ML18003B124

Electrical Separation Analysis.

ML18003B124
Person / Time
Site:	Harris
Issue date:	12/31/1984
From:	EBASCO SERVICES, INC.
To:
Shared Package
ML18003B123	List: IR 05000400/1985008 ML18003B122 ML18003B124
References
NUDOCS 8506250404
Download: ML18003B124 (19)

Text

ChROLINA POMER 6 LIGHT SHEARON HARRIS NUCLEAR POWER PLhNT ELECTRICAL SEPhRATIQN hNhLYSIS EBASCO SERVICES INC.

2 WORLD TRADE CENTER NEM YORR, NY 10048 DECEMBER, 1984 850b250404 850 0400 pa@ >>OC< Oa

ThBLE OF CONTENTS

~Pa e 1.0 Introduction 2.0 Scope 3.0 Conclusion 4.0 Analysis S.O Results 6.0 References Tables Table 1 SHNPP Cable and Conduit Data Table 2 SHNPP Cable and Tray Data Table 3 - Steady State Analysis Results

~Pi res Figure 1 - Cable Tray and Conduit Configuration Figure 2 - Transient Temp vs Time Curve

~ ~

'A

1.0 INTRODUCTION

In accordance with the Shearon Harris Nuclear Power Plant FSAR, the criteria for separation of electrical equipment and circuits is based on IEEE 384-1974 as endorsed by Regulatory Guide 1.75 revision 1 dated January, 1975. %here the minimum separation distances referenced in these documents cannot be maintained, separation of one inch between No~lass 1E raceway and Class 1E raceway is allowed in areas where the hazard is limited to failures or faults internal to the Non-Class lE raceway Paragraph 5.1.1.2 of IREE 384 states:

"In those areas where the damage potential is limited to failures or faults internal to the electrical equipment or circuits, the minimum separation distance can be established by analysis of the proposed cable installation" ~

Safety related cables for the SHNPP meet the requirements of the IEEE

'3B3 flame test.

It is the purpose of this report to provide such an analysis for allowing a one inch separation between an enclosed NonWlass 1E raceway (i.e., conduit) and an'open top Class 1E raceway (i.e., cable tray).

2.0 SCOPE The scope of this report encompasses all Non-Class lE cables routed in conduit which may be located as close as one inch to any Class 1E cables in open cable tray.

3.0 CONCLUSION

The event postulated in this report would have no adverse effect on the operation of Class 1E cables located as close as one inch to a Non-Class lE power cable conduit. This conclusion is based on the transient heat transfer analysis included in this report (using the temperature vs. time curve of Figure 2, and the Shearon Harris Nuclear Power Plant cable insulation emergency rating, i.e., 130'C for 500 hours0.00579 days <br />0.139 hours <br />8.267196e-4 weeks <br />1.9025e-4 months <br /> over the life of the cable).

4.0 ANALYSIS 4.1 Abstract The analysis examines the thermal effect of a three phase fault for a cable installed in a conduit located one (1) inch from a Class lE open cable tray. A single analysis was performed for the maximum heat source conduit (i.e., a 5 i'nch conduit containing 3-750 kcmil conductors, and a fault current of 33.2 kA), and a tray with the lowest thermal capacitance (a six inch control tray, 65.3X model envelopes all Shearon Harris configurations.

fill), since this 4.2 Technical Data The technical data used in the calculation are as follows.

a) NNS Conduit Power cable voltage class, number of conductors in conduit, conductor size, maximum continuous current, maximum fault current, maximum cable outside diameter, conduit size, cable insulation and )acket material. (See Table 1.)

- Tray (poser and control),

fill, and type b) Class~lE 0 en Cabl~eTra height, width, maximum tray configurations, as well as cable insulation and jacket materials. (See Table 2.)

c) Pault Interru tion Time - Interruption togae for the maximum fault current is considered to be the time at which the conductor melts. Credit was not taken for the Non&lass lE circuit breaker clearing the fault. Thia ia considered extremely conservative, since the SHNPP breakers are considered reliable. hlso, for the event postulated, an additional upstream breaker would normally provide backup; therefore both breakers would have to fail before the interruption time assumed in the analysis would ever be I reached.

4.3 Steed~State Heat Flow Calculations In the analysis only resistive (I R) losses were considered since they are the major source of heat. Under normal operating conditions, the losses due to the current carried by a conductor raises its temperature until an equi14brium is established. Lt equilibrium, the heat generated is equal to the heat dissipated.

h worst case hes been chosen from the various cables and cable trays 1isted in Table 1 and Table 2, based on the largest heat capacity.

This case consists of a 5 inch conduit containing three 750 kcmil, 15 kV power cables. Because these cables have the highest heat capacitance (See Table 3), they have the greatest potential to heat the cables-in the tray adjacent to the conduit. When the fault occurs, heat trensfer is primarily due to radiation and convection. The heat capacitance, heat generation under normal and fault current conditions, and the time required to melt the cable conductor are shown in 'Table 3.

The results show that the 750 kcmil, 15 kV power cable hes the highest thermal capacitance and it takes about 25 seconds for the fault current

+hk to melt the conductor. Therefore, the cables in"trey near this conduit would have the greatest potential to be heated to,q high temperature.

The smallest tray is considered to be the worst case because the surface temperature varies inversely with the sire of the heat sink for a given heat transfer rate. Both in terms of a maximiaed heat source (failed conduit) end e minimized heat sink (cable tray), the selected

~ model represents the bounding case.

The steady state temperature distribution in the failed conduit end the cable in the tray was calculated assuming that the failed conductor is maintained et the conductor melting point temperature end the heat transfer is mainly due to radiation. The results show that the cables in'the tray would reach a temperature of 800'o 900'F if, the cable tray absorbs all the radiated heat for one hour. These calculations assume that the cable conductor is maintained at 2000'F once the conductor melts due to short circuit current. In reality, the fault current which would be essentially constant along the length of the run would also cause melting at the svitchgeer cable termination. Thus the actual temperature would be somevhet less than assumed.

0

~ f

The total heat generated due to the fault current is 84,292 STU per hour per foot of the cable. For. a fault current of 33.2 kA, the conductor takes approximately 25 seconds to reach its melting point and, therefore, the 750 kcmil conductor stores 599 $ 19 per foot of the cable. Vith a constant conductor temperature of 1980'F, this amount of heat will be dissipated through radiation and convection in about 2255 seconds. However, the conductor temperature <<ill decrease rapidly due to the heat dissipation and the cessation of current after the conductor <<elts. Therefore, the reduced conductor temperature <<i11 result in a decline in the convective and radiative heat transfer rates. In view of the time scales of the problem and the .assumptions involved the steady state results, though conservative, are not realistic. For this reason the problem was realistically modeled for transient heat transfer calculations using the Systems Improved Numerical Differencing hnalyxer (SINDA) computer code.

4.4 Transient Heat Flow Calculations h model of the conduit and cable tray was developed to study the temperature transient <<hen a cable fails in the conduit (Figure 1).

The model consisted of a 5" conduit containing three 15 kV, 750 kcmil power cables run parallel to a cable tray six inches wide by aiz inches deep filled with cables so that a layer of. cables lie across the top of the tray one inch below the conduit. For simplicity, cables of the same sixe were assumed to occupy 651 of the tray cross-section, with the remainder occupied by air. The tray ambient temperature was conservatively assumed as 120'F. The model also assumes that a fault current of 33.2 kA is passed through the cable until temperature of 1980'F at which point it ceases.

it reaches a The conduit is heated by conduction and the heated conduit radiates to the surface of the cables at the top of the tray.

The configuration was analyxed using the SINDA heat transfer code.

Figure 2 illustrates the transient temperatures. The faulted conductors reach their melting point in approximately 25 seconds.

Insulation and )acket integrity will not be preserved due to the intense heat and gasses generated by the insulation and acket material. This will allow the melted copper to come in contact with the steel conduit causing the conduit temperature to increase rapidly.

The surface temperature of the upper cables in the cable tray will rise to 226'F, and then begin to fall. The cable fault in the conduit raises the temperature of all the cables in the tray to some degree but in no case is the temperature of any cable higher than 226'F.

The transient thermal analysis shows that in a bounding case of maximum heat source, minimum heat sink, the cable tray temperature will not exceed 226'F. (There is a + 10'ncertainty.) larger trays will not be heated to as high a temperature, as there is a larger area for convection and a greater likelihood that there will be a larger spacing between the bottom of the conduit and the top of the uppermost cable than the one inch assumed in this study.

5.0 RESULTS The results of the transient heat transfer analysis show that the surface of the upper control cables in the tray below the failed conduit rises from 155'F (which is the normal steady state )acket temperature) to 226'P + lOX, and then begins to fall in less than one hour after the occurrence of the postulated fault. However, the average temperature of the cables in the tray will be raised only a few degrees above their steady state temperature of 155'F because of the low capacity of the heat source. Consequently no cable will reach a temperature close to the 249'F (226 + 10X) reached by the upper jacket surface, In any combination of conduit and cable tray configuration.

These results are still conservative because the configuration analyzed includes the failure of a highest heat capacity cable (heat source) above the smallest heat capacity of cable tray (heat sink)

Zf the tray under the faulty conduit is a power cable tray (which will have a larger heat sink effect than that of the control tray), the heating effect on the power cables will still be similar to that on the control cables. However,'the temperatures reached by the pmier cables will be higher than that of the control cables because the power cable operating temperature is, by design, 90'G (194'F). This transient state will also be within the acceptable operating conditions of the power cables, since it will result in power cable temperatures well below their emergency rating of 130'C (266'F) for 500 hours0.00579 days <br />0.139 hours <br />8.267196e-4 weeks <br />1.9025e-4 months <br /> over the life of the cables.

6. 0 REFERENCES

1. SHNPP File No. 3D2, Calculation No. 15 dated 12/84

Paya 5 TABLE CAROLIlQ, K~ Cs LXCFN.'OMPASS'HEARS HARRIS HUCLEAR PNER PLAÃF ChbLE DATA Cabla Porc>> Nueber of Xnaulatton/

Cable Conductora Con~ Outatde Con Jacket Conduit ~a Continuoua Fault Circuit Study Voltage in a ductor Mam@ter ductor &ti Current Current breaker Int Ho. Claaa, Conduit Sire (Inchea) Nsterial 'lnehea) (Aepa) (kh) (gN) Time I, a, 15%V 3-C/c 750 keel 1.62 Copper EPR/CPE 5 493- 33.2 Xnftnt t ~ (ace Hog-

b. 15kV 3-1/e 350 %cail 1.22 Copper EPR/CPE 4 330 33.2 ZnBntte'aae Ho%

15 kY 3-1/e 4/o AVC 1.10 Copper EPR/CPE 3 247 33.2 InBnjte (aee+ota 2~ ~o 60% 3 I/e 500 kemt1 1 ~ 19 Copper mX/tk 4 456 42.9 Infinite (aae Note

b. 600V 3-1/c 350 kcatl 1.05 Copper Rra/M 3 367 42.9 Infinite (aee roti 3-1/c 4/0 AVC O.S5 Copper ava/m 3 265 42.9 Xnftntte (ace Note 600'0%

3-1/c 2 AC 0.59 Copper ~n 1.5 125 It.4 Xnfinita (aae Note 600V 3-1/c 6 AMC 0.47 Copper mx/n 1.5 71 IT.4 Xnfinita (ace mote 600'(NP 3-1/e 10 AMC 0.37 Copper amfra 1 39 IT.4 Infinite (ace Rote 3 I/c(T) 3$ 0 beati 2.26 Copper ma/ra 4 367 42.9 1nfinita (cia Nota h, 600@ 1 3/e(T) 4/0 AC 1.84 Copper am/FR 3 265 42.0 Xnfteite (lent Wo 60% 1 3/c 2 AMC 1.1$ Copper ash% 1.5 117 17.4 Infinite '(see lfees 3 600V 1 3/c 6 AVG 0.&7 Copper max/n 1.5 86 17.4 Infinite (ese k.

~

600V '1 3/c 10 AMC 0,65 Copper BTK/tR 1 40 ITo4 Infinite Ne

(

HeTE 1: Circuit ta considered tntertopted at the ttea that the coppe>> te ealted.

T: Triplexod.

Page 6 CAROLINA tOMER 6 LIGHT. COMPANY SHEARS SLRRIS NUCLEAR PCMZR tIAÃX .

CAJOLE AND TRAY DATA A. CA%IX CMSXRUCTION Insulation Materials BTI- High Temperature Xertte Proprietary) thta. insuiation fa ~ dertvattva of IPR RPR- Ethylene PropyIene Rubber Jacket Mstertale Vulcantsed Chlorinated Rubber CPE- Chlorinated Polyethylene

- A11 conductor operating temperatures are o

%0 C

h. TRAY ORIEHTATIONS 6NN Control Tra H > 4 I/2 or 6 inches M ~ 6, 12, 18, 24 or 30 inches

'ax Fill ~ 65.3'X of 6 tnches 600V Pover Tra H > 4 l(2 or 6 inches 12, 18, or 24 inches

'ax Fill ~ 34.8X of 4 inches SkV Pover Tra H ~ 4 1f2 or 6 inches Q ~ 6, 12 or 24 inches Max Fill ~ 33.7L of 4 inches Trays are ateel, aoltd or ladder type

f'aac 7 TABLE 3 CAROLER PSfKR & LICHT CNPASV SHEARS HARRIS NUCLEAR PSfER PLANT STEAUt STATE ANALTSI8 RESULTS 2

Sctf$ 1 Poedt C$ bld Conduetot Condue tot 8$ $ t tault I> R Tfac. ol No. Vojtage Df$ llletet Re$ 1$ tdned Curtent,lr Conduct ot Inch Capacitance C1$ 0$ oh$ NI/Ft bTU/Ft F lA Ht tt Nelting, See 1 ~ 1.07 2.24 R 10 0,320 33.2 8.43xl0 25.6 15 kV 0.75 4.32 R 10 00157 33,2 2.0 lO 6.5 15 N 0,59 6.89 R 10 00097 2.2aao 2.5 2 P ~ 0.98 3.14 R 10 0.268 42.9 le02IIIO 9.2 0.76 4.32 R 10 0.161 42.9 2.21IIto 4.0 C ~ 0.59 6.89 R 10 0.097 44.9 0.22I ia 1.5

a. 0.35 2il3 R 10 0.034 17.4 2,20alO l.O Po 0.23 5.34 R 10 0.015 17.4 2. 02110 0,18 0.16 1.35 R 10 0.007 17.4 104%10 0.034

ESASCO SERVICES IHCORPORATED P .~u-s4 pic>E 6 0<

DE>T.

CHKD. SY o*va J~ ~~ ~~ OPS NO NO.

CI.ICNT hPOt iWA, W~~ w Ll

~ ao)ccT ~ I-lE LION 4 A.P' IS, HUCLEA. ~9 +LANG

'b, LF TFi@ Y AN(O rv 9 i l 0 F)C> uZ A7lOAl-PIC>UCc.

1

A

~ ~

~

~a~.

~

9

~ ~

CAROl.lHA ~1 5 I.ICHOR CNPANV SHKAPAN HARRIS NP VIIITNO. 1

'f50 KCMI. 15 KV PCNER CABAL fN 5 lHCH 0085lJfl TS~mmAmRE WANMNYcue TQ CABLE FAKSRK a S i eel ~ erne s o e ~ seem> ~ ~ ~ 1 IE ~ 1 t

e 1.

e fAILIDCSOLI llICUl'4 tA fAIL%O OJItACe Ot CNLf

~tf OIhlCTLVSBttATII

~ e ~ Seel ~ ~ e y. ~ eyr.' e e 4. ~ e 71 ~ 0 le tRc Rl etc se $ Rc FIGURE 2

All~~&X SHNPP FSAR

~

The minimum separation distance between enclosed racewa o between barriers and the protected raceway(s) is 1 in. "~4SKAT A (b) Cable and Raceway Hazard Areas Analyses of the effects of.

pipe whip, jet impingement, missiles, fire, and flooding demonstrate that safety related electrical circuits, raceways, artd equipment are not degraded beyond an acceptable level.

The analyses are referenced as follows:

High Pressure Piping (Section 3.6)

Missiles 'Section 3.5)

Flammable Material (Section 9.5.1)

Flooding (Section 2.4)

In fire hazard areas outside the cable spreading rooms, where redundant safety related trays or safety related and non-safety related trays are exposed to the same fire hazard, protection has been provided by spatial separation, fire suppression systems, Eire retardant coatings, fire barriers, or combination thereof.

(c) Cable Spreading Area and Control Room The cable spreading area is the space below the Control Room where the instrumentation and control cables converge prior to entering the control, termination, or instrumentation panels. Refer to Section 8.3.1.2.14 for further discussion of the circuits in these areas' In the cable spreading rooms, cabling for redundant safety divisions A and B are separated by three hour fire barriers.

Automatic sprinklers are also provided in the cable spreading rooms. Non-safety related cables are run in separate raceways from safety related cables with a minimum separation distance of 1 ft. for trays separated horizontally and 3 ft. for trays separated vertically. Where the minimum separation distance could not be maintained, fire barriers, and/or enclosed raceways, have been utilized with a minimum separation distance of 1 in. between enclosed raceways or between barriers and the protected racewa (s).

IMMRW 9" The above separation methods are based on the following:

(1) Cable splices in cable trays are prohibited.

8.3.1-38 Amendment No. 15

=. INSERT "A" Non-Class IE circuits are generally separated from Class IE circuits by the same minimum spatial requirements as redundant Class IE circuits are separated from each other. However, where the converging circuits are contained in Class IE tray and non-Class IE conduit, only the non-Class IE raceway circuits are enclosed (i.e., in conduit). Analysis of this conf iguration has demonstrated that there would be no adverse ef fects of the non-Class IE circuit upon the Class IE circuits. Non-Class IE circuits need not= be protected from Class IE circuits.

INSERT "B" Where the converging raceways are specifically Class IE tray and non-Class IE conduit, only the non-Class I E raceway circuits are enclosed (i.e., in conduit). Analysis of this configuration has demonstrated that there would be no adverse effects of the non-Class'lE circuits upon the Class IE circuits.

Non-Class IE circuits need not be protected from Class IE circuits.

(1578GAS/ccc )