L-97-039, Provides Notification of Completion of Corrective Actions for Thermo-Lag 330-1 Fire Barrier Systems at BVPS,Unit2,as Requested in GL 92-08.Response to 970829 RAI Re Ampacity Derating Issues & Calculation DE&S-SR/EM01,also Encl

Provides Notification of Completion of Corrective Actions for Thermo-Lag 330-1 Fire Barrier Systems at BVPS,Unit2,as Requested in GL 92-08.Response to 970829 RAI Re Ampacity Derating Issues & Calculation DE&S-SR/EM01,also Encl

ML20217J828
Person / Time
Site:	Beaver Valley
Issue date:	10/13/1997
From:	Jain S DUQUESNE LIGHT CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
Shared Package
ML20217J832	List: L-97-039, Provides Notification of Completion of Corrective Actions for Thermo-Lag 330-1 Fire Barrier Systems at BVPS,Unit2,as Requested in GL 92-08.Response to 970829 RAI Re Ampacity Derating Issues & Calculation DE&S-SR/EM01,also Encl ML20217J836
References
GL-92-08, GL-92-8, L-97-039, L-97-39, TAC-M85517, NUDOCS 9710220149
Download: ML20217J828 (11)

Text

- - - - - . --------- ------- -- _

1 g n..vm vp.y Poww Swen stw,w.wi.m uorram

• ESNIN..w.m o

ruNU$$.

UZn7. /%.n October 13,1997 L 97 039 U. S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, DC 20555-0001

Subject:

Beaver Valley Power Station, Unit No. 2 Docket No. 50-412, License No. NPF-73 Thermo-Lag 330-1 Fire Harrier Upgrades and l Resporse to Request for Additional Information Regarding Thermo Lag Related Ampacity Derating Issues at Heaver Valley Power Station, Unit No. 2 (TAC No, M85517)

As requested in Generic Letter 92-08, "Thenno-Lag 330-1 Fire Barriers,"

notification is being provided relative to the completion of corrective actions for the Thenno-Lag 330-1 fire banier systems at BVPS Unit 2. liardware modifications and engineering analyses to assess the Thermo-Lag 330-1 fire barrier systems at BVPS  !

Unit 2 have been completed with the exception of receiving the results of the chemical composition analysis of the final Thermo-Lag samples. The upgraded Thenno-Lag fire barrier systems utilized at BVPS Unit 2 for the protection of post fire safe shutdown circuits on electrical conduit raceways, boxed enclosures and their associated supports provide the required fire resistance capability based on qualified, tested configurations.

Minor anomalies or differences with tested configurations were evaluated and justification provided in supporting engineering analysis. Attachment I provides a summary status of the Thenno-Lag fire barrier systems at BVPS Unit 2.

This submittal also forwards the Duquesne Light Company response to the Nuclear Regulatory Commission's request for additional infonnation dated August 29, 1997, regarding Thenno-Lag related ampacity derating issues at Beaver Valley Power Station, Unit No. 2. The attached enclosure (Auschment 2) provides the response to the NRC questions and a copy of the Duquesne Light Company calculation of the effect on cable ampacities in fire-wrapped conduits. In addition, a copy of Okonite Bulletin EHB 90 is provided as requested in the RAl. This completes the ampacity evaluation perfonned for Beaver Valley Power Station, Unit No. 2 associated with the Thermo Lag installati ns as requested in Generic Letter 92 08. i 9710220149 971013-PDR ADOCK 05000412 P PDR hfh

Beaver Valley Pcwer Stati:n, Unit No. 2 BVPS Unit 2 Thermo-Lag 330-1 Fire Barrier Upgrades and Response to NRC RAl: Thermo Lag Ampacity Derating issues Page 2 If you have any questions regarding this submittal, please contact Mr. J. Arias at (412) 393-5203.

Sincerely, Sushil C. Jain c: Mr. D. M. Kem, Sr. Resident inspector Mr.11. J. Miller, NRC Region I Administrator Mr. D. S. Brinkman, Sr. Project Manager

.l

A'ITACilMENT 1 Status of Thermo-Lag 330-1 Fire Barrier Systems at HVPS Unit 2 Thenno Lag 330-1 fire banier systems are utilized at BVPS Unit 2 for fire protection of safe shutdown circuitt on electrical conduit raceways, boxed enclosures (electricaljunction boxes and pull boxes) and their associated supports. 3M Interam fire protection system was also utilized to protect portions of the conduit support system, flexible conduits and is used for protection of the cable trays. Thenno Lag is not used to protect cable trays at BVPS Unit 2 Based on the results ofindustry testing, the boxed enclosures utilizing Thenno Lag 3301 fire barrier systems cwdited for post-fire safe shutdown capability were either upgraded orjustified to meet the required 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> fire resistance capability. Engineering analyses were also perfonned to assess both the upgraded and non upgraded boxed enclosures. The upgrades were perfomied in two phases using industry testing to provide the basis for a 1-hour fire rated banier. These upgrades, utilizing additional Thenno-Lag trowel grade material and stress skin along the butted seams, were completed during the third quarter of 1996.

Duquesne Light Company has been actively pursuing a resolution to the Thenno-Lag fire barrier issues for BVPS Unit 2. As noted in our previous submittal dated February 14, 1997, a conservative decision was made at that time to upgrade an additional 900 feet oflarge size conduits (3" to 6" diameter) encapsulated with Thenno-Lag 330-1 fire banier systems which are credited for post-fire safe shutdown capability.

The larger size conduit raceways (3" to 6" in diameter) encapsulated with Thenno-Lag 330-1 fire banier systems were, at that time, in the process of being evaluated by comparison to industiy testing which had successfully qualified as a 1-hour fire rated assembly. One of the parameters previously credited in the engineering analyses based on a sampling ofinstalled Thenno-Lag was the installation method of pre buttering the joints with trowel grade material vs. post-buttering after the assembly is configured. In December of 1996, during the upgrade project for small size conduits, a 2" diameter conduit was discovered to be void of the trowel grade material along the seam of the pre-fonned half shell installation. Acceptable industiy testing on 3" diameter and larger sized conduits has been based on utilizing pre-buttered joints. Industiy testing of post-buttered joints on smaller sized conduits (two tests of 2" diameter conduits) had mixed results. The discovery that post-buttering was used on certain installations at DVPS Unit 2 had a direct impact on the acceptability of the engineering analysis efforts that were being developed for justification of the as-installed 3" to 6" diameter conduit configurations encapsulated in Thermo-Lag.

Approximately 400 feet of the smaller sized conduit (1%" and 2" diameter) encapsulated in Thermo-Lag 330-1 fire banier systems ciedited for post-fire safe shutdown were upgraded to meet the required 1-hour fire resistance capability. This was

~ _

- - . .- = . - _ _ _ _ _ . _ - _ , .. - - -. - - -

Attaciunent 1 Progress Sununary of Thenno Lag 330-1 Fire Barrier issues at BVPS Unit 2 accomplished by adding an additional 3/8" thick Thenno Lag pre-shaped half shells with I additional trowel grade and stress skin material on top of the original baseline existing Thenno Lag installation. The larger sized conduit (3" to 6" diameter) encapsulated in Thenno Lag 330-1 fire barrier systems were upgraded with a 3/8" (minimum) layer of Thenno-Lag trowel grade material, overlaid with stress skin material, which was added to the original baseline installation.

The Duquesne Light Company objective has been to ensure a 1-hour fire rated barrier utilizing Thenno-Lag 3301 fire barrier systems by bounding our installations at BVPS Unit 2 to a qualified, tested configuration. Minor anomalies or differences with tested configurations were evaluated and justification provided in supporting engineering analysis. Engineering analysis providingjustification of the Thenno-Lag installations at BVPS Unit 2 including ampacity derating evaluations have been completed. As noted in Generic Letter 92-08, all analyses associated with the Thenno Lag 330-1 fire barrier systems ul be retained for future NRC audits or inspections.

The results from final chemical composition testing, have not been received from the test agency. Compensatory measures (fire watches) will remain in piece until satisfactory results are received.

Page 2 of 2 e

NITACllMENT 2 IIVI'S UNIT 2 ItESI'ONSE TO TIIE NitC ltEQUEST I<Olt ADDITIONAL,INFoltMATION Dated 8/29/97 (TAC NO. M85517) l

~

U%

l.0 llACKGROUND This is the response to the NRC Request for Additional infonnation [Ref.1] regarding the heat transfer model used by Duquesne Light Company to calculate the ampaity derating factor at lleaver Valley Power Station Unit 2 (llVPS2)[Ref. 2). The NRC stairs request for additional information and Duquesne Light Company responses are provided in the paragraphs below

} following a briefintroduction of the revised methodology adopted by the Duquesne Light Company.

The subject calculation [Ref. 2) has been superseded in its entirety by a new calculation [Ref. 3) which follows the same basic heat transfer concept as the superseded calculation but employs a different set of heat transfer equations to calculate the pmtected cable ampacity and the corresponding ampacity derating factor ihr the fire protected raceways at 11VPS2. The essential features of the heat transfer model used in the new calculation are summarized below:

The heat transfer model is developed from basic heat transfer principles and correlations fbr conduction, convection, and radiation that can be found in popular heat transfer text books such as incropera [Ref. 8), Otisik [Ref. 9), and Iloiman [Ref.10]. Where uncertainties exist, effbrt has been made to bias the model toward calculating a higher thermal resistance so that the calculated ampacity is lower, and therefbre, a conservative ampacity derating factor is calculated. The heat transfer model has been verified by comparing its predictions against industry test data. In addition, sensitivity studies have been performed tojustify the assumptions and approximations used in applying the model to the raceways at IIVPS2. The heat transfer model has the following essential features:

1. The model treats a protected conduit in distinct regions starting with the cable conductors where the heat is generated and ending with the ambient where the heat is dissipated. The model has the ibilowing elements.

. Ileat generation rate in the conduit Ileat transfer from the cables to the conduit wall

. Ileat transfer in the air gap region between the conduit wall and the fire barrier

. Ileat transfer through the fire barrier wall Ileat transfer from the fire barrier to the surrounding

2. The conduit wall is not treated separately in the model for two reasons. First, the thermal resistance of the conduit wall is negligible due to the high thennal conductivity of the metal wall. Second, the thermal resistance of the region from the cable conductor to the conduit wall is back calculated using the IEEE/NEC [Refs. 4,5] baseline ampacity data which includes the thermal resistance of the conduit wall itself.

3. The wetted perimetsr approach followed in the original calculation [Ref. 2] and the equation Ibr the overall heat transte coefficient (U,. defined by Equation 4 in Ref. 2] are not used.

Response to the NRC RAIITAC NorUi7) Page 2 of 7

2.0 RESPONS13 TO Tile QUESTIONS 2.1 Ileat Trainfer Ratefrorn Ihe Cable Conductor to the Conduit \\'all and the Recornmended Options.

l With regard to the wetted perimeter approach and the equation Ibr the overall heat transfer coefficient (U, on page 3 of the calculation [Ref. 2] ) it should be noted that neither the wetted perimeter approach nor the equation for U, are used in the revised calculation.

The revised method calculates the cable ampacity without Thermo lag material from the cable conductor to the conduit wall using NiiC 1993. The ampacity is considered base line ampacity. The cable thermal resistance is calculated from cable conductor to conduit wall using baseline ampacity data. This thermal resistance accounts Ihr the resistance of the cable bundle in the conduit, the resistance of the air space between the cable bundle and the conduit wall, and the resistance of the conduit wall itself. The applicable equations are given below:

lleat generation rate in the conduit:

l q, = n,(n,,l'R) / C (1) where q, heat generation rate per unit length of the raceway, Blu/hr fl C unit conversion constant,0.2931 W hr/ Btu

/ conductor baseline current, Amps R conductor resistance per unit length, ohm /11 n,, number of conductors per cable n, number of cables in the conduit c subscript for cablu n subscript for conductor r subscript for raceway (i.e., conduit) lleat Transfer from the Cable Bundle to the Conduit Wall:

q, = U,(nd,)(1; - T,) (1) where U, overall heat transfer coellicient (conductor to conduit wall), Btu /hr-ft2 ,7 T,, conductor temperature, T T, conduit wall temperature, T d, conduit outside diameter,11 Response to the NRC RAI(TAC No 833I7) Page 3 of 7

lleat Dissipation from the Conduit Wall to the Ambient:

q, = t nd, )(h, .,,, + h,,,,s. ,,, )( T, - T,, ) (3) where h,., convective heat transfer coefficient from the conduit to the ambient, Btu /hr f1' *F h,,,,,, radiative heat transfer coefficient from the conduit to the ambient, Btu /hr-ft' *F T, ambient temperature, *F Equations 1,2, and 3 are solved for the overall heat transfer coefficient U, and the raceway surface temperature T,. Since the heat transfer coefficients h,.,, and h,,,,,, are temperature dependent, the solution is achieved by iteration.

A direct approach to estimate the cable geometry parameters and calculate a heat transfer coeflicient U, based on first principles is not followed due to its complexity and the uncertainties in the arrangement of the cables in the conduit, the contact area. and the contact heat transfer coefficient. A method that accounts for these details is described by Neher-hicGrath [Ref. I l], and is adopted by IEEE [Ref. 4] to develop ampacity tables for conduits with various cable sizes and cable fills. The conduit heat transfer model used for IWpS2 takes advantage of these published ampacity tables and back calculates an overall heat transfer coefficient U, from the cable conductor to the surface the conduit wall, The method of back calculation, while indirectly complying with the methodology described by Neher-hicGrath (Ref. I 1], also ensures consistency with the IEEE ampacity tables, i.e., in the absence of any fire barrier, the ampacity calculated by the method matches the IEEE ampacity value.

2.2.a Valklation Case and the Use of TUE Test Data The revised heat transfer model was validated against test data from TVA [Ref.13] and TUE test programs [Ref.12]. Two validation cases were used from the TUB test program: a 2 inch conduit with 3/4": nominal barrier thickness and a 5" conduit with 1/2' nominal barrier thickness. The barrier thicknesses used in the TVA validation problem are based on the actual measured values reported in the test report [Ref.13] Attachment A of the Themo-lag ampacity derating analysis [Ref. 3] compares the analytical .nodel versus test cases.

2.2.b Cable Fill Datafor Conduit 2CL6070D The cable fill for 2CL6070D consists of three 3/C 4AWO cables and one Triplex 250hiChi cable. The calculation for 2CL6070D in the revised calculation is based on these cables.

Respome to the NRC R.4I(TAC No 833I7) Page 4 of 7

2.3 Justification)or the Ampacity Loadsfor the Cables in the Conduits 2CL21JND and 2CL6070E The conduits 2CL213ND and 2CL6070E were analyzed using the heat transfer method described in Section 1.0 above. The results, based on the assumption that all of the cables in these conduits are loaded at their protected ampacity, indicated that some of the cables are overloaded beyond their rated ampacity. A supplemental calculation was performed for these raceways by considering the diversity of the tables installed in them.

Itaceways 2CL213ND Examination of the full load currents for the cables in conduit 2CL213ND has indicated that some cables are lightly loaded while the others are heavily loaded. When this diversity of loading is considered, and a heat balance for the entire set of cables in the raceway is performed, it was seen that the thermal conditions at the " required" current are mere favorable than the themial conditions at the " rated" current (i.e., "motected" ampacity).

This is demonstrated in Table 1 below where the total heat genued at the maximum required current is compared with the total heat at the protected ampacity. The total heat generated at the maximum required current is less than the total heat generated at the protected ampacity. Since the cable conductor temperature at the rated current conditions is 194 T, this implies that the ceble conductor temperature at the maximum current will be less than the cable conductor temperature at the protected ampacity (i.e.,194 T). It has, therefore, been concluded that the cables in raceway 2CL213ND have su0icient diversity so l

that overloading of the two cables (2SCANNLO25 and 2SCANNLO26) is compensated by the underloading of the remaining three cables. DLC has initiated condition report 971661 to evaluate the life expectancy for the cables listed in this paragraph.

Raceways 2CL6070E The situation with 2CL6070E is similar to the situation with 2CL213ND discussed above.

Ilowever, there are only three cables in this raceway with two of them heavily loaded. One of the heavily loaded cable (211VSAOL210) is used to power a heater and is normally not energized. This cable is energized automatically on high moisture content (70%) at the intake of Supplementary Leak Collection and Release System (SLCRS) fan and initiation of CIA signal. The low moisture signal is annunciated in the control room. A supplemental calculation was donc for raceway 2CL6070E with cable 211VSAOL210 deenergized. Under these conditions, the remaining heavily loaded cable (211VSAOL200) has approximately 2 percent ampacity margin, in addition to the 2 pere:nt ampacity margin, there is sunicient diversity to prevent overheating of this cable. This is demonstrated in Table I below where it is shown that the heat generated at the maximum required current is less than the heat generated at the protected ampacity. Okonite bulletin EllB 90 states that " operation at emergency overload temperature of 130C (266F) shall not exceed 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> in any twelve consecutive months nor more than 500 hours0.00579 days <br />0.139 hours <br />8.267196e-4 weeks <br />1.9025e-4 months <br /> during the life time of the cable". The heater is energized during Operation Surveillance Testing (OSTs) for approximately half hour per OST, twelve times a year. DLC system engineering records shows that the heater was automatically operated once when the moisture detector went out of calibration in 1995 and second time during calibration process the heater was energized by I & C Tech in 1997.

DLC has initiated a condition report CR 971662 to evaluate the cable life expectancy.

Respome to the NRC RAI(TAC No 855l') Page 5 of 7

Talite 1 Evaluation of Raceways 2CL213ND and 2CL6070E Ilused on llent Generation Rate (details are given in Reference 3)

Racew ay Ampacity Catile No. llaseline Protected Required lleat lleat Generated No. Detating Arnpacity Ampacity Ampacity Generated at Required r l' actor (Arnps) (Amps) (Amps) at Protected Ampacity Ampacity (litu'br ft)

(lituhr ft) 2CL213ND 15.79 2LAl'NNL010 65.0 54.8 29.8 4.5 1.3 2SCANNLOl6 65.0 54 8 8.3 4.5 0.1 2SCANNL017 65.0 54 8 9.8 4.5 0.1 2SCANNLO25 37.5 31.6 44.6 3.8 7.5 2SCANNLO26 37.5 31.6 $ 1.5 38 10.0

'I otal lleat Generated 20.9 19.0 2CL6070L 26 0'e 2LilSAOL245 253 6 187.7 90.1 21.4 4.9 2ilVSAOL200 307.2 227.3 223.3 23.0 22.1 2ilVSAOl.210 0.0 0.0 0.0 00 0.0 Iotal lleat Generated 44.3 27.1 Respome to the NRC RAI(TAC Nos5517) Page 6 of 7

____~

3,0 11EFEllENCES

1. Enclosure I to the letter from D. S. lirinkman (Omce of Nuclear Reactor Regulation) to J.

11. Cross (Duquesne 1.ight Company) dated August 29,1997.

Subject:

Requestfor ildditionalInformation (R.II) Regarding 1hermo Lag Related Ampacity Derating issues -

Beaver l' alley Power Station. Unit No. 2 (BI'PS *J (TAC NOS. AINS$17).

J. Attachment I to the 1.etter from Sushil C. Jain (Duquesne Light Company) to U. S. Nuclear Regulatory Commission dated Alarch 25,1997.

Subject:

"# caver l' alley Powcr Station.

Unit 2 Docket No. 50-.ll2, License No. NPF-73. Response to Requestfor information Regarding Ampacity Derating ofCables Enclosedin 7hermo-Lag 330-1 Fire Harricts. "

3. Beaser Valley Power Station Unit 2 Calculation Titled "Ampacity Derating Factors offire.

Protected Cables," Calculation No. DE&S SR/Eh101, Rev.1, October,1997. DE C analysis 10080 l!-138 rev. 2 addenda A2.

4. IEEE/lPCEA Standard S-135/P-46-426, Powcr Cab /c Ampacities: l'olume I-Copper Conductors and l'olume II Aluminum Conductors, I984.

5. Early, hi. W. et al., National Electric Code //andbook, National Fire Protection Association, 1993.

6. ICEA Standard Publication, Ampacities of Cables in Open top Cable Trays,ICl!A P 54-440 (Third Edition), NEh1A WC 51-1986.

7. IEEE P848, Draft IEEE Standard titled," Procedure for The Determination of The Ampacity Derating of Fire Protected Cables," Draft 16 January 1,1995.

8. Incropera, F. P. and Dewitt, D. P., Introduction to // cal Transkr, John Wiley & Sons,1990.

9. Ozisik, hl. N., Baalc // cat Tran3fer, hicGraw llill,1977.

10. Iloiman, J. P., Heat Transfer, hicGraw 11i11,1963.

I1. Neher, J.11. and hi 11. hicGrath, The Calculation of The Temperature Rise and Load Capability of cable Systems, IEEE Transactions, Vol. 76, October 1957, pp: 752 773.

12. Omega Point Laboratories Report Titled,"Ampacity Derating of Cables Enclosed in l Conduits with Thermo Lag 330-1/7701 Upgrade Electrical Raceway Fire Barrier Systems (ERFDS)," Project Nos. 11960 97337 & 97338. Tennessee Valley Authority. August 21, 1995.

Omega Point Laboratories Report Titled,"Ampacity Derating of Fire Protected Cables,"

Project No. 12340-94583,95165-95168,95246, TU Electric, hlarch 19,1993.

Response to the NRC RAI(TAC No 853 t 7) Page 7 of 7

~ n,