, it is only in piping that i passes through the zone. Scenario 1 This fire scenario models non-safety related swit2p 1B self-igniting. The T-L wrap is not in the plume of such fire and would only see the hot gas layer region. However, there are three 24-inch wide cable trays stacked on top of each other 1.ft. apart in the plume of such fire (directly over the back vents of the switchgear 1B, about 4 ft. above, and run parallel to the west wall along the back of the switchgear). It is assumed that 20 ft of the first cable tray immediately above the switchgear 18 ignites (being in the plume of the ser ;. gear fire). This subsequently leads to ignition of 22 ft. and 24 ft. of the other cable trays above. The time required for these sequential fires to occur is conservatively ignored here. In addition to the vertical propagation, the fire is also assumed to propagate horizontally along the exposed cable tray with a flame speed of 10 feet per hour (per Ref. 2, Appendix l-C). The total amount of cable tray assumed to be available for this fire scenario is 50 feet per tray (for 3 hour fire duration). The heat release rates considered for this scenario are as follows (by fire source): 1. electrical panel (switchgear 18) with nonqualified cable - 400 BTU /sec (Ref.1); this HRR is applicable to scenario time t=0-15 minutes;

2. first cable tray - 23.4 BTU /sec-ft2 (Ref. 2, FIVE Table 1E, cable #5 adjusted to consider

'l Equation 2 on FIVE manual, page 10.410); since 40 ft2 are initially buming, the HRR for this source alone is 936 BTU /sec; this HRR will increase with time, as fire propagates horizontally along the tray (i.e., 6 minutes into the fire, the HRR for this source will be 983 BTU /s, which includes the HRR from 2 ft2 of additional cable tray);

3. second cable tray - 23.4 BTU /sec-ft2 (Ref. 2, FIVE Table 1E, cable #5 adjusted to consider Equation 2 on FIVE manual, page 10.4-10); just as with the first cable tray, the HRR for this source alone is 1030 BTU /sec (since 44 ft2 are initially buming); this HRR will increase with time, as fire propagates horizontally along the tray (i.e.,6 minutes into the fire, the HRR for this source will be 1076 BTU /s, which includes the HRR from 2 ft2 of additional cable tray);

4. third cable tray - 23.4 BTU /sec-ft2 (Ref. 2, FIVE Table 1E, cable #5 adjusted to consider Equation 2 on FIVE manual, page 10.4-10); just as with the second cable tray, the HRR for

/ this source alone is 1123 BTU /sec (since 48 ft2 are initially buming); this HRR will increase / with time, as fire propagates horizontally along the tray (i.e.,8 minutes into the fire, the HRR for this source will be 1170 BTU /s, which includes the HRR from 2 ft2 of additional cable tray), i 3 j

The total available heat (Otot) from each fire source is: 1. electrical panel --360,000 BTUs (it assumes that the panel contains enough fuel to sustain a 15 minute fire); 2. first cable tray -20,000,000 BTUs [it assumes 10,000 BTU /lbm of cable insulation (per Ref. 6, page E2-1) x 50 linear feet x 40 lbm/ linear foot (engineering judgment for 100% filled,24 inch wide cable tray)); 3. second cable tray -20,000,000 BTUs [it assumes 10,000 BTU /lbm of cable insulation (per Ref. 6, page E2-1) x 50 linear feet x 40 lbm/ linear foot (engineering judgment for 100% filled, 24 inch wide cable tray)); 4. third cable tray -20,000,000 BTUs [it assumes 10,000 BTU /lbm of cable insulation (per Ref. 6, page E2-1) x 50 linear feet x 40 lbm/ linear foot (engineering judgment for 100% filled,24 inch wide cable tray)) 1 The floor area for this fire zone is 2,666 square feet (Ref. 3) with a height of about 20 feet (General Arrangement Drawings). It should be noted that this zone contains fire areas TB-FA-3A, TB-FA 3B, and TB FA-26. The floor areas for TB-FA-3A and TB-FA-38 combined is estimated at about 1000 square feet and the floor area for TB-FA-26 is about 250 square feet. The height for these fire areas were estimated at about 10 feet. Results The time temperature profile scenario calculated by the model are presented in Table 1. The model results indicate that hot gas layer effects in the zone due to this postulated fire cause the temperature to exceed 1000*F at time = 19 minutes. It should be noted that these HGL temperatures in this compartment are expected to be very conservative due to the number and size of the openings that exist in this fire zone and the tortuous path that the hot gas needs to take to reach the T L wrap conduit area vs. other direct openings out of this fire zone. Past the 1000'F HGL temperature, it is assumed that a fire compartment temperature will follow the ASTM E-119 time-temperature relationship (Ref.1). Appendix B contains the basis for this assumption. A sample model worksheet is provided in Appendix A, Figure A-1. l 4 j

Table 1. Temperature-Time History for Fire Scenario 1. Time Haa to I HHH lo UIVI Temp. Temp. Target HGL (min.) (Blu/s) (Blu/s) (Blu) Target - (F) HGL - (F) O 3459 3459 0 100 100 5 3529 3tuu 1305440 302 302 12 Jrow wow h==> 559 559 15 3509 3509 7_- > 953 953 19 3532 3532 4135440 1025 1025 25 13Uu-13vu" 50 1DJU" 1DJO" 90 1750" 1750" 120 1535" 1535" 150 15cu-1552" 150 1900" 1900" " Temperatures follow those of AdlNI 1d 119. Scenario 2 All the cable trays in this f;re zone are at a higher elevation than the T-L wrapped condult. However, this scenario models self-ignition of one of these cable trays and evaluates the time-temperature profile in the room for the case when/if the HGL temperature reaches 1000*F and ~ it becomes a fully involved fire. The most bounding configuration for this scenario was found to be three cable trays stacked on top of each other,1 ft. apart, and about 4 ft behind and 2 ft. above the T-L wrap conduit. The cable trays are 24 inches wide it is assumed that 2 ft. of the bottom cable tray self-ignites and this ignites 4 ft. of cable tray'in the middle, and consequently,6 ft. of cable tray on top, it is conservatively assumed that all the cable trays ignition occur instantaneously for calculation simplification. In addition to the vertical propagation, the fire is also assumed to propagate horizontally along the exposed cable tray with a flame speed of 10 feet per hour (per Ref. 2, Appendix l-C). The total amount of cable tray assumed to be available for this fire scenario is 30 feet per tray (for 3 hour fire duration). j The heat release rates considered for this scenario are as follows (by fire source): l

1. first cable tray - 23.4 BTU /sec ft2 (Ref. 2, FNE Table 1E, cable #5 adjusted to consider Equation 2 on FNE manual, page 10.4-10); since 4 tt2 are initially burning, the HRR for this j

source alone is 93.6 BTU /sec; this HRR will increase with time, as fire propagates horizontally along the tray (i.e., 6 minutes into the fire, the HRR for this source will be 140.4 BTU /s, which 5

o. includes the HRR from 2 ft2 of additional cable tray);

2. second cable tray - 23.4 BTU /sec-ft2 (Ref. 2, FIVE Table 1 E, cable #5 adjusted to consider Equation 2 on FIVE manual, page 10.4-10); just as with the first cable tray, the HRR for this source alone is 187 BTU /sec (since 8 ft2 are initially buming); this HRR will increase with time, as fire propagates horizontally along the tray (i.e.,6 minutes into the Are, the HRR for this source will be 234 BTU /s, which includes the HRR from 2 ft2 of additional cable tray);

3. third cable tray - 23.4 BTU /sec-ft2 (Ref. 2, FIVE Table 1E, cable #5 adjusted to consider Equation 2 on FIVE manual, page 10.410); just as with the second cable tray, the HRR for this source alone is 281 BTU /sec (since 12 ft2 are initially burning); this HRR will increase with time, as fire propagates horizontally along the tray (i.e.,6 minutes into the fire, the HRR for this source will be 328 BTU /s, which includes the HRR from 2 ft2 of additional cable tray);

The total available heat (Otot) from each fire source is:

1. first cable tray -12,000,000 BTUs [it assumes 10,000 BTUAbm of cable insulation (per Ref.

6, page E2-1) x 30 linear feet x 40 lbmSinear foot (engineering judgment for 100% filled,24 inch wide cable tray));

2. second cable tray -12,000,000 BTUs [it assumes 10,000 BTUAbm of cable insulation (per Ref.

6, page E21) x 30 linear feet x 40 lbm/ linear foot (engineering judgment for 100% filled,24 inch wide cable tray));

3. third cable tray -12,000,000 BTUs [it assumes 10,000 BTUAbm of cable insulation (per Ref.

6, page E21) x 30 linear feet x 40 lbm/ linear foot (engineering judgment for 100% filled,24 inch wide cable. tray)); i Results ) The time temperature profile scenario calculated by the model are presented in Table 2. The / model results indicate that hot gas layer effects in the zone due to this postulated fire cause the / temperature to exceed 1000"F at time = 34 minutes. It should be noted that these HGL temperatures in this compartment are expected to be very conservative due to the number and 4 size of the openings that exist in this fire zone. 6 4 ,-,w-

c. o. Past the 1000"F HGL temperature, it is assumed that a fire wiii, ... temperature will follow the ASTM E-119 time-temperature relationship (Ref.1). Appendix B contains the basis for this assumption. A sample model worksheet is provided in Appendix A, Figure A-2. Table 2. Temperaturo-Time History for Fire Scenario 2. j l Time H MM to HMM to uIUi Temp. Temp. Target MGL (min.) (Stu/s) (5tu/s) (5tu) Target - (F) HGL - (F) 0 552 552 0 100 100 5 tur 702 znzrzu 171 171 ) 12 542 542 555540 255 255 15 952 952 909350 400 400 24 1122 1122 1314dau 550 550 30 1252 1252 1(o/ovu 525 525 34 1355 1355 2093040 1042 1042 50 1550" 1550" 90 1704" 1704" 150 1550" 1850" 150 1858" 1555" " Temperatures follow those of A3IM E-119. i 5. FIRE MODELING

SUMMARY

Fire modeling was performed for two fixed fire Source-Thermo-lag wrap configurations in Oyster Creek Fire Zone TB FZ-11C. The analysis results are presented as time-temperature profiles. These results are used in the following section to develop total heat load and then corwort to equivalent ASTM E 119 exposures. 6. APPROXIMATING EXPOSURE TIME-HISTORY OF THERMO-LAG SUS-CONFIGURATIONS - THE TOTAL HEAT LOAD CONCEPT Calculation of an exposure time-temperature profile for a specific T-L Sub-configuration is based on the methodology presented in Reference 1. The time -temperature profiles are used to develop the total heat load (Reference 1 Appendix lil A) for each fire exposure. The total heat load concept holds that the area under the incident heat flux vs. time curve, or total / heat load, can be used as a measure of fire severity. Comparing the total host load of a fire Scenarb exposure to the total heat load of a test exposure (ASTM E-119) allows one to predict the response of a Thermo-Lag barrier to the scenario exposure. The total heat load per unit area of fire barrier surface is equal to the area under the incident heat flux-time curve, corrected for convection effects as described in Reference 1, Appendix lil-A. 7

6.1 Total Heat Load Results Location Scenarlo Description Peak Total Heat Load No.(GA Temperature at Equivalent E-119

DWG, Target (T)

Exposure Duration App.D) (min.) is 1 switcngear is ano 1900 > 150, < i so Cable Tray 13 2 Gable Tray 1888 > 150, < 180 Appendix C contains tables with Total Heat Load values calculated for Thermo-Lag scenarios 1, and 2. 7. REFERENCES 1. EPRI Report " Methods for Evaluation of Cable Wrap Fire Barrier Performance", July, 1995 2. EPRI TR 100370, " Fire-induced Vulnerability Evaluation," April 1992. l 3. GPUN, " Fire Hazards Analysis Report - OC," Doc. No. 990-1746, Rev. 8, May 24, 1995. i 4. GPUN - OC Procedure No.120.5 (Revision 5), May 20,1993. 5. Thermo-Lag 330-1 Combustibility Study, NEl,1994. 6. GPUN - TMI-1 Procedure No.1035 (Revision 24), August 23,1994. l ) l / l

l i l Appendix A Sample Scenario Worksheets 6 0 0 1 1 i , / e

_.. _. _. ~. _ _. _ _.. _ _ _ _ _ _ _ _ _. _. _. _. _....... - n. 1 i FGure A-1 1 Temperature calculation at time min 6 Maximum ambient temperature F 100 Floor Area ft2 2666 l l 1 Location of Target Het Gas Layer i 2 Height of Target above Fire source ft 2.00 l 3 Height from Fire Source to Ceiling, H ft 10.00 l 4 Ratio of Target Height / Ceiling Height 0.200 5 Longitudinal Datance from Fire Source to ft NA i Target, L l 6 Longitudinal Distance to Height Ratio, UH NA i 7 Enclosure Width, W ft NA 8 Height to Width Ratio, H/W NA l 9 Peak Fire intensity Stu/s 3629 10 Fire Location Factor 1 11 Effective Fire intensity Stu/s 3629 12 Plume Temperature Rise at Target F 1600 13 Temp Rise Factorat Target 0.00 i e { 14 Temperature Rise at Target F 0 15 Temperature at Target F 302 3 16 Temperature in HGL F 302 17 Qtot to HGL Btu 1306440 1 18 Estimated Heat Loss Fraction 0.94 \\ f 19 Calculated Onet Stu 76386 4 20 Calculated Enclosure Volume, V ft3 26660 21 Calculated QnetN Qnet/ft3 2.94 l 22 HGL Temperature increase F 202 i e

4. Figure A-2 Temperature calculation at time min 6 Maximum ambient temperature F 100 Floor Area A2 2666 1 Location of Target Hot Gas Layer 2 Height of Target above Fire Source R 0.10 3 Height from Fire Source to Ceiling. H A 5.00 4 Ratio of Target Height / Ceiling Height 0.020 5 Longitudinal Distance from Fire Source to ft NA Target, L 6 Longitudinal Distance to Height Ratio, LJH NA 7 Enclosure Width, W R NA 8 Height to Width Ratio, H/W NA 9 Peak Fire intensity Blu/s 702 10 Fire Location Factor 1 11 Effective Fire intensity Btuts 702 12 Plume Temperature Rise at Target F 1600 13 Temp Rise Factor at Target 0.00 l 14 Temperature Rise at Target F 0 15 Temperature at Target F 171 16 Temperature in HOL F 171 17 Otot to HGL Btu 252720 ] 18 Estimated Heat Loss Fraction 0.g4 19 Calculated Qnet Blu 15163 20 Calculated Enclosure Volume, V A3 13330 21 Calculated QnetN Qnot#t3 1.14 22 HGL Temperature increase F 79 l/ i -wyr-y -.e 7 . + -s a.

s v 1 Appendix B Basis for Following ASTM E 119 Time-Temperature Relationship When T(HOL) is 1000 Y e 0 l //

1 I i TRANSITION TO ASTM E 119 TIME-HISTORY IN THE HOT GAS LAYER i ISSUE: The Fire Hazard Tool currently under development in the EPRI Cable Wrap Fire Barrier Tailored Collaboration assumes that the fire compartment temperature will follow the ASTM E-119 time-temperature relationship as soon as the compartment hot gas layer reaches 1000*F. - BASIS: The Handbook (BgL, pg. 6-75) states that "Flashover of an enclosure is likely to occur if the temperature of the upper gas layer reaches approximately 1000*F (600*C). The EPRI assumption of 1000*F is therefore conservative. As stated near the top of page 6-67 of the Handbook, the intensity of the fire will be somewhat lower when the walls and ceiling absorb significant amounts of energy rather than when act as insulation or radiation barriers. Since power plant compartments are typically concrete barriers which absorb significant amounts of heat, the intensity of fire will tend to be less severe than fires in compartments with insulated barriers. ASTM E 119 has been the accepted standard for testing fire barrier systems since the early 1900's. The standard, which was developed from actual fire tests, has been used to evaluate the fire endurance capabilities of fire barriers since it was developed. The time-temperature relationship represented by the curve represents a fully-involved fire compartment. According to Fred Mowrer of the University of Maryland, plume and ceiling effects disappear in a fully involved fire compartment as the plume / ceiling jet gas are no longer more buoyant than surrounding gases. Use of the ASTM E-119 time-temperature relationship to represent a fully involved power plant compartment fire is therefore consistent with industry practice. Assuming that that occurs when the hot gas layer reaches 10007 is more conservative than the 11007 noted in the Handbook.

Reference:

NFPA Fire Protection Handbook, Seventeenth Edition, Section 6/ Chapter 6. 4

s-e s Appendix C Total Heat Load Results E e / ___m_.______.._.

_ -. -.. ~... 3. @ m 41 e Total Heat Load Scenario 1 Target Hot Gas Layer ASTM E-119 Time Temperature inc Heat Flux Total Heat Load Total Host Load Total Heat Load (min) (F) (Stu/s-s0 (10008tu/s0 (1000Bluts0 (1000 Btu /s0 0 100 0.4 0.0 0.00 0.00 6 302 0.4 0.14 0.14 0.41 12 589 1.0 0.40 0.40 1.65 18 953 2.4 1.02 1.02 3.56 19 1025 2.9 1.18 1.18 3.92 25 1300 5.3 2.65 2.65 6.31 60 1638 10.1 18.81 18.81 25.06 90 1750 12.3 38.93 38.93 45.76 120 1835 14.2 62.74 62.74 69.38 150 1862 14.8 88.85 88.85 95.02 180 1900 15.8 116.46 116.46 122.34 b O P I / ,c

. _.. _ ~ _ _ _. 4 Figure C-2 Total Heat 1. cad Scenario 2 TarDat Hot Gas Layer ASTM E-110 Time, Temperature inc Heat Flux Total Heat Load Total Heat Load Total Heat Load l (min) (F) (Blu/s-sf) (1000 Btu /s0 (1000 Btu /sf) (1000 Btu /s0 1 0 100 0.4 0.0 0.00 0.00 6 171 0.4 0.13 0.13 0.41 12 268 0.4 '53 0.28 1.65 18 400 0.7 0.47 0.47 3.56 24 580 1.0 0.77 0.77 5.93 I 30 828 1.8 1.27 1.27 8.64 I 34 1042 3.0 1.84 1.84 10.54 60 1550 8.6 10.87 10.87 24.95 90 1704 11.3 28.76 28.76 45.65 1 150 1850 14.5 75.25 75.25 94.54 180 1888 15.5 102.30 102.30 121.87 l i i e

s Appendix D Ceneral Arrangement Drawings 1 e e O 6 e 4 e

_ ~ _ _ e p L o a l l \\ L A-PLANT T^ le0RTM k .a k ~ i O acATdM t .. [ T li AAA [ % 1% ,t i / 4 me M[ 2,,y i i ( i.,g. I rrr r- '96,,, R b.... i u i / u_t bt~ L < p: Ls-- 3 g-w d-t EE,',S..y - - - ~ -.i l 1 h j' U 6' g s . 4 4 -- Ema N -9.1w::=;u. [.Y 'i U C_[l i i t

i h --

g, k; --M =3b-e v__ f m ii ~L_ W ) - g---.. gA7Taf n. ~~, w,. g' r_=,d.)'G.5N, E ~ ?-> f . g: i . =. - __ y a ~- 7 ~/< = = 'l-t Ni 6 -~~

ii;;;:;;-

i y he fi _. c if __ I r p . =4 ag: l: t

;. a r E f Q--

i,. J::g\\,,, ~;;.: = C 1.N E -~ .l 0 l .io -imai I l 1 t I ar t8 '1 g Co .,i i. /_.- .\\;; l-j; ,,A, e W f i (-- r* j~ ~ "

r p '-

ar.;a =tr /m x / p '.: G, x-' w.- 1 r-t-- . v. y. 'p,;fi[3,7.- a -- Ma=<r, i3~a.- y Lu?_'. _<y , g~;-~s j. t.,., e ~.. =. 4 ,.t_ a _~_ 'i .n J 0 ~ gw., 4 g @-r-p I L 3.. g 4 1 mamwan cs w 2+ i =. =. = - - .a. " L e. w /m N.mm, w L,,,,, ,y.,y e2-ass = ne ,J. Lw : q+-- m.. ~._. __, 1 ~ = - 4 4 l c..., 3-- q .t.:::::::...e. ><Z.J, __ O, i i.,. li "l g 7-9 g;,.. - - - - - g \\ l 2 V_ - 1 PL AN e El 2V-6* ~7TE - F 7 - i / c. TB NG~2~w lN& &a ~ I I I t

OVERSIZE DOCUMENT PAGE PULLED ~ SEE APERTURE CARDS NUMBER OF OVERSIZE PAGES FILMED ON APERTURE CARDS %ot3ooSS di 15(o APERTURE CARD /HARD COPY AVAILABLE FROM jg /hQ MECORDS AND REPORTS MANAGEMENT BRANCH i -.. - -}}