ML17262A549
| ML17262A549 | |
| Person / Time | |
|---|---|
| Issue date: | 02/12/1979 |
| From: | Levine S Office of Nuclear Regulatory Research |
| To: | Harold Denton, Minogue R Office of Nuclear Reactor Regulation, NRC/OSD |
| References | |
| RIL-0046 | |
| Download: ML17262A549 (12) | |
Text
.. *** UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D. C. 20565 MEMORANDUM FOR: Robert B. Minogue, Director Office of St~ndards Development Harold R. Denton, Director Office of N~clear Reactor Regulation FROM: Saul Levine, Director Office of Nuclear Regulatory Research
SUBJECT:
RESEARCH INFORMATION LETTER # 46, "EFFECTIVENESS OF CABLE TRAY COATING MATERIALS AND BARRIERS IN RETARDING THE COMBUSTION OF CABLE TRAYS SUBJECTED TO EXPOSURE FIRES AND IN PREVENTING PROPAGATION- BETWEEN CABLE TRAYS (HORIZONTAL OPEN SPACE CONFIGURATION)"
References:
- 1. SAND78-0518, "Preliminary Report on Fire Protection Research Program Fire Retardant Coatings Tests, 11 March, 1978.
- 2. SAND78-1456 (NUREG/CR-0381), "Preliminary Report on Fire Protection Research Program Fire Barriers and Fire Retardant Coatings Tests, 11 September 1978.
- 3. Memorandum from Saul Le~ine to E. G. Case and R. B. Minogue,
Subject:
Research Information Letter #14 - "Physical Separation Criteria for Electrical Cable Trays (Horizontal Open Space-Confi guration), 11 dated November 9, 1977.
- 4. NUREG/CR0376, "Models of Horizontal Electric Cables and Cable Trays Exposed to a Fire Plume, 11 September 7, 1978.
1.0 INTRODUCTION
This memorandum transmits the results of a completed portion of the NRC Fire Protection Research program relating to the effectiveness of cable tray fire retardant coating materials and barriers in retarding the combustion within and propagation between horizontal cable trays. The research results include a test method that can be used to evaluate the relative performance of these protective measures. Data were obtained on the effectiveness of candidate
. fire retardant coating materials and barrier designs that are in use or are being con.sidered for nuclear power plants.
I.
I 1 J R. B. Minogue ~nd H. R. Denton 2.0
SUMMARY
I I The research results support the following conclusions:
- An acceptable test methodology was developed for use in evaluating fire retardant coating materials and barrier designs with hori-zontal cable trays.
There was a significant decrease in fire damage to the cable in both the fire retardant coating and barrier tests with cable qualified to the IEE_E-383 flame test standard.
- Moreover, with cable qualified to the IEEE-383 flame test all fire coatings tested prevented propagation between trays without the use of additional barriers.
- However, fire did propagate between cable trays with one fire retardant coating using cable not qualified to the IEEE-383 flame test standard.
- Fire propagation between cable trays was prevented with all barrier designs tested with.cable not qualified to the IEEE-383 flame test standard.
- All fire retardant coating materials and barriers studied offered some additional protection for the cables in a tray subjected to the test fires. *
3.0 BACKGROUND
- 3. 1 Support for Staff Positions The specific research upon which this RIL is based is outlined in the Research Support Branch Plan (Enclosure 1) for Fire Protection Research. The overall Fire Protection program is*
based on the research need identified in the research request SD 77-10 (August 19, 1977), the NRR program support letter from B. Rusche to H. Kouts dated June 25, 1976, the review of the Browns Ferry fire, as well as through consultation and formal review with the NRC user groups. The Fire Protection Research Review Group has been and continues to be the focal point for both the formulation of the research program and evaluation of program results that form the basis of this RIL.
(Related to this is the NRR/RES task force to evaluate the impact on the fire protection research program of includi_ng full-scale replication tests.)
.~. .....; ~* .-. ......: .. - :
R. B. Minogue and H. R. Denton The specific user requirement for the research conducted to date with cable tray fire retardant coatings and barriers is based on the need to provide test data to confirm the guidance given in Appendix A to the Branch Technical Position APCSB 9.5.l, "Guidelines for Fire Protection for Nuclear Power Plants," and in the draft Regulatory Guide 1.120, "Fire Pro-tection Guidelines for Nucl.ear Power Pl ants. 11 Regulatory Guide 1.75 ("Physical Independence of Electrical Systems")
requires only consideration of electrically initiated fires without consideration of any fuel source other than the cable insulation itself. The staff position for evaluation of the fire protection capability is that exposure fires must be considered. The type and size of the exposure fire are 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 conclu-sions presented in this RIL comprise an assessment of the effectiveness of cable tray fire retardant coatings and barriers in preventing the propagation of exposure fires in horizontally oriented cable trays.
- 3.2 Testing Rationale Previous exposure fire testing showed that even with cable qualified to the IEEE-383 flame test additional protection beyond the 5-ft (l.5-m) vertical and 3-ft {0.9-m) horizontal separation distance defined in R.G. 1.75 may be required between redundant safety divisions. Work was undertaken to study the effectiveness of cable tray fire retardant coating materials and barriers in preventing the spread of fire b~
tween redundant safety divisions.
- Specifically, the scope of this RIL covers evaluation of the effectiveness of certain cable tray fire retardant coating materials and cable tray barriers in retarding combustion and preventing propagation of closely spaced cable trays subject to exposure fires.
R. B. Minogue and H. R. Denton 4.0 TESTS 4.1 Test Configurations The test configuration simulates cables placed in the horizontal open space of a plant where the effects of reflected heat from the walls and ceiling are minimized. One and two tray tests were conducted each using 18-inch-wide (0.45-m) open-ladder steel cable trays 12-ft (3.7m) long. As with previous hori-zontal open space cable tray tests the separation between trays was 10.5 inches (26.7 cm).
Fires were started using both propane burners and diesel fuel.
The propane burner teit fire had previously been shown to be of sufficient intensity and size to consistently result in a fully developed fire in the ignition tray (SAND71-1424).
Measurements made of the tes*t diesel fuel fire show that it is comparable to the propane burner test fire in terms of heat flux produced if the fire surface area and burn times are similar. Both ignition sources exposed a 3-ft. (0.9-m) section of the cable tray to the flame. Air flow was maintained to simulate normal ambient plant air flow in plant open space areas. All tests used cable trays that had a random fill pattern with the cable loaded to the top of the side rails.
This resulted in a cable loading less than the maximum volumetric loading of 40% permitted by the Insulated Power Cable Engineering Association. Previous tests showed that the random fill patterns (with minimum fuel density and maximum air space in a fully loaded cable tray) represented the worst allowable case with regard to combustion and propagation and was used on all cable tray coating and barrier tests.
Tests were conducted with cable that was qualified to the flame test standard of IEEE-383-74 and also with cable that could not qualify to this standard.
In an effort to obtain a more basic understanding of the combustibility of candidate fire retardant coating materials,
- _, small-scale furnace tests were conducted on all coating materials used in the full tray tests. The tests were conducted using a 6-inch-square (15-cm-square) sample of coated cable *. The test furnace provided controlled air flow and radiant heat flux.
The specimens were heated in a controlled temperature environ-ment in the presence of a pilot ignition source and basic combustibility measurements made* to establish the time to ignition, time to maximum heat release, and cumulative heat
'release. Radiant heat fluxes similar to those encountered in the full tray .ignition testing were used (approximately 4 watts/ crrf)
- R. B. Minogue and H. R. Denton The full-scale test used six different fire retardant coating materials; five of the six coatings were sprayed on to the
- manufacturers' recommended wet thicknesses. The sixth fire retardant coating was applied with a trowel in accordance with the manufacturer's specification. In addition -to the cable tray fire retardant coatings, tests were conducted using solid bottom cable trays wi thou*t covers, sol id bottom cable trays with vented covers, cable trays with solid covers (open ladder bottom), ladder cable trays covered with a ceramic fiber blanket, and two open ladder cable trays with a solid fire barrier between trays. The full-scale testing covered both one- and two-tray tests. The single tray te.sts were conducted to study the degree to which combustion was retarded and the two-tray tests were conducted to study the degree to which propagation between trays was prevented. (N.ote: In the two-tray tests, a barrier above the upper tray simulated the boundary of a third tray.) Propane burners and diesel fuel exposure fires were utilized. The propane burners were arranged to provide an exposure fire identical to that utilized in prior unprotected cable tray tests, and as in previous testing the propane burners were used in 5-minute burn cycles with a*
thermal barrier above the ignition tray during the burn period.
The diesel fuel exposure fires were contained in an enclosure 36 inches by 18 inches (0.9-m x 0.4-m) the same distance below the tray as the propane burners. Two gallons of fuel were ignited to create a fire which burned for about 13 minutes.
(There was no thermal barrier above the ignition tray for the diesel fuel fires.)
- 4.2 Test Specimen Standards The significance of the research covered by this RIL should be viewed primarily as the development of a test methodology by which passive fire protective measures can be evaluated. The tests developed can be performed by suppliers and plant operators to justify alternate fire retardant coatings and barriers not tested by NRC or to demonstrate the effectiveness of those measures tested by the NRC in situations where the des_ign basis fire is significantly different than the test case fires used in this research. Furthermore, suppliers of fire retardant coating and barrier materials may change their. products or recomiTiended practices which may requfre further verification of their effectiveness. In conducting this research, currently accepted design practices and materfals were used when available.
In the absence of accepted design practice, supplier recommendations were used. With the exception* of the sol id barrier between trays, no NRC test guidance is currently available for any of the passive fire ~rotective measures evaluated.
e
- R. B. Minogue and H. R. Denton 4.3 Test Limitations In evaluating these or other tests results for the fire retardant coatings and barriers, particular care should be exercised with regard to (1) the thickness of the fire retardant coating materials used, and (2) the method of installation of the
- ceramic fiber blanket. No attempt was made to determine the amount of uncovered blanket or degraded fire retardant coating that would alter the test results with regard to the propa-gation of a fire between cable trays, nor was there any attempt to evaluate in a quantitative way the significance of the vent area in the test of solid bottom trays with vented covers.
Unrestricted use of the cable tray fire retardant coating test results also requires caution since there are many untested variables which can influence the selection of a cable tray coating. The following items were not covered in the research compl.eted thus far:
- l - No evaluation has been made of the aging characteristics of the fire retardant coating materials. This topic has a number of different aspects. There is the possibility of a long term reaction of the fire retardant coatfog material with the cable jacket. There is also the question of chemical and mechanical stability of the fire retardant coating material itself as it ages. The fire retardant coatings ~11 have different trace materials and exhibit different mechanical properties when cured. The significance of these differences was not evaluated, although work is planned on the aging of fire retardant coating materials.
2 - The method of installation of the fire retardant coating requires specific consideration and depends upon the supplier's recorrmended thickness for cable tray application and the method of application (i.e., sprayed or troweled). The wet consistency (which may be affected by shelf life) may have a significant effect on the ability to apply the material evenly.
If the material is sprayed, the spraying equipment may also play an important role in obtaining a uniform layer of coat-ing. The current practice is to measure the thickness of the wet coating material on the top cables. However, depending on the spraying equipment used (or trawling technique) and the wet consis.tency of the material, the degree to which the coati.ng material is forced down into the tray between cables will vary. The d.egree to which the coatfog material can fill the voids in the cable tray was shown to be an important
- variable affecting the ability of the coated cable trays to resist damage from exposure fires.
r .
R. B. Minogue and H. R. Denton
- 3 - No attempt was made in this research to evaluate coati.ngs
with equal thickness applications. The research was intended to be confirmatory of the products being sold and with the specifications recorrmended by the supp.lier.
4 - The question of intumescence was not evaluated except as it affected thereffectiveness of the fire retardant coatings with the fully developed test fires. It is possible that the properties of a coating could change because of intumescent behavior before it is exposed to *the intense portion of the fire. The fire retardant coatings were not evaluated with
. regard to their effectiveness i~ slowly developing fires.
Also the slowly developing fire might also be important for a fire redardant coati.ng that depen.ds primarily on the release of water and or gas to retard the combustion of the cable.
5.0 RESULTS Prior test data and the RES evaluation in RIL #14 (Ref. #3) showed that the most important factor in determining if a cable tray fire will develop is the spacing between cables within a given tray since this spacing establishes the air-fuel mixture at the burning surface. It was established by measurement and observation of the prior test films that the fire propagation from one tray to a tray above it depended on the collection of hot gaseous fuel released from the cable exposed to fire. Fire propagation occurred by ignition of the gaseous fuel above the cable tray to which the fire is spreading, with a subsequent spread of the flame down into the tray itse 1f.
Subsequent to the writing of RIL #14 this conclusion has been verified by additional tests without fire retardant coatings and barriers and also during the fire retardant coating and barrier tests at Sandia Laboratories. Furthermore an analytical .report NUREG/CR-0376 (Ref. #4) "Models of Horizontal Electric Cables and Cable Trays Exposed to a Fire Plume" completed by the Applied Physics Laboratory (APL) verifies these conclusions. The APL research provides mathematical correlation of the experimental determined favorable conditions for the development and propagation of fires .in horizontally oriented cable trays. Specifically the equation governing mass flux of the flammable gas indicates that it is roughly propo.rtional to the mass flux of fire plume gas from the tray whose bottom side is being heated by the ignition tray. It is the spacing between cables that enables the fi're plume gas to pass through the tray and collect above it
- I --- . *- ... -~ ,., -- .. *' - . *..... ' *-- - -- .
-~.~ . ~ *~
I.
R. B. Min_ogue and H. R. Denton These experimental and analytical conclusions verify that the Sandia fire retardant coating and.barrier tests covered worst case conditions for horizontally oriented cable trays insofar as the cable spacing and loading are concerned.
The following specific conclusions can be drawn from the tests and analysis conducted Ref. (1) and (2):
1 - All coating materials studied offer some protection against the
-
- test fires; however, there was a wide range of effective.ness among the coatings in both their ability to retard combustion when
- exposed to a fire and in their ability to prevent fire propagation from one tray to another. Two of the coati_ngs tested (type* D and type E) prevented _ignition for six five-minute burn cycles.
2 - With cable qualified to the IEEE-383 flame test standard, all coatings tested prevented propagation between trays; however.
with one *coating tested {type C), relatively high cable temperatures (in excess of l000°F) were measured in the upper tray. With another coating tested (type A), cable temperatures in the upper tray in excess of 800°F were measured.
3 - With cable not qualified to the IEEE-383 flame test standard, one of the coatings tested (type C) did not prevent fire propaga-tion between trays.
- 4 - There was good correlation of data between (1) the small-scale, (2) single-tray, (3) two-tray propane burner, and {4} two-tray diesel fuel fire tests indicating that any of the test procedures developed as part of this research can be utilized for further evaluation. However, verification of the ability of a cable tray fire retardant coating or barrier to prevent tray-to-tray propa-gation can only be accomplished with one of the two tray tests
- developed. A comparison of the diesel fuel fire and the propane burner fire showed that if the surface area and burn times of the fires are the same the two are very similar with regard to the heat flux produced.
- 5 ~ All barrier designs studied offered some protection against the test fires. For the tests of propagation through barrie.rs, the experiments were conducted only wfth cable tha*t is not qualified to the IEEE-383 flame test standard, and in all cases propagation was prevented.
- 6 - With all barrier designs tested with cable that is not qualified to the IEEE-383 flame test, fire did develop in the _ignition tray in either one or two-burn cycles.
R. B. Minogue and H. R. Denton
- 7 -,'The barrier designs ("sol id bottom no cover" and "sol id bottom vented cover") and the fire retardant coatings (type D and E) tested in single tray tests with cable qualified to the IEEE-383 flame test offered protection against the ignition and electrical failure in the cable tray exposed to the test fire.
8 - There was a distinct improvement in performance in both the coating and barrier tests with cable qualified to the IEEE-383
-flame test standard.
6.0 EVALUATION AND RECOMMENDATIONS The data presented in this RIL, and an advance draft copy of the RIL, have been reviewed with members of the Fire Protection Research Review Group. There is general agreement with the contents of this RIL with the exception of the RES *recommended minimum thickness and inspection to guarantee total blockage of all space between cables.
The data discussed in this RIL are of primary concern in those cases where a permanent rated barrier has not and cannot be utilized between safety divisions. Previous testing as discussed in RIL #14 (Reference 3) showed that use of cable qualified to the IEEE-383 flame test standard is not in itself a sufficient defense against the propagation of fire across a 5-ft (1.5-m) vertical and 3-ft (0.9-m) horizontal separation distance if exposure fires are con-sidered credible.
The data discussed in this RIL showed that protection against the propagation of fire between cable trays can be obtaine*d with both cable tray coating materials and various cable tray barrier designs.
However, there is a distinct improvement in the degree of prote*c-tion offered when cable qualified to the IEEE-383. flame test standard is used.
For those cases where the cable is not qualified to the IEEE-383 flame test standard, all of the coating materials and barrier designs tested exhibited poorer perfo.rmance with regard to their ability to retard combustion. However, with the exception of coating Type C all coatings and barriers prevented fire propagation between cable trays. *
- If fire retardant coatings are utilized the quality control of the application is importarit. Based on the experience gained in con-ducting the fire retardant coating tests at Sandi~ and observations of ac.tual cable tray appl ication*s it is concluded that a minimum
'*.; thickness should be established for use of fire retardant coatings
.i in nuclear power plants.
- R. B. Minogue and H. R. Denton Therefore, if cable tray coatings are used it is recommended that a 1/8-in (0.3-cm) dry thickness be required. The thickness recommended by the suppliers of coatings A, B, C, and G is 1/8-in (0.3-cm) wet.
The suppliers of coatings D and E recommended a coating which is 1/4-inch (0.6-cm) thick when wet. The supplier of criating E changed this thickness specification to 5/32-inch (0.4-cm) wet before the last portion of the test program without any noticable change in the relative performance of the coatings tested. However, if all coatings are required to be applied to the same thickness it may be that the differences in their effectiveness will be reduced.
Because of application and the amount of water in the wet condition (30 to 40%) it is unrealistic to expect thickness tolerances better than + 25% after drying.
Based on the experience gained in conducting the cable tray fire test and fire retardant coating tests at Sandia, it is concluded that whatever protective measure is utilized it should present a solid barrier to the fire and not allow the passage of combustible
. gas through the cable bundle within the tray. If fire retardant cable tray coatings are used consideration should be given to surrounding the cable with the fire retardant material, and a suitable check should be required to ensure that all air passages through the cable bundle have been blocked by the fire retardant coating material. A visual check above the cable tray with a high intensity lamp placed below the tray may be useful.
- Fire retardant coatings and barriers can be utilized to prevent tray-to-tray fire propagation with cable capable of passing the IEEE-383 flame test standard with cable tray configurations and fires similar to those on which the fire retardant coating tests were conducted. For coating type C which exhibited only marginal performance with cab 1e qua*l ifi ed to the IEEE-383 flame test standard at the nominal recommended thickness of 1/8-inch (0.3-cm) wet, verification from applicants using the Sandia two-tray configuration should be requested with the 1/8-inch (0.3-cm) dry thickness to ensure an adequate margin in preventing prop.agation.
Fire retardant coating types A, B, D and E can be utilized to prevent tray-to-tray fire propagation with cable not qualified to the IEEE-383 flame test standa.rd with cable tray configurations and fires similar to those on ~hich the fire retardant criating tests were conducted. *
- -~- -.;.---......-.-** '-'<-:,*~ ..:.._, -*-*****"' *. ~-**- *~ -**
R. B. Min_ogue and H. R. Denton The barrier designs tested can be utilized to prevent tray-to-tray fire propagation with cable not qualified to the IEEE-383 flame test standard with cable tray configurations similar to those on which the barriers were tested at Sandia.
7.0 COORDINATION CONTACT For coordination of any further evaluation of these results and for discussion and future experiments the reader is advised to contact Mr. Ronald Feit, Fire Protection Research Program Manager, RES, telephone 427-4272. *
- k-d~*
Saul Levine, Director Office of Nuclear Regulatory Research
Enclosures:
As stated
- ~
1
- i
~
..*,.~-,
~
- ' * -.R.
- B. Minogue .and. H. R*. Denton
.f
.I . *' i*...
-~
- The .bar.rfer designs ~ested can be utilized to prevent tray-to-triy *; : : * *
- '* . . ' ffre propagation wfth cable not qualified to the IEEE-383 flame
- test standard with cable tray configurations similar to those on *-* .:* ' :, .
- . wh1Ch the barriers were tested at Sandia. -* . * . ~
-' ~\ ,.
7.0 COORDINATION CONTACT For coord1.nation of any further evaluation of these results and for discussion and future experiments the reader is advised to contact Mr. Ronald Feft, Ff re Protection Research Program Manager, *RES,
-*Telephone 427-4272.
- Original Si~ ii Saul Lev.lDJt ~
Saul Levine, Director Office of Nuclear Regulatory Research
Enclosures:
As stated bee: ACRS PDR DISTRIBUTION RECORD NOTE.: Comments were received SUBJ and incorporated from SD (Notley).
CIRC Comments were received from Matthews (DSS)
CH RON -and incorporated. The only significant
/ co1TU11ent not incorporated was the recommenda-.
- l
. .' BRANCH RF
. RFeit RF tion by Matthews that the minimum thickness ... --*
RFeit CY and applications recommendations be d~Jeted.
GLBennett CY CEJohnson CY*
LSTong CY TEMurley CY RScroggins CY RBudn i tz CY -.
. Slev*i ne CY
/
/' .I l
. RSR:W: RS (})J_. * ,/, .*
- RFeit/pg(~
~
- .... l'f- l'l'f ..
.