ML17262A014

Research Information Letter 0014, Physical Separation Criteria for Electrical Cable Trays (Horizontal Open Space Configuration.

ML17262A014
Person / Time
Issue date:	11/09/1977
From:	Levine S Office of Nuclear Regulatory Research
To:	Case E, Minogue R Office of Nuclear Reactor Regulation
References
RIL-0014
Download: ML17262A014 (7)

Text

1.,..

_\.

NOV 9 1977 MOl>RANDUM FOR: Edson 6. Case, Acting Director Office of Nuclear.Reactor Regulation Robert B. Minogue, Director Office of Standards Development FROM: Saul Levine, Director Office of Nuclear Regulatory Research

SUBJECT:

RESEARCH INFORMATION LETTER #14 - *PHYSICAL SEPARATION CRITERIA* FOR ELECTRICAL CABLE TRAYS (HORIZONTAL OPEN

>>' SPACE CONFIGURATION)"

Introduction

-~::

This memorandum transmits the results of a c0111pleted portion of the NRC Fire Protection Research Program relating to the adequacy of cable tray separation to prevent the spreap of electrically initiated and exposure fires. It indicates that currently used criteria for cable tray separation appear to be adequate for electrically 1nftiated fires but that changes may be required for fire due to external ianition sources.

Background

The specific research upon which this RIL is based is outlined in the Research Support Branch Plan (Enclosure 1) for Fire Protection Research.

. "i\' The overall Fire Protection Program is based on the research need

-~:

identified in conjunction with the* review of the Browns Ferry fire and through consultation and fonnal review with the NRC user groups. The Electrical Standards and Fire Protection Research Review Group was the focal point for both the formulation of the research program and evaluation of pr~gram results that form the basis of this RIL.

The specific user requirement for the research conducted to date with electrically initiated fires is based on the need to verify Regulatory Gu1de 1. 75, "Physical Independence of Electric Systems." Exposure fire testing employing external fuel sources was conducted to provide data for the current NRC staff position as documented fn the Appendix A to the Branch Technical Position APCSB 9.5-1, "Guidelines for Fire Protection for Nuclear Power Plants" and in the draft Regulatory Guide 1.120, "Fire Protection Guidel foes for Nuclear Power Pl ants. 11 Regulatory Guide l. 75 requires only consideration of electrically initiated fires without consideration of any fuel source other than the cable insulation itself.

.,,...* I****---------------********

• • • • • • • • -- - - I - -- -----*-1-*****-**-*-*-*-*-*--*-*****

---**----***** 1-- --*------* -------

DATE. ................................... .... .. . . ............................. .................................... .. ....................................

NRC FORM 318 (9-76) NRCM 0240

• U, ** QOVERNMENT PRINTINCJ OFf'ICEr 11178 - 02CMIM

Edson 6. Case Robert B. Minogue NOV 9 1977 Howl'ter, the staff position for future plants is that exposure ffres 11Ult be consfdered. The type and size of the exposure fire fs to be based on a fire hazards analysis and will vary from plant to plant and will also be different for different locations within the plant. The tests conducted to date and the conclusions presented in this RIL support an assessment of the effectiveness of separation distance in preventing the propagation of both electrically initiated fires and exposure fires.

A complete systems test including all cable tray configurations and cable tray components such as conduits, barriers, breaks and penetration fire stops was not possible because of the lack of unifonnity in existing and proposed plants, the size of the test that would be required, and because there is no single design basis fire that can be utilized for design purposes. The total electrical cable system fire problem is covered by

.... the first five program elements outlined in Enclosure 1. This RIL covers results from tests and analysis for the horizontal open space cable tray

-~::

configurations portion of the first program element. Subsequent RIL's will cover data on vertical cable tray configurations and enclosed or tunnel areas.

Specifically, the scope of this RIL covers evaluation of the effectiveness of cable tray separation for typically loaded horizontal cable trays in a building approximately 60 ft. long, 19 ft. high and 24 ft. wide simulating an open plant configuration where the effects of reflected heat from the walls and ceiling are minimized. Eighteen-inch wide open-ladder steel cable trays, typical of those used in power plants, were used. Horizontal and vertical separation distances between single.trays representfng different safety divisions were varied from f1ve ft. vertical and three ft. horizontal (which are required 1n Regulatory Guide 1.75 for separation between safety divisions) to a minimum of 10.5 inches vertical and B inches horizontal.

Tests were also conducted with a single division comprising 14 closely packed trays. The fourteen cable trays simulating a closely packed single

~, . division were spaced 10.5 inches vertically and 8 inches horizontally (Figure 1). Allowable separation distances stated in Regulatory Guide 1.75 were used between redundant divisions.

Fires were started using both electrical intiation (simulating an electrical short between two conductors) and an exposure fire using propane burners. Air flow was maintained to simulate normal ambient plant air flow in plant open space areas. Some initial t~sting utilized cable trays that were uniformly loaded up to the top of the side rails, which resulted 1n a 60% volume fill. Later tests, including all of the tests with the packed 14 tray division, used a random f111 pattern with a m**** omo* ..1-- --1--- -*- *---- *---*-1-----**--*-----1-----------

DATI! * ............................ ... .......................... .... ............................... ... . . ... . . ....................................................................... ..

NRC PORK 318 (9-76) NllCM 0240 RU. a. GOVERNMltNT PRINTINO OFP'ICIE1 111711-11-M

• ~'

Edson G. case Robert B. Minogue NOV 9 1977 loading of approximately 40% volume fill recOfllDl!nded by the Insulated Power Cable Engineering Association (IPCEA). Rated electrical load on cables was utilized initially but was shown to have no significant effect on the propagation of fires in open ladder, uncovered trays.

The cable fill pattern was varied from a uniform fill pattern representing the maximum fuel density and minimum air space between cables to a random pattern with minimum fuel density and maximum air space. Fire retardant cable was utilized that is in compliance with IEEE-383, "IEEE Standard for Type Test of Cl ass l E Electric Cables, Field Splices, and Connections for Nuclear Power Generating Stations."

• Results The test data and analysis {Enclosure 2) in support of this RIL show that there are a number of key parameters that determine whether or not a .

-~:: cable tray fire w111 develop and propagate to an adjacent tray. Probably the most important factor in detenn1ning if a cable tray ffre will develop is the spacing between cables in the tray which establish the air-fuel mixture at the burning surface.* Since the cable pattern cannot be specified and maintained during power plant construction, a random pattern w1th maximtnn and uniform air space between cables was cho~en after it was shown to be the most vulnerable to fire propagation. Tnis pattern with a 40S volume fill was used for inost of the tests.

It was established by measurement and observation' of the test films that the fire propagation from one tray to a tray above ft depended on the coll ect1on of hot gaseous fuel re..1 eased from the fuel. Propagation 1n thestests conducted occurred by ignition of the gaseous fuel above the cable tray to which the fire 1s spreading, with a subsequent spread of the flame down into the tray itself. A necessary condition for fire propagation

., is that the fire become fully developed in terms of the cable bundle temperature and the equilibrium heat fluxes reached. The burning surface area is probably the most important factor in establishing a fully developed fire. This fully developed condition ITlJSt be sustained long enough to allow for the collection of the gaseous fuel. Also, suff1c1ent fuel ITl.!St be available to support the combustion long enough to allow for the collection of sufficient gaseous fuel. The air flow over the cable trays and the extent to which the gaseous fuel can be trapped by ceiling walls or other barriers also affect the degree to which the gaseous fuel collects around the cable tray. It was shown that a fully loaded cable tray is itself a sufficient barrier for the collection of the gaseous fuel.

"":::::: I :: :  :

..... J .... ... ... . .

1.

I m mI *  :  : .:* \ *: : : : : : : : : : : : *

.. . . . . . . . . -** * * * * * * * * * . ---- I ::~::~::~=~~m \. : : : : : : : : : : : : : : : : : :

NB.C PORK 318 (9-76) NRCM 0240

• U. IS. OOVERNMltNT PRINTING OFP'ICE1 Ul7lll - GZ-M

Edson G. case Robert 8. Minogue NOV 9 1977 Two propane burners rated at 70,000 Btu/hr., burning for five minutes beneath 1 cable tray, provided sufficient conditions to yield a fully developed cable fire. However, it .cannot be concluded from the tests that these are minimum conditions.* Also, the test data show that for the configuration tested, a fully developed cable fire will propagate ff the f1re covers at least an 18 x 36 inch area and there is sufficient fuel to sustain the fire for about five to ten minutes. As is the case for conditions detennined to obtain a fully developed fire, these conditions for propagation may not be a minimum set.

In sunmary, it is beneficial to consider the cable tray fire question 1n two parts. The first consideration 1s to establish the possibility of a fire developing in a cable tray. If a cable fire develops in a cable tray, the next consideration is the possibility of the fire propagating to an adjacent tray. In keeping with this problem definition, the

• t following specific conclusions can be drawn from the tests and analysis

-~::

conducted.

1. It 1s possible to initiate a fire electrically in a fully loaded open ladder cable tray by generating an overcurrent in two contiguous conductors. The rated cable current must be increased by about a factor of five to start the fire.

2. With an electrically initiated fire fn a cable tray, propagation to adjacent cable trays fs un11kely fn horizontally oriented cable trays filled with electrical cable capable of passing the current version of the IEEE-383 standard. The apparent reason that cable tray fire propagation 1s unlikely is because there is less chance

• of an electrically 1n1t1ated 'f~re becoming fully developed.

The tests showed that an electrically fnit1ated fire started at a single location and was followed by other sporadic point source fires. For the tests conducted with electrical 1n1tfat1on, the fires did not spread over a large enough area to become fully developed. For the cable tray configuration tested, ft isuunlikely that an electrically initiated fire could cause concurrent burning o over a large enough surface area to establish a fully developed fire which is a necessary condition for propagation. It 1s possible that further testing with electrical in1t1ation under different

. .1...

conditior~, such as sustained overload in covered trays may result in fully developed fires *

......* 1* *-------

• URN AME.

--*--1* --- -------1**------- *-*---.. ,. . --*-----*--1 * - ----- -1 * ---.. . . . :~.

• • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • ...................... ,,,,,_.,._............ . ...................................-......... * *************n***-*****-*0 ************* * . . . . . . . . . . . . . . . . . . . . . . . . . .- * * * * * * * * * * * * *************-*--**-**--

DATE.. .....*.................................... ..............*....*.*.........*......... ............................................ ............................*..*........ . .........................*....*......... *******************-*****************

NJlC FORM 318 (9-76) NRCM 0240

• U: II. OOYltRNMltNT PRINTING Ol'P"IC!ti 1976- ea-z.i

Edson G. Case Jjobert B. Minogue Nn" " ion 3*. Fully developed cable tray fires were obtained using two propane gas burners each having a nominal rating of 70,000 Btu/hr. w1th a burn time of five minutes. ~IEEE-383 specfffes a single 70,000 Btu/hr. with a 20 minute burn time). Such an exposure fire is not considered unreasonable and has been estimated to be equivalent to the fire that would result from a small oil spill (approxfmately one gallon) and subsequent ignition.

4. If a fully developed fire is obtained in a cable tray, propagation to adjacent trays located 10.5 inches or less above the ffre fs likely to occur in plant open space areas. Tests were not conducted to determine the minimum space required between trays to prevent a fully developed ffre from propagating to the tray above it *

5. A fully developed fire inftfated in a typically loaded safety division (consisting of two hor1zonta1 stacks of seven trays with a spacing of 10.5 inches between each tray and the tray above and a spacing of B inches between the two stacks), after tray-to-tray propagation within that d1v1s1on may propagate to a redundant safety division across the 5 ft. vertical separation distance specified fn Regulatory Guide 1.75.

Evaluation The data obtained confirm the or1gina1 licensing position as taken fn Regulatory Guide 1.75 requiring a vertical separation of 5 ft. between

• safety divisions 1f only electrically initiated ffres are considered.

Testing was limited to electrical cable capable of passing IEEE-383 installed 1n horizontally oriented.cable trays 1n a test cell simulating a plant open space area.

., Regulatory Guide 1.75 1s based on consideration of internally electrically initiated fires, and the supposition 1s made that additional fuel or ignition sources need not be considered. Appendix A to the Branch Technical Pos itfon APCSB 9. 5-1, Guidel1nP.s for Fire Protection for 11 Nuclear Power Plants," fssued in May 1976, and the proposed Regulatory Gu1de 1.120, "Fire Protection Guidelines for Nuclear Power Plants" both require that fuel, in addition to the cable insulating material,

..,,.. ~ 1--- "m m mm

• I

• uRNA.Mt:.. ....**....*.............................

mm -1

"" ---- mm I mmmmmmm*----1-- mm *mm*-*m I mm *m**""" m DA.TE.. .......... ........ .. ...............*.*......... ...... ... .......*....................... ***********************************

NRC PORM 318 (9-76) NRCM 0240

• u. l!I. oov .. RNMENT PRINTINO OFP"ICEt 111715 -1!12-24

Edson G. Case Robert B. Minogue NOV 9 1377 including external ignition sources be considered for generation of exposure fires. The data from the exposure fire tests show that additional protection, beyond the 5 ft. vertical and*3 ft. horizontal separation distance may be required between redundant safety dfvis1ons.

Work is scheduled to confinn generic barrier and coating designs and materials that will provide adequate protection for the redundant safety divisions when subjected to exposure ffres.

Original Signe(\Q Saul Levint ,,....,;,,)

Saul Levine, Dfrector Office of Nuclear Regulatory

... Research

Enclosures:

-~ :: 1. Research Support Branch Pl an

2. 8 Quick-Look Reports (Sandia Laboratories):

July 76, August 76, October 76, November 76, December 76, February 77, March 77, July 77 Distribution:

Subject CIRC CH RON Branch rf RF rf GLB cy LST cy TEM cy PGN cy SL cy Jupiter cy Beck cy

• See previous yellow for concurrences.

RE~

C~er 11/7 /77 OFFICE~ R.~.?. . ~ . ~?..~.'.. ~~*~*-* . ~;.~. ;J§R. ~.R.~.~ .................................... RES:RSR:D RES:RSR:W RES:SA:D R~

SURNAME~ R.f.J~tt. ;.P.~.t.~....... GL.Bn.e.ne.t.t:*.. Benn.ett/Iong T.(f.1~:~J::~;~:::::.: . ..::::::P.G.N:~~~f.~~ *: : : :~: : . . . . i:~:~::::

DATE~ 9./.Z~/.77. .... 9/28/77 ... 9128/.77. ..... 10/ 27/77 .. JJ..11. ../.Z?... . . . JJI1..!Z?..

NRC FORM 318 (9-76) NRCM 02.0 "fl U, 11. GOVERNMENT PRINTING OFFICll!:r 111711-112-M

---

~

5 FT. VERT I CAL SEPARATION BETWEEN SAFETY DIVISIONS 3 FT. HORIZONTAL SEPARA-TION BETWEEN SAFETY

.* ~I VISIONS >'

• .** -,)

I I

I I

_y ___ ~

~ ~

CABLE CABLE TRAY

-**----~

~~Ril~~~HSEPARA~ ~ 8 INC~_ HORIZONTAL SEPARATION BETWEEN TRAYS TION BETWEEN 1 ~ ~* ~

TRAYS I ~

-K"--

/~ IGNITION TRAY FLOOR LEVEL FIGURE I. HORIZONTAL OPEN SPACE CABLE TRAY ARRANGEMENT (STACKED TRAY TESTS)