ML17354A430
| ML17354A430 | |
| Person / Time | |
|---|---|
| Site: | Turkey Point  |
| Issue date: | 03/07/1997 |
| From: | FLORIDA POWER & LIGHT CO. |
| To: | |
| Shared Package | |
| ML17354A428 | List: |
| References | |
| PTN-BFJM-96-028, PTN-BFJM-96-028-R00, PTN-BFJM-96-28, PTN-BFJM-96-28-R, NUDOCS 9703130426 | |
| Download: ML17354A430 (66) | |
Text
Page i CALCULATION COVER SHEET Ca 1 cul ati on No: PTN-BFJH-96-028
Title:
Fire Barrier ACF for T-La 330-1/770-1 Assemblies Initial Issue Descri tion REVISIONS Form 82A. Rev 6/94
'F703130426 '770307 PDR ADOCK 05000250 i P PDR
0 Page ii LIST OF EFFECTIVE PAGES Ca 1 cul ati on No. PTN-8FJH-96-028 Rev.
Title Fire Bar rier ACF for T-La 330-1/770-1 Assemblies Pa e Section Rev. Pa e Section Rev.
.1 0 ~
ll 0 ill 0 1 1,2,3 0 2 4 0 3 '4,5 0 4 5 0 5 5 0 6 5 0 7 6 0 Al (23 Pages) 0 Form 828. Rev 6/94
Page iii TABLE OF CONTENTS CALCULATION NUMBER PTN-BFJM-96-028 REV.
SECTION TITLE PAGES Cover Sheet List of Effective Pages Table of Contents 1.0 Purpose/Scope 2.0 References 3.0 Methodology 4.0 Assumptions/Bases 5.0 Calculation
'6.0 Results ATlACH NO. T-ITLE PAGES Fire Endurance .and Ampacity Testing of One 23 and Three-Hour Rated Thermo-Lag Electrical Raceway Fire Barrier Systems Form 82C, Rev 6/94
0 0
0
CALCULATION SHEET CALCULATION NO. PTN-BFJN-96-028 REV 0 SHEET NO. 1 1.0 Purpose/Scope Ampacity correction factors (ACFs) were developed for Turkey Point use based on TSI Thermo-Lag (T-Lag) 330-1 material for 1-hour and 3-hour rated fire barriers (References 2. 1 and 2.2). The purpose of this calculation is to develo ACF values t'or T-Lag 330-1 with T-Lag 770-1 overlay (330-17770-1 composite) assemtlies for 3-hour rated fire barriers for power cables.
2.0 References 2.1 Calculation PTN-BFJM-96-005. Revision 0. "Fire Barrier Ampacity Correction Factors-Extrapolation of Test Results for 3 Hour Barrier" 2.2 Evaluation JPN-PTN-SEEP-96-011. Revision 0, "Review of Ampacity Rating for Power Cables in Conduits and Trays with Thermo-Lag 330-1 Covering" 2.3 Report, "Fire Endurance and Ampaci.ty Testing of One and Three-Hour 'Rated Thermo-Lag Electrical Raceway Fire Barrier Systems" (included as, Attachment 1) 3.0 Hethodology The approach of Reference 2. 1 was to use heat transfer analysis to extrapolate from Texas Utilities Electric test data and develop ACFs for single conduit. cable tray and banked conduit. Heat transfer was calculated per foot of raceway length as a convenient relation to,the test results. The analysis was taken a step further by Reference 2.2 which evaluated power cable ampacity ratings.
This calculation begins with a similar approach to that of Reference 2. 1 and evaluates the effects of augmenting existing T-Lag 330-1 material with T-Lag 770-1 material. However, this evaluation continues further by also considering the findings of Reference 2.3 to establish appropriate derating factors based on testing and to include boxes among the assembly configurations.
The analysis starts with the values calculated in Reference 2. 1 for trays and conduits, then considers the ampacity effect of overlaying with T-Lag 770-1 wrap.
As such. the heat transfer equation used in Reference 2. 1 is revised as follows:
q
~ (T,-T,) / (R5 + R + R, + R5) where: g . Rate of heat transfer from raceway Temperature of conductor (fixed 9 90'C/194'F)
T, = Ambient temperature (fixed 9 40'C/104'F)
= Thermal resistance of all items within the raceway R,
- Thermal resistance of the air gap between the raceway and the T-Lag R, = Thermal resistance of either T-Lag 330-1 alone or with the T-Lag 770-1 overlay R, = Thermal resistance at the surface of the raceway The heat transferred from the raceway under steady-state conditions is essentially equal to the I'R losses within the conductors. These heat transfer values were determined from the test data based on the measured current and size of conductor used; as documented in Reference 2. 1.
Form 83, Rev 6/94
0 0
CALCULATION SHEET CALCULATION NO. PTN-BFJH-'96-028 REV 0 SHEET NO. 2 The thermal resistance values of R,. R.. R, and R, are also based on those determined in Reference 2. 1 for 3-hour>ated t-Lag 330-7 assemblies. The'value for R, with the T-Lag 770-1 overlay will be calculated assuming that the thermal conductivity for T-Lag 770-1 is the same as for T-Lag 330-1. After thermal resistance values have been established. the heat transferred can be calculated for the raceway with the 3-hour rated 330-.1/770-1 composite fire barrier.
Since the heat is a function of the current squared. the ampacity correction factor (ACF) will be determined by the following relationship.
ACF = I,/I = (q,/q)" where the subscript "p" refers to the protected raceway 4.0 Assumptions/Bases 4.1 The effects of inductive losses in the raceway and cabl'e sheath are negligible with respect to applying the test data to the Turkey Point configurations.
4.2 Surface emittance for cable, raceway, and T-Lag is assumed to be equal to 0.9. Note that a high emittance value will reduce the heat transfer at the surface having an overall effect of maximizing the ampacity de-rating.
4.3 Heat transfer is assumed to flow perpendicular to the surface. This allows one-dimensional analysis, is reasonable for single conduits. and is conservative for cable tray and banked conduits because no credit is taken for heat flow through box
~ .. corners'eat transfer through the sides of cable tray is assumed to be zero. This reduces the heat transfer equation for tray to one dimension. As the tested cable tray is relatively wide (24"), this is expected to be a good approximation for all cable tray. Also, the horizontal tray configuration is more conservative than vertical in terms of reduced heat transfer, based on typical emperically developed formulae for natural convection coefficients over horizontal and vertical plane surfaces.
4.5 Convective heat transfer flow is assumed to be laminar. This is reasonable for "stagnant" rooms and conservative in outdoor areas and rooms with forced ventilation, in terms of reducing the heat transfer rate.
4.6 Pre-buttering with trowel grade T-Lag material is assumed to virtually eliminate air gaps between T-Lag layers such that interface resistance effects are negligible.
4.7 The 3-hour...T-Lag 330-1 assembly is assumed to be at the 1/" nominal thickness in accordance 'with the manufacturer's tolerance (+ 4"). It was judged to be excessively conservative. considering application techniques and material cost, to assume a maximum thickness'f l~g".
4.8 Each layer of T-Lag 770-1 material is assumed to have a finished nominal thickness of 3/8" (including buttering). Two or more layers of wrap are assumed to be.
consistent with the tested configurations (Reference 2.3).
4.9 Raceway is made of rigid steel. which is typical for power. plant installations.
~ 4. 10 Conduit which is banked in tray.
section and it a single plane Both configurations involve is conservative to a cable assume can be assumed an mass air to arranged in be a
equivalent to cable shallow rectangular gap between cables in the tray and the T-Lag material.
Form 83, Rev 6/94
0 0
0
CALCULATION SHEET CALCULATION NO.~ PTN-BFJM-96-028 REV 0 SHEET NO.~ 3 4.]1 The thermal resistance values for all items within the raceway and for the gap between the conduit and the T-Lag material are assumed to remain constant as additional thickness of T-Lag is installed. Considering that the geometry of, these areas is not changed, this approximation is ,reasonable for the purpose of extrapolating the thermal resistance from raceway with T-Lag 330-1 only to raceway with composite 330-1/770-1 wrap.
- 4. 12 The thermal conductivity of T-Lag 7?0-1 material is assumed to be the same as for T-Lag 330-1. This is a reasonable approximation since the two materials seem to be chemically similar. Also, based on discussion with Kent Brown (co-author of Reference 2.3), material property data provided by TSI has been inconsistent, and is considered unreliable.
- 4. 13 The radiant heat from the sun does not adversely impact ACF values for wrapped components because the insulating-effect of T-Lag reduces sun load with respect to an exposed component.
5.0 Calculation Formulae-, constants and parameter values are as presented in Reference 2. 1 unless noted otherwise. ,Detai'led calculations are performed by spreadsheet and the results presented in the table at the end of this section.
5.1 Determination of test heat loads Test heat losses for unwrapped raceway were calculated (q=I'RN) with the following results:
Raceway (Conductor ) Heat/Ft Size BTU/Hr (1-3C/¹'10) 22. 6 tl 2
Il (1-3C/¹6) 23.7 5 (4-750 kCNil) 99.7 Tray (126 -3C/¹6) 382.1 5.2 Determination of Thermo-Lag R values (R,)
For heat transfer through T-Lag cylinder:
R, - ln(r,/r,)/2vkL r0 Outside Radius r Inside Radius Thermal Conductivity = 0. 1 BTU/Hr-FT-'F kayo kv Thermal Conductivity - 0. 1 BTU/Hr-FT-'F L Length = 1 Ft. (Per, Foot)
For heat transfer through T-Lag sheet:
R,
= l/kA 1 Thickness 4~ Thermal Conductivity = 0.1 BTU/Hr-FT-'F kno .
Thermal Conductivity. - 0. 1 BTU/Hr-FT-'F A Surface Area Form 83, Rev 6/94
0 0
CALCULATION SHEET CALCULATION NP. PTN- BF JN-96-028, REV 0 .SHEET NP. 4 5.3 Determination of surface R values (R,)
The surface resistance considers free convection andiradiation heat transfer:
=
qs qc + qr
= .Heat transferred by convection q,
q
- Heat transferred 'by radiation For free (laminar flow) convection:
q, - AT 0.27(hT/l )"MT Convection heat transfer coefficient:
Horizontal cylinders in air h 0.27(BT/L)~
Vertical planes in air h =
planes facing 0.29(hT/L)'orizontal up in air h 0.27(hT/L)"
A Surface Area L Characteristic length in feet -(diameter or width)
For radiant energy:
0 cr
,A
,q, = crAe(T,'-T,")
=
=
1.714x10 Constant Surface Area
'TU/Kr -Ft'-R4, Stefan-Bol tzmann Surface Emittance = 0.9 T = Absolute Temperature. R (460 + 'F)
For total heat transferred from the surface:
= 0. 27(hT/L) "MT + 1. 714x10'(0. 9)A(T,'-Tz')
q, q, = f0.27(DT/L)" + 1.714x10'(0.9) (T,'-T,') MT]MT hT/qs Rs 1/ LO 27(GT/L) + 1 714x10 (0 9)(Ti Tz )/hT 5.4 Calculation of ACF The ACF is 'calculated in accordance with the methodol'ogy described above. A description of the headings follows:
OD/M This is an input value of the conduit outside diameter or cable tray width.
Conduit diameters were developed in Reference 2. 1.
TH This value is the thermolag thickness. For each raceway size a thickness representing 330-1 wrap (TH330) only and the 770-1 wrap (TH770) is entered.
ODT This is the outside diameter of the raceway with calculated from the OD and TH. For cable tray QD'is not calculated since, it wi;11 always be equal to W.
Form 83, Rev 6I94
0 CALCULATION SHEET CALCULATION NO. PTN-BFJM-96-0 8 REV 0 SHEET NO. 5 0 A The outer surface heat transfer area. Note that for raceway both the top and bottom areas are included. Area is calculated on the basis of a one foot length of raceway.
Ri Insi~thermal resistance as defined above. The value was calculated from the test data in Reference 2. 1 and is not recalculated'ere.
Rg Gap thermal resistance as defined above. The value was calculated from the test data in Reference 2. 1 and is not recalculated here.
Rt Thermo-Lag thermal resistance. The value .is calculated in accordance with the following equations which were developed above.
Conduit Rt = 1n(ODT/OD)/2rrk k = 0. 1 Tray Rt = TH/kA k '0.1 Rs Surface thermal resistance is calculated in accordance with the following equations which were developed above. Note that the BT in this equation, is between the surface and ambient and the 7'alues must be in 'R. The ambient temperature used is 104'F/564'R.
,Rs = 1/[0. 27((Ts-104) /ODT)'+ l. 714xl0'(0. 9) ((Ts +460) 4-5644) /(Ts-104) 3A Ts Surface temperature of T-Lag, determined by iteration until q
- q'.
q Heat transferred, cal'culated as follows:.
q
= hT/(Ri + Rg + Rt + Rs), Where hT = 90 F q'eat transferred from the surface. calculated as follows:
q' hT/Rs, Where BT = Ts - 104'F From continuity, the heat transferred from the surface is the same as the total heat transferred. In order to solve the various cases. Ts is adjusted by iteration until q - q'.
ACF Ampaeity correction factor calculated by the following equation which was developed above.
ACF (q,/q)~
The ampacity correction. factors for 3-hour T-Lag 330-1/7?0-1 composite assemblies extrapolated'y calculation are as follows:
Item ACF 3," Conduit (2 wraps) 0,84 F" Conduit (3 wraps) 24" Tray 0 '3 0.54 Banked Conduit Form 83, Rev 6/94
Cl 0
I i 1 I ~ ~ ~
) ~
~RIREBIKHI~~~~~~~~~~KKK
~WWW~~~~~~~
~KKRKRZRMRMRRlaRISEEIEKESIQEBSKHES~HERIKBZa~
EiiKZIQESRK$9l&3WKSK'E5%RXER&KHME%&KRRE~RHHSRHH5&K%
KEG'HQK'lKKQ'RHEKE5%KKKRKK5iR&KQRKEESR~ZHRRERE%%5EESRKMK'Egg QQiigsmgi@iigi~i'gQ~~~~~ggQ~QQ~~gQQQQJ~~ggmmiigg EK~KHRRRHE9RRRRKR%KKCKRM2!%~~ME'%%3KRKEEKQKKERNRK%
~WWW~~~
MHRRRESEIQKEBRXGRREKRKRHSKKSRKMRKMRHK1KXRSK'ERR KRSRQRBHIHKKSRxMRKGRRB&RKMREkRRR)28%KEQKK~KIKEHQK~EB~Q KRRZQHH)U~~m<<RESa3Umammaxmam5mazmazmaIEEI25aK~2RREEE~xm5%
~~~MR~~
%52tSM9KRH9RKt29RXEERISKkRKE2%RXGRREWRZERIMElEREKSRHEIRR2%
KRR~RKRKEHREHIRERRRKESRXEQRXKSK'EERREHSIKKQ'RRXRRZRIRZE5 ERRZRIKZlHRUCKBRXQHRZKSKKER~EWKZMREHSNHE15RKRRRKGSMIM aEZmiK23iEHRH99aKmaERaaKKamZmaKHaaEHXaEIEEIatE9%SEKRWRXZa EHQ>55iK23i5EHRE3M'ZZQRKKRKKESRXBRRRkMRXERKKER1SH'E5528KEKEEg
~RiRRIR~~~MHM~~WEW
~gggggg~~~s~ggggQgggmgggJgmgggg~~ggrimgglggg~MEW~
amiaEIRlKel~~aKeamKHam5mmxamHarHKKSRma mam REfRBK8IKW~RK9%
~+igigggiig~igiJ~~~i~ii i i RRSRRRSRRRRERNHKH NEER RES
~i~i~igg+~i.g+gp~i: ~i~i gg~i~~i.g~
~
0 0
~O
CALCULATION SHEET CALCULATION NO. PTN-BFJM-96-028 REV .0 'HEET NO.
TYA performed ampacity tests as described in Reference 2.3. The testing was performed using an even number of conductors. pre-formed T-Lag 330.-1 dry-fit to the component. and other such criteria per IEEE P-848 Draft 14. TVA identified key parameters. presented a reasonable rationale for:what constitutes conservative test configurations, and maintained key parameter limits while establishin ACF value for these configurations based directly from testing.
The ampacity correction factors tor 3-hour T-Lag 330'-1/770-1 composite assemblies based on tested configurations are as follows: .
Item ACF Conduit 0.87 Tr ay 0.52 Banked Conduit 0.74 Although actual testing of conduits produced an ACF of 0.87, TVA established a value of 0.82 as their standard to allow for variations in emissivity for i eren t or differe conduit materials used.
Banked conduit testing was performed in box-type configurations. The ACF value of 0.74 is appropriate for banked conduit assemblies and conservative for larger boxes.
Therefore, since increasing box size increases convective cooling rate and the void fraction is larger than typically found for cable trays, an ACF value closer to that for banked conduits is reasonable for wrapped electrical boxes. Accordingly, an ACF v'alue of 0 '0 is selected for boxes.
Also, the tested configuration for 1" conduit appears to be with 3 wraps of T-Lag 770-1 material. whereas Kent Brown (co-author), indicated that only the 2-wrap configuration was used for ampacity testing. ,Even so. the di.fference in calculated values, is not significant, and is conservative with respect to test values.
6' Results Calculated results for conduit and cable tray are reasonably close to those established through testing. However, reliable testing automatically accounts for the myriad variables and the effects of their interactions, whereas calculated results depend on the validity of the model analyzed as well as the assumptions employed. This is particularly the case here with the assumptions for material thermal conductivity and using a conservative value for emissivity. As such, test data is considered to be a more rel.iable basis for determining ACF values. Since conduit testing included only 1" and 4" conduits, the calculated value is used for the 5" conduit. Therefore, the following ampacity correction factors shall be used for 3-hour rated T-Lag 330-1/770-1 composite assemblies:
ACF Conduit (<4") 0.82 Conduit (5") 0.?4 Tray 0.52 Banked Conduit 0.74 Boxes (selected) 0.70 Form 83, Rev 6/94
41 0'
~v~e I SSyy Oa 4V\ 813052488231 ' 2 Calculation PTN-BFJM-96-028 Revision 0 Attachment 1 FIRE ENDURANCE AND AMPACITYTESTINCPQF1 o< 23 ONE hND THREE HOVR RATED THERMO-LAG ELECTRICAL RACEWAY FIRE BhRRIER SYSTEMS Mark H. Salley Kent O'. Brown Senior Engineering Specialist Senior Elccuical Engineer Tennessee Valley Authority Tennessee Valley Authority ABSTRACT The Nuclear Regulatory Commission (NRC) has determined there werc dcffciencics in the design, testing and installation of Ihermo-Lag 330-1 Electrical Raceway Fire Barrier Systems (ERFBS). Their two primacy concerns are:
~ The currently installed,Thermo-Lag 330-1 ERFBS may not ensure that tho eiecuical cables aro Gee from fire damage for the required rating and, The ampacity derating factors currently used for Thermo-Lag 330-1 ERFBS may not bc correct.
Thermo-Lag 330-1 has been utilized in thc design of the Tennesseo Valley Authority's (TVA) Watts Bar (WBN),
Sequbyah (SQN) and Bmwns Ferry (BFN) Nuclear Power Plants to protect redundant cables required for safe shutdown.
of th) reactor in the event of a fire. Such protection is a licensing requirement per the Code of Federal Appendix R. After reviewing other avaiiablo ERFBS options, TVA determined that redesign and Rcguiationa,'0CIIR50, qualipcation of the Thermo-Lag 330-1 systems would provide the best available',ERFBS and minimize cablo replacement due to ampacity effica'-. Thc'to+Vmj.fire'and arnpacity. test prague'mYdeveloped one and three-hour rated Thermo-Lag ERFBS for cable trays condditiI,'iutdropj'ifaCtion'b'oxes an@ichir'uniqtio raceway configurations. The imptcIved designs and installation.iec&illucs.also Siuilinad ilrorrettuired.u+kip'hiaimg. These new designs will be used in inidal instalhtipns at WBN and for upgrading other systcins'g other TVANuclear Power Plants (NPP), The TVA research. and testing'also pr'ovided useful input to thc NRCs criterion for ERFBS'fire testmg and to an IEEE stanchrd being'eveloped which'gav'ertts ERFBS ampacity tosting. Tho TVA'lnpacitj'estsrals'o <<stablishcd methcIdologics for use withiriceway eitifigurations which fal14utsidc the boundi of those'ccivered by tho draft IEEE standard. Results of ampacity tests performed by Texas Utilities Electric (TU) were also compared to an existing mathcImatical model for cable trays which havo an ERFBS installed. Modification to kcy parameters of that model ate While suggested which result in; good correlation with the TU tests.
""-'-""-- q
~ ~
I L~ODUCTION - '= - ." '- - NPP werc'pcriting or in'fmal'stages'of construction:
i redundant shutdown cucuits were a part of their Thoro Lag 330-1 Electrical'aceway Flii Sarrleri asfc'dciign, ieduhdant FSSD circuiti'were often not SysteiIss (ERFBS) hive 5ecii used in the'frtajoriiy'of . separated by rated fire baniers. Thc most cost. ffectivo United States NGclhr'RhkYPlanls'ks iyaislvo ftro' 'bickfit'to meet the now rettuiremcnti was to protect bamer to protect. redundant electrical cables and . ono of the installed redundant FSSD circuits. Areas in equipment rHgUgd
' fhfe ShutdoiIjn'(FMpuf" -; the NPP that have," iiis6died automatic f&detcxtio'n:an'd the rijc6x. (d'"' .~ tho reqIliremedH'df th'"" '"'suppiession rsyste'tns refy'.ori 'odikloitr'ra'tcd'RFBS' App&dbms 'Rr; tB'd'h
,f480. ',!
separinng"rediid8ait's'i cl" 1ated ERF5S'."'ha'dCutfonfeme'cting thyfiro endiirancol FSSO cablhr,a54't'quipitienL'mu'st. hiiaa'three Sik'" "- rc'qiiirements', the'EIPBS'6iuldnoc id'vcisilly'impiict-'aieas'without rating ib &Au'tomatic'fire'.de jTe&i;""!-'-ttie'impacitj'dctat'iilgoAhec'abf5or'pr'esenr an impacti and"s pprcss fon-'or'a Wi"Sour'.iiimjfoi'are'ai'%noh '-'r'-'-"hiza8 "4 '"sif~fe~"-'cqttiptneat duritig an insiallcd "sutomatic"'fire'-'detbdtorr"and si" ppfl$5 toaP "1"'-"- -earth quak '."-".'-"'-'1 -~" .'-"8 whciiTAppendbt R.was prontulgaled,'tho titajority~F'-":
' '\> ~~ ~ ~ ~ w ~ ~, ' ~ l ' a4 ~ ~ ~, w ~
0 45 0
vI ~ ~ ~ IJ ~ '4 1V OO' 1 1 5 I&M outa. Engr. 8L. svcs. 813052488231iu 3
~ ~
\
Caicu1ation PTN-BFJH-96-028.
Revision p Attachment 1 Page 2 of 23 Thermo-Lag 330-$ is a" proprietaiy'aterial barrier This Bullctm was then followed by Generic manufactured by Thermal Science Incorporated (TSI) Lager 92<8'n December 1992. With all tho conccrtt of St. Louis MIssopi Thc material is best Involving.Thermo-Lag 330-1 ERFBS and indcpcndcnt ch rizcd as a "sub ing ablator.". When exposed testing being performed by utilitics and the NRC the fo Icvatcd ih c materiaLtransposes following question evolved: "What is the correct fire temperatures from'h a 5 lid into a vapor vithi results in heat blockage testing criteria 2:".
(ab tive shielding/cooling ). The material originated in the space industry d ,was primarilyiusrd as a'heat " At the time. this issue came to light the TVA was in the shi Id for vehicle r~ into'ihc aunosphcre'. -
final, stages. of.completing construction and. licensing WBN Unit. l located near Spring CityTennessee..The CO CERNS ...original.. design .of,,WBN. utilized... an . ERFBS
.. manufactured,,by the3M Company. After comparing .
Fi Protcctioa -, .;. the ampacity deratfng factors. of both 3M and Theimo-
.Lag;330-'I;, the decision.was made ta switch from the Begprupg m Augustaf I99I thc NRC alerted:.utilities .: 3M.ERFBS to Thermo-Lag.330-I with its lower to tcntial problemsassociated with Thermo-Lag 330- ampacity derating factors. Since WBN relics on I E S'. Based upon fire testing performed by Gulf Theimo-Lag 330-'1 ERFBS, TVA then volunteered to S s Utilities'GSU) kiver Bend Station in October: assist the Nuclear Utilities Management And Resource l9 8, deQciencies were identified in their Thermo-Lag Council (NUMARC) .with engineering and craft 33 I installation. ln the GSU testing,.30-inch wide personnel in performing vendor sponsored'Ihenno-Lag alu inurn cable trays detected with 3-hour Thermo- 330-I fire testing that could be used at WBN as well as Lag 30-I ERFBS exceeded pass/fail tcnipcratures and other NPPs. (Note; The Nuclear Energy Institute (NEI)
Iost ircuit integrity witfiinapproximately 60 minuter. is the successor organization to NUh4QLC). Afler phic: failure 'atId collapse of .the cable. tray 'umerous'" meetings:. and discussions; with the C'S within 90.minutes.'he purpose of the GSU :NUMARCstaff and other utilitics 'represented oa the.
was to evaluate tlteir as-itistaHed.configurations.: NUMARC advtsoty.comnuttee, it was determined that lmEdetermincd~g maintenanceactiVitiet dao aconscnsttson:thc'appropriate fire testingmethodology the bcontractot'who inktalled the Thertti'~330-I: an4 ttcceptance':criteria:could not: be reached'ia the rcmov'ed:the-"sfsldn"'from thi:: firer barrier: sho'rt'term.iequiied.to support WBN licensing. TVA mat a}..'Str'ess skin is a:steel wiie'mesh iYistalledhy. ~
would have to 'proceed on its, own in qualifying dor.@apart:ofihe'ication of -: Thenao-Gag:330'. is'a rabbet ERFBR . '
~;
Se%hermo'-Lig'nc'-hour 33+i'. Thc'rIrio-Lhg'.330-'P l1as'stliss skin 33~l'hree-hour has on the inside itress skin Inpalied .on both Nermo-k.ag the-inside and H outs de surfaces GSU en repeated: the, fire.test by Thd'ecLsi65 toi~llrotect.tfLc:4kles,at.%9M'with aK m biles.in~rdancewith - ERFBS'occurred:afler Ihe'Cab%i w'eie installed'ut their cting MHtionai the endori published'al Iatiott;:manuaL:':Siinilar raceways.'v'tiiuatioii of tlio e~of thc ERFBS on'he fall" occurred'ltcrNKC Icon"tinftcd initalled caSfcsrw'as" initially'p'eifortnet in themid
'ntl
~
19804 utiliziiigaiiipacity cbttectfott actors based on to7oifow"up-on'Diforrnaiie7Ndtkcd this eject issttittg thi iesults of a+ducted by oi'forHs mattufhcturer; T Utllidc'iEsca@ 'in'su'ppoit ot'the"Iicen'sing TSf I'n thi earIJJ 1990s Notices, Bulletins and Generic of anch'e 2";contlucted-'ad4itio'n'al fHe. Leneri fro'm the HRC atuf position'papers'rom various cn -tcst5rg of 'hour "-FScrinWlt-:330-'P industty~upficKntified:potential probletns with the ERF : Based upoii lcd'Mures 'Of stnaIVcd'nduit and originaV'ampacity'ert'rith- the'mbst 'significant I 'cable tray's dtukgf 'VLF'testing, tfiiPNRC Bsued coaccrni fn the tuea of piotected'trays' comparison Bull 92-Ol'~dsup cmen t 'l" ilrJi&lo and August of tho'cotrocdotr facrora o'rigirutlly utilized is TVAanti l This Bulletiid~dminctf Qiit Therltio-Rig-330-" thermo:which were snbscquendy published is yven in I;.as instttlled o~Ot condu'Ihsmaller tKm 4-'inch and ie trays:large'r d~k'l~ wide;-wistma5Ie to:
rated'fbi'd Its destgitMsis:t~tijjrequ Ra ta.>> i m "ont irz:tisane ~ "
lg ue
~ p \ ~
( "$ ( tl ~
. ". ~ .!i"""c.':~'~~ia< '"!'>"ws se:W' ~">:"
0 0
0
'V I I ~ ~ Ll I ~ 0 ~ M ov ~ I ~ 'JAg~ ~ uuxo engr. o ~vcs. >> 81305246623.1:o 4
~ I
~ t
-' Calculation PTN-BFJM-96-02B
'Revisiori 0-Attachment 1
.Table t
~
Page 3 of 23 Comparisob orAapacisy Corrccsios Fiorors'(hCF) derariag, a layer was included in rhc TU ampacity tests.
The TU"correctioa" factor;(Table 2) was therefore conservative: for uscaat.ZVA'asWBN Calfiounaks OI1 TVA ACP~ OAer SowesACt plant"where no siliqa.biankoawoufd bp isrilizecLl oi TSsyosaOVC, '", -0.475 ~ ~: ~
V S OTIC"'
5
. TU (fata Hoar,.Tbcrspo-Lag 330-t Tray. ~
l4oie'l. ~
ACF Wrappafhmpi'PSasclino imps"
.Notaz. Baiiut on TSi sOoososcd sassing.'.. " n ..'.C~asa!SO4)-' . Trsy aosO4 as)-: -. ~ca i A review of the early tests reveals that there was '2.45 4 i ~ o 445 between the various protocols us'cd for they iiule'onsistency amp/rcity assessment., This is not surprising:given thc Quubljl.~:~The, TVA and TU condui< configuratioas absentee at that time'of an industry consensus method. differed,, Both"used the same-.&erma.-.Lag 330-1 for the perfonnance'of such tests. Diffcrenccs existed preformed sections. however. TVA procedures dictated in tlIe length of the test specimens;.tlie control'of end that the interior of the barrier have a complete layer of heatmg effects"the method'of ambient temperature trowclable grade Thermo-Lag 330-.l applied prior to control, the size and nmqber of cables used; the size of installation., TU used a mare conventional method, raceway, the location! of thcrmocoupies and the employing only prebuttered joints. Because of the defujition of equilibrium conditions. Variances in'such resulting differenc in Qt, TVAdetermined to pcrfortn kcy Iparameters inevitably resulted in its own.ampacity, tests on canduits (using Draft 12) in the derived correction factors.o:
significant.'iffcreaccs with the same configurations as those which were i
evaluated during.,thri.fire endurance. testing as follows:
ln jrcparing;;for- theo,suhj act, test .,prograar..:TVA-prougufs" which'"h&e ~-
pcrfcIrtned a parameter by paraineter review of-'th' been issued'yhce=th
~ '.,, 'ia,sioaioktksch bye.,
38Rnc5 base covered Qiils i 3/S.hctf origipattcstiiqp"T1re &a;:Uh 17M'.was-an issietl 1984l ychile.the second, IEEE P-848'as still:in draft?
4 c.~~.i .o<< '~iA (zyNach',bag;covered wtrh a 3/Bach C'
form[ Based.oa a 'camparislt of I As perttutrecfbyhft12, separiita coaduits were used the'key"paraectcrs,'Eicti is"dexsscrribed'in an coariier'paper~'TVA'chSh 'to utilize the draft IEEE dopuncat,";:~-; ....'.:...-',< <~;=.'-, - ~
for each. of, thise assembltes.. Tbo. utilization of differen-conduit:: segments.for cadi con'figs'uration At the same time that TVA was evaluating the varioui aHowed.thetesbprograin~ymcccd'abc much;faster methodologies, TU w'af'prepatmg to pe'rfoim-'aaIpac~ pact, thanj+they.bye,beyn. condgctpt.sertucntially on tests" .TU had also deterniined'that""title-'F448:draft"" ihc same segment, due to the 30 day cure time (for the the best'available'guidahce tmwelablc grade matemd), between eachitesr. Given:.
provided Based its progr'am'ott Draft'1'I ---"" i""'
fdr=the:testssm5'ad that-.the thermal'esistrvity. of/water is fairly, low, its elimination..&m. theiseuns and= interior.-.prior to.
XZ~;- Draft 11 reqttk5f Sti'use'hf a 4=incH'x'24'-in'eh" ampacity'testing'is required'in 'order. to:obtain results.
tray filled w'8K 1205A%9tWG'600'ihNco~ cabh5 which accaratelydepict the effects of theERPBS...
arranged in four 3~~TU baiiier'designs utilized a 5/8-iitdr layer'otdS%9ctmM+330-'1 jiirieli'withiF - ln the early stages of the i"~:A~g singlcI layer'of MiMesPMbhtiikct placed'over th' program development it had beea assama'hat the configuratians:autlinedln.PA4$ meuM;envelope those top of tha fo'-m; th~slmet wa's aged to priv'id'idditiana1protectioa'Yor'cables . which wouidibeiencarmtcredaduria.construction at in high/ filled trays from the heat transmitted through the WBN): Asactual,hisailbahaef the ERFMl.neared, lt ba'meta'ff tlie firis- Sin'ce4e.tNerttfhl.asistiyity became cleat that. many unique, enclosures would be of tho sill@a oiotht" is about Advice ttuttwf: t5'e itic!IARd'coulci'. tli'ctii7oRMuit"-iti'%YmSMdi%M".
cabi~ rcquirecL.'These.unique eaciosttres Jlt hto,two broad categories. Fust~:thc qiidug4&odcfuitspacing; was.. not, suQicicat;".to . gcgp'k,instalhtsait, of, the 3
hg
0
~O
t ~ ~t ~ ~et ~ 4 I 4Vg jjLOOLd 1 t
Calculation PTN-BFJM-96-028 Revision.0
.Attachment 1.
Page 4.of. 2$
~ '
sections'he only viable alternate was the 'reformed ouidancc itt Generic Lett'cr 86-10<<, ASTM E 119>>
enclosure of mangle conduits within-a canimon alid the American Nuctcar tnsurcrs guidance provided ERFBS. Second, where cabtes.in trays extended. above in "ANI/MAERP RA Standard Fire Endurance Test sideraits such that enclosure of individual trays was not MetIiod'to'Qualify a'Protective Envelope for Class lE 1.
pospible it.was determined,that multiple trays within Electrical Circuits""., Undctwri teis Laboratories (UL) the lame vemcal'stack would have to.be wrapped in a also'had a methodology. avaIlable foriteIting ERFBS.
conImon ERFBS, .Therefore, given'he:myriad"of 'IYieu cfocumcnt, UL Subject f724'r, vias reviewed and combinations which can exist, onemajor component of ,detcnnined to be the best starting point for developing TVA'.s test program became: tlic'. identifications of tti'a methodology.'."TVA cfeqidcd that testing would bc boupding test configucations for.:such<<non-standard 'erfo'ttned to the S tandard Lime/Temperature curve as arrangements ancl the development df some general iisted'in,most fire testing standards (i. e. ASTM E119).
rules, for,extrapolation of the fmdings .
For'he TV4 Pfiase I testIng the decision.was made to TVk APPROACH XO.RESOLUVfOY.: tn -". test'the.racev/ass without cables...in their place woukl be a single bare /t8 AWQ stranded copper conductor.
This provides two.benefits: fiat, thc fire test would be t conservative since'there is no thcntial mass ofcables to As IIrcviously stated Thermo-Lag 330-l is used at all act as a heat sink during thc thamal exposure, and three TVA NPP..Since thc problem was generic to second, the testing would bc mde pendent of cable types more than onc TVA 'NPP,, the "ownership" of the and fills. The bare conductor is instrtunented at 6 or pro/em resided with Corporate Engineering (CE) l2-inch intervals and acts as a surrogate cable,. It' Chief Engineer. The 6 rst step in addressing the issue thennocouples provide a profile of the temperature was to assembk a self-managing taun of engineers to environment, inside the raceway during the. fire test.
ownl this issue. In order. to be successful Ae team: The exterior of. the raceway. is also insuumented.with woufd have to work at a'fast pace,'quickly overcoming thcnnocouples at expected high:temperature locations'nd any-mistakes or failures-'s'. At:TVA;thc respottsibility- 6 'oi. I2-Inch s'pactngntherciftcr,= Because the for Fire Protection Engineering/Appendix R is assign'cd: ERFBS'encloses the entire riccway, there'is no easily to tbsp Mechanical/HucIcar. Enginecririg. dcfmable cold side of thc-btftricr as thercr is with fire wall tests. Thcrefote by definition, thc exterior. oftho Department'M/NE):
Because of. the poteatuti:impact'onccabling.
and pn raceway seismic'integrity, theiEIcctr'ibad-'and raceway is the cold side of the ERFBS:"Thc test data is.
CiviltEtiginming Departments;iri a'dditlon to'Nucleai thea i'ntapreted ae follows - -- '. '...::".
LicetIslng, would 'hy: -key roke in isucces'sfully: ~ ~ ~ t resolfing4m Issue.'h@ is "appa'rent'sinco Thcim&- l)"Sinco: the'.'ERFBS'tis ittstallcd-'for. physicaL La'g )30 1 w'oald have tp.meet tmuitiple perfdimancc. ~
"scparauott-'and'fi?actions~-a,fire w~ if the criteria to bc successfuL 'TTiaefote; lntorder to address average tempaature rise is ass than 25'12VC) and rcso 've all the tss ues.assoc'atal i vnth Th erm o- Lag
'" =-aiid the.highest single'thartnocouph risc is less';
330-'f, 'thR spcc1aliiccL'self trfatiigiiig'fBmiappt'oach tfian 325; V:(l63.'C) me'iisu'rcd:oii the 'exterio was scca as thc most ajpropriatc choicii; %ith thi ASTM EH 9:Crt tertts); ifshaH bc ofhc'raceway,'(ri.
NIfCtasuaiicc of "RiqiMfor Additionil Infoimatiot'i . 'consider'ef rited" for iIItypcjs'aad fills'of cables.')
Rcgafding Gcncria: LcCjr. 92<5, Issuai Pursuant to 10CPp0.'54(f) &5ekfa6er 22; 1994". aa-eifginecr. -
Iftile'ERFBS'excccch these tanpjetature liraits; ts
. - time/temperature proSe cats be developed and thc.
to theC teatts! " '~"
from thc CE IvtaQrh5c&.5ipcctionMpaitmcnt wtN specific 'cabl6'types then:qualified in'accordance with ttie requiiemcnt5 of UL'124:Appendix B
"(with'sctmc.enhancement/ .: -.:
jnn'tjnttnnjninjj tjtenp'ptttpt jete:ftiet:tint'lttitjtit'tjojnjp- Fire:bamer tcsttng~icallp Iticludes:anpcxposutts to a and acceptance criteri fbr ERFBS'thI. et'arting potntp'asp hose"stream 'to modcI itn pact; aoeiots'-and. cooling.
tci perform a lit'eraturts review-aF tHc=<<vailibS. effects. TVA;detaiminccL that a rcaHstic hose stteatnt stan '.. t7VA EngTti'cering'rc'yiewecf the:NRC test eh'ould bc'pcrfonned atthecnd'of the'ftreexposute; n
I I
t " ln
~ et
~
e
~ r
0
~
~
ae:hi aY 6 10-66 t.i.i ."
BA4M Gute Eager. 4.6~cs....,. 8,a306246623i 'c 6
~ w ~ -'=" "'CaTcuTat'tcIh&TN-"BFM<6-t}28-Revision 0 Attachment 1.
Page 5 of 23 on the experience of the March 2~, 1975, BFN two craAsmen to install thc ERFBS), thc assemblics toe, the electrical raceways did not receive any severe must cure for a minimum 30 days. This curing time is impact force fitxtr'Mlingobjects induced by the fire. required'or the .trowel grade Thermo-Lag 330 1 This can be attiibuteif ro thc solid ctcsigned'suppon material to,dry. The fiat decks consisted of two each:
systems used for eitithquake requirements. The one- 1-inch conduits, 5-irich conduits. and 2-inch au drops.
hap"ERFBS will bc installed. in sprinkler protected Varia'tians in the ERFBS Cosign would be ippfied to areas of the plant to meet Appendix R requirements. fn each ser. Thc'goal 'was to determine the effects of an )tctual NPP fire the ERFBS will bc'exposed'to the conduit siae.on identical ERFBS. The first test was run mechanical impact, erosion and cooling effects'of water on December 21, 1992: Thii test had a baseline, spray fram sprinkler'system opcrauon and fire'brigade nominal 5/8'-inch Tliermo-Lag 330-1 ERFBS installed hose streams. Therefore, it was determined that thc 'ased on each'set ofcomponents. Additia'nally, the second sct most appropriate hose stream test that would represent of components'(I-inch and 5-inch concfuits and 2-inch theimechanical impact, erosion and coaling effects in. air drop) had an external'layer of stress skin applied a T~VA NPP environment would bc the one and.one- over the completed. Thermo-Lag 330-1 ERFBS. The half inch fog nazile testas describe'd in NRCs NUREG purposo af t&is.variatioa'was to ancmpt to achieve 0800' ~ ~
'etter. performance, as suggested by, thc smaII scale test rc'suits as reported in Information Notice 92-55'. This Aft/r the completion af the TVA phase I testing and additional stress skin would. also act as an cxtcrnal prior to the start of Phaso li, the NRC issued skeleton should any structural problems develop durmg Supplement I to Generic Letter 86-10'~. In this the test. The test demottstratcd that a single layer of Su+lemon! thc NRC defined the appropriate nominal 5/8-inch Thermo-Lag 330-1 was adequate to methodology and acceptance criteria for ERFBS. protect the 5-inch conduit but insufficient to ptatecr the tcstilig. Comparison of Supplement 1 to TVA's testing l-inch. conduit or tho 2-inch air drop for the required pas4iaa demonstratef:a great deaf of similanty,-the period of tinlo'(i.oone hour) majIIr diffcrc'nce being the nomimtf'spacing, of.
theqno'couples. TVA's Phase 4 testing'tttilizetf UL'. An'intc'rcsifng prebfetn was dBcoverek'during some 1724 guidance of key'locationi on the raceway'.ind'I2='nchI earltcr- 'tcstlllg "Lnvol~"..ianotnalies'weal: thc spacing thereafter. Supplcmeiif %requires 8-inc}i* thcimocou'plo. r'eadingi. -TU'-*ia cxifcricficing erratic spacfng. Based Qpon the uiuform temperatuie prbfiies. temperatuio dibs in sornb oAMir'tests. Aftcrsotne dev)hped in the Phascl I',-'tc'sts'lh'h differenc'~as: researcher by'c- li boratory'On@a" Point) it "was:
detcimined to be insignificant'".- IIIevcrdiekis,'5%ii'o fF dctetmutc'tf-'r fhtf..'twitch. thc i ThctTttyiLag ancf )R testing would'utilizti'th'o 6-'inch thenaaca'uplo'V undergo'tisubliming ablaifvo shielding process the tniiertal saturates the fiberglass insiihited~acoupio lea& causing theth to:short. 3 oftifngthc testing, TYA FERE. TEST RESUKTS. -:"-" -: ha'd expeiienced. tfti.'same an6ehUes: with fiberglass
-lrisulafcd thmn~h iastalJctf Me+er eoastrtlctcd:
PhsuIe I Fire Testa"' ':" ~:Aaatcsuif:of these'probletni+yiiigg4+
a Tea Chap@ .
~
The fire testing waa dfv+xl into three phases,.Pttaee I basic elcctsfcak raceways such as; conduits, jungian boxes anil~ profcctedwith OOH~
'rcabhv.trays'nd'ultiple:
A1iteik TVA:continuetf'constructing: Phaso I test decks and performin'g~tesfs4iniugh:tho spring-'o$ 1993 For the ERFBS;.
i candItits in sing& eniIQuro proWted witft a~M first "upgrade ~'a nomfnaf 3/8-inch preformed ERFQS;" -Pftaeai.IN tcstttd cotufuits and caMo. fnya: section-wae ahtef aves the nominal 5/8-inch first prouictccf with "tfae&NQS ERPHK TVA bcgais laycrx'.i An interesting discovery.wis made during the.
~etio n o'fhc first three Pha& I:test decks in:fifer. first:5/8-incft+ 3lMxJi tese~ccrning tho two layer.
t fait. f-1992.c "Ttr'erforming fcstfttg't mttit., 'Ofcttno-Kag330-'I ERFBS perfortnance (see Figure 1).
lead'fimotS'bit%KenM'Gtg'330-1'hifsra:hhg Ther two- layer ERFBS CHtnbcd stcadtly up to with each'test hdtv tbo'nstalbttlaa-waa approximately.2~ (40PC} (asnbfcnr tcmpcntturc,~
co leto (appioxitnatep ono w'eek"pei'test'decTtt for tempera&to sfre). Ac that point the curve Qattctiof whh
~' U
41 a ~ N~Qc40023i 4 ~ 7
~ " '
~P
~ . lc.
' i""='"""(ad julftao~Tt7-'fggM-Obr028'Revision 0
~ - - Attachment'1 page 6 of 23" only a small rise for the last 1/3 of the test. Upon Tabic 3 disassembly it wa! also noted the outside layer of 'rhctlao Lac 330.1 Coaltllit S~ acaiaaa material was cocnpiacaly consusncd and l/2-inch to 3/8>>
inch of the insidle layer was un-reacted. Based upon this tost, subsequent tests wcte performed and qualified a Wo.l ~~
using a 3/8-inch ~ 3/8-inch ERFBS. design. lt is 5' 55' )/p Vl +l4 theorized that the interface resistance bctwecn the two layers coupled: with the outside hyer" charring and X i stress skin maxiinizes the efficiency o'f the subliming '.!'
x ablative cooling on the protected electrical raceway.
The graphs shown in Figure 1 were dupficatid for all i x two;layer Thermo-Lag 330-'1 ERFBS installed on '5, x i '
varying sizes of conduitL Thc only difference in the curves is at what time into the test the sample reaches 5 apprOximately 212'F (10'}, Due'to their smaller X' X X therntal mass (inertia+ smaller cond'uits. (and air drops}
4 : )c i'-;1-' 5 reaclI this point faster th3n larger,'conduits.,Table 3 shows the TVA Thermo-Lag 330-1 ERFBS. conduit ' !':
designs qualified in the test prognun.
r Nba i. Retina caMe qualNcadoa.
5 Nar Cease ~
Ffipua I r Nt CeHas luactfoti box testing demonstrated a single layer of nominal 5/8-'inch Ihertao-Lag 330-1 with an exteriia}
hyerofstainlc'ss steel stress shnind trowel grade skim cow was adequiie to protect sizci rariy'tig from 6-ia*
g5 0 ~
x 6-inchx 6-inch through 24-inch-x-.t8 Inchz; }2<<inctL..
i
'-i~ '
Phase'I also. determined junction 'boxes" with a diiri'ension gteatcr=than 24-.inch.up.through.48-hch x
~
c555555I 5,,0 X 1 ls5555" r.flIloI I 36 inch x 12-inch can be success5dly ptotccted with a 5 ~, I'il nominal:.5/8-inch-"+'3/8-inch:-- Thettao-Rag "330 1 I ~
5
~
~
5 ' ~ 1 ~i lI5e5'hasci Afterthe successful compietioa of Phase I testing, TVA.
catered into an agreemeat with the, manufacturer, TSI,.
for the Phase I15and IIItesting, In this atrangemcnt TSl would provide ail accessary.'Thermo-Lag materials and I also performed tests to de'tertaine ifthere was laboratory fees-whili TYA.. was.-responsible for any difFerence in thcrtaal erfartaance.'bctwcctk;: '~..th ~gm ~g,a m~ndiaion~
p aluminum aad steet. cotiduiti protected.by.ifctitical ~.
ERFHS. Steet is apprtaiaatidy three times heavier equal )ized contQlita 55gehttninutnhas approxiniatsly;u,:
~ thc necessary electrical raceways. 'the results of thii testing would bir made avaihible to any interested NPP'"
through NEL,. This.'5aizangetnyg would provide a ce the;
> hest:carr exposed to the same ~ t:~g';firtt5;1css on 3 utch conduits with the saaSa 'Ihamo-Lag 330-1. ERFBS heat fhx demonstrated thc-aluminum conduita wiR:experience slightly higher .
temperatures (Reference ~A Test:6."1 A).
"Win/%iii"philosophy for thi prolect and all patties involvecP'.-. In addldoo to the-fhwtests, corresponding ampacity dcratiag would'e pcr&rmcd on the configutationL5 The shy'f sevctr"orie-hottr5firi tests in Phaia 8 aalu-two/threeti'-
the.'testhtg was sec a .:
~ \"
three hour fire tests in Phase K':: -: =
~ a'0 iV 5
~ ~ ~ 'ZAVt"'
%1
'I
~ ~
~ ~ ~
Calculat>on,PTN, BFJH,96 028 Revision 0
~ - 'ttachment 1 Page 7 of 23 Phase II Fire Testa not including the weight of the Therma-Lag 330-1 ERFBS which was constant for each tray) versus its The design for the seven one-hour tests was based on temperature at 60 minutes an'expression for tho effect exiping installations it SFN and SQN in addition: io. the of cables can be developed; expected configurations'at WBN'-- Based upon: ihe r success of the TVA'h'ase: t: testingr: the design
-,philosophy was ta keep the "upgrades"'minimal aad geqeric as possible. The tests would also aaempt to 1 Her t~ ~
Figiire?
CANIS LyylIIIy establish installation r principals through bounding configurations. The "Generic TVA upgrade"- consisted C+
of adding a layer of external stainless. steel stress skin r .r lryyy with a skim coat of Thermo-Lag 330- l trowel grade.
Thq external stainless steel stress skin and trowel grade skim coat would:
I I ~
r W%
thl Oah ILKSeO I
- ~ Provide a thermal boundary layer during thc ~
fire exposure. This will maxuniae the ~ O yr S O S e ablative shielding provided to thc raceway I by the nominal 5/8-inch Thermo-Lag 330-1, i~ Ensures the 'RFBS rcntauts sttucturaHy The followiag equation is based on a "best fit curve intact during.thc dynamic process of the approach with a logarithmi relationship (i.er result of material subliming during the fire exposure liner regrcssioa, method of least squitres).
and subsequent hose stream test, and e l - Provide the necessary strength aad sttucturai Final Temp.'87&'- 86.752 Log'(Weight)
~
integrity needed to meet seismic requireinents.'VA verified.. this'design featureh thesummei of'19944y canthcting ~
Final. Tenqa ': Degrees Fahrenheit"
-'".. "'- fall sade. shake ta8'at Wyie Liabo'ratorics,' Weight -'ounds/Foot of cable tray
"" Huntsville, Alabaini. aaC cables r*
This equation is'valid far eighteen-inch cable trays were.dedicated ta cable triy configuratiotts." TVA'-'Test protected with the TVA dciigned Thermo-Lag 330-1 6: l:0 consisted'of thr'ee N-inch wide;hdder back, ERFBS having-cable.fiHi ranging from 624 Lbs/ft up with ldcatfcihupgraded ERFBS and Varying.
s&l'ab'rays thru 69.36 Lbs/k.
I cabl) fiiL The left tray uz the.test deck'represented'rf ~
mm u. f-if~~(i.c.'2894CW6.AWOyn6.Lbe- Further review waa. performed'on the. results, of the cabti/'linear Ft.)). The center tray in the test deck single layer fiHed cable tray (624 Lbs/ft of cable) and represented a single layer filled tray (i.e. 26 4/C 416 the empty cabR'triy (0.0 Lbs/ft of cables). This was AWG (624 Lbe cibkf haatr Fto; TSo rijh'I'tray in the determined necessary since the cf5xts of adding cables test deck repreaseihk'i4 empcy tray'(Le. ni'cables): over thc first layer becotnes less importaat due to the The results are'shavikltm Figuri-2 " " cable insulation skrwhg the hear transfer to the capper conductors. Conservatively, a plot was constructed of The only theitnot~IIlea to 'exceetf -tSe the temperanues for tbc empty cable tray (0,0 Lbs/ft of weri the Me h iniide tfte:empty tray. This: cable) and the single'laya cable tray (624 Lbs/ft of acceptance'riteria occurred 56"mm~ into" the-test. -The caMo). The:resultiag linear'equatioa givctt below
'at th'e ~;of'thb tist was 83'6 TMs ambfcnt'ern coaservanveiy predicts the system's thermal response di a mIiiXitniim'avetago'fempiratiire'of 33$ 'F at at law cable'Qih (he. les! than 624'f4'/ft of cable).
I ~
60 mIautes (ambient tcmyeriiture plus 250'F allowabler tcmpcntture~J Bypkltthtg the weight of+M1Jcabliet Find Z'emp ~ 385;10'- 97756 ~ {Weight) tray system-'(f.e the weig&of the'tray and cables attiL' r ~ ~ ~ I r ~ ~ ~ I ~,
I '
I ., ~
~
a '
I
~ ~
0 0
I'I s Cr Cs 3 v C S. J OhOOoo S
~ 'e Ca 1 cul a ti on PTN-BF JM-96;028 Revision 0 Attachment 1 Page 8 of 23 where: the asscnl biles. Tlris test was perfortned
'Final Temy... Degrees Fahrenheit'" horizontal position.
Weight ~" - Pounds/foot, of cable tray and cable . TaMe 4 Two Sieicat Eadocarca Sriadarat Coadoit Coafiaaradoaa Solving this linear equation in the range of acceptable temperatures indicates that a cable tray system.with a weight of 5.33 Lbs/Ft would maintain acceptable Coehcsreeess' hvmye 'fasyaaare Oaesseeea rem at 60 minutes. Subtracting the weight of I& Tmeerasen aies FFI the Ie tray (4.00 Lbs) from ihk system yields a cable Seaeai ass eeeea Tss I $ eeea eeeceeee l ing of 133 Lbs/Ft.. Based on the cables used in the test(4/C 816 AWG>> 024 Lbs/Ft) a minimum 6 cables seeaa aces ushers Fraae are Iteeded to produce acceptable temperatures (i.e. b, T
<2 0;F). - Lares seas ssCa Seas s'L Coeaaoeo Test 6.1.7 also contained a 3-inch conduit with Laras aos I/siesso onethalf protcctecLby Thermo-Lag 330-1 and the F~
remaning half protected by 3M M20-A. The different fire farrier materials were butted together and a layer TVA Test 6.1.11 demonstrated thc performance of of3IVt M20-A was wrapped around the joint. The test "three sided enclosures". This type enclosure is demonstrated compatibility between the materials with typically used where conduits are located agrunst a wall no failureL The thermal performance during or ceiling. The conduits are enclosed oa one side by tho t t was representative of each'unique materiaL concrete and three sides by Thermo-Lag 330-1 aaached to thc conduits, Tiblo 5 shows thc average aad TV Tests 6. LS and'6;19 demonstrated'that bilge cable maxhnum temperatures of thc assemblies. This test tray Ifittings (doublmross), vertical stack of tHW was performed in the vertical position. This test also
. contained a large junctioabosr(60-inch x 36-inch x 24-1g-'nch) trays, and two side by siCk 18.inch trays can succCssfuUy be ptotkcteckin a single Therm~330-l inch) protected vtith the Gihcric TVA Upgrade. Me encgure upgraded with thc TVAsystem.-These'tests:. success of this junction box itesiga eliminated thc need wereI all performed with nir instaHed cables; TVA Te'st- for-a'secondlayer of nominal 3/S-iadr.:thermo-Lag 6.1.$ 'also demonstrated that'a cabli tray with a raised. 330-1 as previously deterauned la TVA Test 6.1.5; cover can bc protected with Thermo'-Lag MO-1:."Thar test Simulated a "randota Glled'cabfe'Tray (ieb, Table 5 Tiraa SiatcaS EacSocarca are all located in the'center of the:triy the'ablg
'V with'the'eist ofthc bundle above the side raigr Onk-inch drop .Cabla .Sandie's owere also Cos/iocsase hsseaos Tmyease a'nd'fvk-tach'ir
~
dern asrrated acceptable in this test, -"
alas rFS
~I stoa " "'.IJ The gext four tire~ of Phase tt'were dedicated T .I Sax~ :-" I)a-. "- t --"o-i1a "
condug'~guraaoas," Each 'assembly'ons~
ta-'ultgle "Tine 1'veksees So,r a ~
I Iis' P
of aomhsiL544fch'Tlienitd-'Lragj330-1'idi'he Coooic i Peneric TVA 0nBe'.'-VVX,. Vest - ~,
the'erfankiiii'""of'twi 6:l.l0'ern
,3?a area".'hhr \ rr encl type-;ciidosuie.is typrceally"usxf. ~
w 'conduiti are kcatcrlM a c'orner,'ic'alorig"w'hats TVA Test Kl".Ik- dhaoastraRd'the plarformance of:
the c)iling and wall meet. Thc coaduits ark <<nclescd "four sideYenclosurcs>> - This-type-.enclosure ls.
on twp sides by coricrato irtd two sideiby'7heimo-Laj: typically used wltese coaduits are free:standmg..Thk 330- panels attachert to"a lrnistrut rtrame.'iMe conduits 'ar6 enciosedea all four, sides by Thermo-Lag.
330-1 aaached'o the. cottduita; The-. CoaBguratioas-4"'he of' average aad alaximum'tern'peraatrc'ase
, ~,n' ~ vr ', ~
~ ~ ~ e v ~
0 0
~ ~
~
~ ~
~ ~
~ P I, . '
~ 'I
~
~
. ~
. ~ ~ ~ I
~ ~ ~ ~ I tI ~ ~
~ ~ ~ I a ~ ~ ~ ~
I ~ ~ ~
~ I~ ~ ~ ~ L ~ I ~ ~ ~ ~ I ~ ~
~ ~
~ ~ '4 ~
~ I ~
~ I~ ~ ~
~ ~ ~ ~ I ~ I
~ al ~ ~ ~ ~ ~ ~ ~
I ~
~ ~ t ~ I ~ '
I ~ ~ ~ ~ I I ~
~ ~ ~ ~ '~
. ~ I ~ ~ ~ I .~~ ~ P ~ PL ~
I ~ ~
I ~ ~ . ~ ~ ~ I ~ ~ I ~ I ~
~ PI ~ ~ I ~ ~ ~ ~ P ~ ~ ~ ~ I ~ ~ I I ~ ~ I
~ L ~ ~ 4 I ~ ~ ~ ~ ~ 'I I ~ II~ ~ ~ ~
~ ~ I ~ ~
~ ~ ~ I ~ ~ ~ ~ ~
~ I ~ ~ ~
~ ~ ~ ~ ~ ~ ~ ~ 'I ~ '
I' ~ I I~ ~ ~ I ~
~ ~ '
I ~ ~ ~ ~ I I I II ~ I ~I I '.L ~ ~ I~ ~ ~ ~ ~
~ ~
I ~ ~
~ I~ ' P~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ I. ~ 'I ~ ~ ~ I. ' II' ~ ~
~ I~~ I~ ~ P ~ ~ ~ ~ ~ I ' P ~ I ~ ~ I
~ I ~ ~ ~ I I~ ~ I~ ~ I ~ ~ I ~ ~ ~ ~ ~ ~ ~ II
~~ ~ ~ ~ ~ . I ~ ~ I ~
~ I ~ I~ ~ I II ' ~ ~ ~ ~ ~ p ~ ~ ~ I ~ ~ '
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ I P ~ ~ ~
~ I ~ ~ ~ ~ ~ I ~ I '
~ I ~ ~ ' I
~ ~ II~ I II ~ I ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ II ' ' ~ ~ I II~ I ~
'I ~ I~ I ~ ~ ~ > ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ I ~
~ ~ ~ ~ I~ ~' ~ ~ ' I ~ ~ ~ ~ ~
~ I I ~ ~ ~ I ~ ~ I I ~ I ~ I
~ ~ ~ ~ ~ ~ ~ ~ I ~
~ ~ ~ I ~
~ ~ ~ ~ ~.~ ~ ~ ~ ~ I
0 0
0
a0 ~ ~ J <<1e av 4v 'S. 44044 I ~, ~
~
- Calculation PTH-8FJM-96-028 Revision 0,-
Attachment 1 Page l0 of 23 using nominal l-i/4-Inch Ther)no-Lag 330-I. "Worst Prior to Ihe start of the second test it was decided to run case" was detinod m utstalling configurations with die d!e test until the first assembly was close to its lease desirable auributes. This included post-buttering allowabl'e maximi)rn average temperature rise of 250'F and stainless steel banding 12-inches onwenter. Thc (l21'C). 7YA.Test 62.2 was run for 4-hours and 10-second step was defined as "R'einforoemcnt'.- ln %is minutes. The assemblies easily passed the hose stream step, areas of known weakness in thc worst case'design test. Test results are listed in Table lO were reinforced with external stainless steel stress skin and Thermo-Lag 770-1 trowel grade materiaL The third step in the installation process was the "Upgrade". Reseal!4 TAQa 0 orTVA Tcs~ "
fn this step, a ncw material Thettno-Lag 770--1 Mats",
manufactured by TSi, would be used. The The'rmo- h>> 4 Ma<<h>> 4 Mea.
Lag 770-1 mats arc a fiexible, nominal 3/g-inch sheet ta<<aa Ahern') 'acy. Nae rS)
<<)SSS . aa)9) ~
material that is installed over the exisung worst case Thermo-Lag 330-1 installations. The number of layers I) cay I 45 A>> l4I h>>
, IS)aeaev InM ofInat applied to each raceway is given in Table 8. All Iear~
layers werc, pre-buttered with Thermo-Lag 770-1 2I'aZ, )5 h>> IS) h>>
I a) MvaL I al Mea, trowel grade and assembled using nominal 1-inch staples and 16 gage annealed stainless steel tie wire. l)a h>> IS I h>>
Ia) laaa I SS Maa, DiKerent assembly techniques were used on the cable trays i,e. multiple pieces vs. continuous wrapping. The t
final step of the installation was a skim coating of Tcatc 10 Thcjmo-Lag 770-1 trowel grade. Rca!tta or TVATc¹ +2 Tabti S.
4gcra oC,TTLLovce bcocltee l:"c,,,a
~ ~
., Ceal!cava<<aa,, '>> 4 Mae t~A>> a
.. ciaerf)
Maea
~ ~
a T~,. <<110 I Neers)
~
. aa)DX!~
tYAfa<< ' 4Cc)clat)O l1 ~.. v,, lee A>>' IS)A>>
l$4 Meea . 4 -
)OI Mass I2 aae<< Iaea<<4.).4.21I ~, ~
iiaC can rrt , I)S A>> 'It) A>>
', 4.).l ~ ~ .y ~ 1 .'. ~~
421 )'aaeal eea4<> . --."...2)4 ' Aw.
',<<OMa<< ': '.))SM>>
11'a ll a40' 4J.!" ~ . a) ' .
psactaa loa
, 1'wel eaec<>,!I.
14l Maaa
)44h>>
)14 Maac: '
. Il~eeehsk'; 4111,, ~ 'v.q I ~ a vvvvtv
'" '22 ' 1"
)'<<<<SvaeaJaea,,..... I44A>>
~ . I.;I%eh>> ')4) 1'aaael'caaafaia
) aaeel 1'I'w~,
eeaaMI ...421
':. 411
'2 1 ah'iaaf "" '
144 I)4 A>>
Mec,' ')l a
leS lee Mia A>>
v Mac'he composite '?70't'ce& 330-'1'deiIg'ii A third test assembly had been prepared using only the out jerfoase'4'rab~'cq)cctlioe. 'TVA new Theimo-Lag'75 l materials"tri conastruct the Test 52.l was iud 'S~ and'~iiiutts.'t tha1 ERFBS. This test assctnbly served as a Backup" in paine the test was stopp'cd sinco no one ltad prepared to the event thc previous Phase iEtcst asscmblics did not run loiigcr'han thc'equired three-hours. The perform satisfactorily. Based oa the successes of the assemb Hes had a large thermal margin remaining,and. previous hvo.tests,~assctnb~vgs not used.
easily passed the hose streitam ppt.. Test'results are l~ l in Table 9.
I 10 P
~
~ Q Y. v
~ ~ ~
v
45 Ci
v ~ ~ 7Q 4 'H c<oocs I
~
Calculation PTH-BFJM-96-028 Revision 0 Attachmegt 1
~ .
Page ll of. 23 TVA ELECTRICALTEST RESULTS Standard Condula C'onfIguratlons I
,Table I l ~
.TSl IVreppcd Coodeit - TVA DorIng the early TU am pacity tests. which were based Tutf on graft I I of P-848, it was noted that rhe'use'of three
~ ACP I ~ iVo 4'CFR Ale conrIuctors connected in series resulted in excessive inductive coupling bctwcen the cables and the conduit. 1 Ori:I.Oi fit'-.
The resultant conduit hiLiting and surpresscd currents in their early tests led Ttr lo add'a fourth conductor to minpnize the conduit'rIffecrs. These findings, later supIIorted by TVA testing led thc P-848 working Ja sn }s iO) i04 "
~
j r 'ol'tT
'.,I group to issue drift l2, IArhich stipulated the usc of an While the correlation between tests conducted'using everI number of conductors. the conductor and power supply artangcments 'ifferent was good, the differenc between thc two Hav ng witnessed a po ion of the TU work, a variety baseline conduits was greater than is desirable for of were peffoflned TVA's labs in Chattanooga, repeatability and seemed to indicate that some conduit test TN. specific effects were still being seen.
Whig the bulk of the d 'Is and findings of that early Given thc confinnauon provided by the variety of TVA effort werc re in our earlier paper, there configurnions evaluated, TVA concluded that ACFs at wcrcI several key ations in that program which or near unity were indeed possible. TVA bclicves that may[now bc updatecL gn the Qrst progrant, because a significant contributor to this improvement was TVA had assumed thd the clror'ce of test conduits the elimination of thc annular air space between wouitrI make litt!e diffcreIice in thc detertnination of the th~
conduit outer surface and the inner surface of the finalj correction "factor;( no effort'made to "match" Thermo-Lag. This was accomplished by thc complete Ii condliit segm eats (ie. sc((ect all from the same vendor). prebuttering ofthi intaior of the'preformed'sccnons of in orIfer to confirm that a'peatable results werc indeed: Thetmo-.~~or: ta.placing. them over the conduit.
being obtained, a second <'-inch baseline"conduit was prepared and tested with['a variety of cable'and power= Having i:i[mutated the air gap ai describ'cd supp y arrangemeati, as rfollows:-' thermal resistince from installation of the" above,'the'ddftiorial Thermo-Lag.330-1 appears to have been.ofQct by thc.
t/c aria corinecied'powered iriigfc'phese:"
siggifrcant it~ace ia die..overall.surhcc area and.by.
4-t/c saia coo~4 ypeeed iiiisre 8-3/o series coeqted pmraed siaaie
~.;
ph+
.~ ...' the inaeased surface emissivity of thc ERFBS 8 3/c powered direc piief4r compared to bare conduit.
'iI A coasistent 3A%'iffc'ppco in the tesuhant correction Duriitg-'a literatu're 'mirchv tV was-observers that the:
factoI was scen for the[ two baseline conduits. The recommended value for cceduit emhsivity haschangcd resu!Is are sho'wtf ih'TgO':I l -
signiflcintlyova the yeat3. Whai the Neher-McGrarh connected sihgle p6ii: I frtr'ihe'8-'9(c'cables'ulations'or thc diffcrttt work~was performccVin l957 a-value of 0.9$ was baseline conduitt- ee abclled ACFI and ACF2, utilized: In':-I962; when':ICEA P~26 was respectively. TbaI. of the I-inch conduit test generated, a value of 0.82,was appiioL kore recently, (whep only"a'singfi llhe condlllt'w'as used) are IREE 835'as utilized a value of-OW. Discussions also iven in Table I'I;" with those fimiiiar- with the conduit"m'anufacturing.
proces's reveals'hat:the chinge waa driven by TVA noted 'thac though,th c span of ACF8'mcasrrrek's improvements'in'he-hot~ ppxcsa and by a trend-.
close y mirrored'an performed'using chtssical towards the use'of electro-galvanized Intermediate McGrath mcrhodo lUgkfjthe corlectiotl'factors' Metal Conduit (IMC) and Electrical. Metallic Tubing.
wer~w ell above that ex baicrf on-TSI an'd'-K (EMT) in lieu of hot~ rfghf steel.conduits.. Thus .
it appears that the. difference observed between ACF I LL I ~
~
~~ ' v I
I
0 41 0
acssa I aY: o- u-oo s s s 3: a 'sr 4 ".
GtaKQ Ktlgc. 3 aves. ~ 3 a3u 240aC3 I ~
' Calculation PTN-BFJH-96-028 Revision 0 Attachment 1 Page 12 of 23 and ACF2 was a %action of variations in the surface The results are given in Table 13 emissivities of the, segments used in making the baseline measurementi. Tottic D Three Hoor Coodilit Upgrade Syslcla' During the course of several test programs. TVA has measured the surface I',mlssivities of thirty-two rigid her steel and five aluminum conduits. All measurements IalaaraeeaO>> oo werp made with a Mikron model M80AL-2FH hand ne I louse hei) gun. This device has a target zone of 0.2$ -inchres I g distance of 2.0-ijlches. The results of those at measurements arc shown in Table l2. Several
- Consistent with the P-848, observations are in order. First, thlt cmissivity.'of the the worst'case ACF was.identified for one aad four inch conduits for each thickness of ERFBS. Those rigiIIsteel conduits varills over a wide rangr . l! may be.
thatl this disparity wasI the result of. the age of the were furih'cr reduced'by f /o to account for 'actors various specimens or( perhaps reflective of their p'ossible vanations in the 'surface 'emissivities of respective storage conditions, ln either case, the installed conduiti.'iTTie"fhal 'values, which were included in TletA's intert tal design standard, are shown impprtance of "matcHing" conduit specimens (ie in Table 14.
draping from the e vendor stock or verjfying emijsivitfes) can be c scca. Second, the readings Tabie l4 for ]teel were frequ below that assumed in lEEE
'.n Scloctlee or Haai ACF- Coa4at t 835) For aluminum, the difference is even greater. ~
WhQe the low emissi itttas will result in favorable corri:ction factors whee testing an ERFBS, the results do sums to warrant somp consideration wlicn choosing SIP I.tease ' O.O0 I basclampacities. Since Ihanufacturing stindards'do not llr tel e IIter>>r ~
on specify an acceptable range of conduit emissiyity; P- al tli0. ~ Iboar..r v...',,ay...a.i.. an 848 has been revised to require that emissivity readings be taken to facilitate in)formed compa'riso'n of results ~l I sheet .... aof I
I an f'ronI differen tests: "'
7obi0 l1 ~ ~: ~ " '
~
Hoa-Stan datdrCo adult.CoaOgaratlons.
Coadtljt Kmisavtucs:,,
The. TVANPL'rogtaia also inoluded'assemblies to lees~ riaarei>>- - . isro, evaluate muMple conduit! encased'withh a common rs I ae eeeeI. e O~ Thermo-Lag ERFBS ..Two types of such.enclosures ert I IL .I enact i. 44t may. be utiUzcd. rn the, fortner, the Qat Thermo-Lag r
330-l.panels. are mounted directly on the surface of the 10-'i " 'gael f. -
l4 LS eiial '84 400 conduits, with aaintaatlonal pciiphetal air gap. This construction will be used oe all three- aad four-sided
~
0 4 v ji .Z'e0aoaw, O.N boxes (thc additional side of thc three-sided box being TVAI -, " -'"A: Sas) '"!" OS'a4inaaaa t .'m'. provided by a concrete wall or ceiling). The second enclosure type consists of Thermo-Lag 330-1 panels
"- %Re subicqucnt test pmgfaii-a!
aaachcd to a Unistrut fhune installed itouad (but aot touching),gabe conduita,, This augaucaoa win bc used 0 inclu dcd'2-iieet 44bch'caiId5 ~~4'with where the conduits to be prptcctqi ere routed such that TSP '-:thieve'0edt" "sy~ '&> phrviouatf only two sides of the box wilt be Thermo-Lag 330-l, b~"llo48of w~~uub -14 leering fQtc8'hf cotnplhaw
':wltich'includcdwlh of the lessIIns learned Ifhini'that'U.and earlier TVA efforts the other, sider being concrete walla or a wall and the ceiling. The. substantial heat,rapacity nf the concrete and rebar, coupled with its low thermal resistivity raeaItbntt'-specimen. configuration. aad'.connection s I I
a s 'sa sl ~ M s
0 0
Ca 1 cul ation PTN-BF JM-96-028
.....Revision
- 0 Attachment 1 Page 13 of 23 compared to Thermo.Lag 330-I, w>li ensure that such protected power conduits (which could noc bo
'two-,and three-sided enclosures are not thc limiting individually wrapped) within a typical generating cases and may beseprasented during testing through station was-low; the)use of four.-sided assemblies. "
I I Thc presence'of multiple Ealduits in ciosi proximiy to There is no specific guidance for the performance of ono.another (even without a barrier) results in mutual such tests in P-848. Thus, TVA first identified the kcy heating ance tea inaoductioa" of'hat Nch'cr an't parameters regarding such arrangcmcnts and'selected McGrath described as an "interfereace tempcraatro theivariables in a manner as to ensure conservative rise"., This eO'cct is typica0y accounted for through the results. Those parameters are thc selection of conduit uso of a grouping factor, such as is given in PA&426.
.size; num&sr of coadiihs (and their anangement into The tendency of the effect to-'dominate the correction rows and/or columns), conduit~lag (whero multiple factor when maay conduita~ wrapped'ogether, conduits are used) and. box size (in the case of two- further contributed to TVA'i decision to uso small sid@ boxes on Unistrut &ames). Those'ssues were conduit banks (lx3:aaC2x3).
addressed by, TVA's selection in the following manner, I
A spacing between coaduita of on&talftheir aotaiaal
&cfosurcr, Using drc. Conduir Jor Supporr - When diameter was utilized: Lesser spacings would result in encfosurcs are constructed of panels mounted directly a greater interferenco tetnperature riso effect, as to the conduit, the air gaps (between adjacent conduits) dcscabcd above (and'hus appear to minimize the ten< to vary in accordance with the <<nciosed conduit effect of the barrier) and are not feasible below oao-size,'Wth small conduits, tho conespondingly smak fourth of the nominal diameter bccausc of the physical air gyps result la tho heat uaasfcr across the gap being interfcreaco of coupHngs aad supports and as a reek of a fbnctioa of tadiatioa aad coadactioa only, rather thea the need for tool clearance Larger spacings wouM a I'ncaa of radiatiao,ccevcctiaa iad~xhctsNL generally support individual wrapping. of the conduits.
Th tho ACF derivedusiag "sma11"- c'oaduits would be
- e. ln kcepia'g with the P-848'philosophy; &closur'es on a Unltrrut Fraimt -'Ia contrast to the TV/ chose to utilize.1 inch Geduits since they aro the boxes formed by Thermo-Lag 330-l panels in direct smaQest size ia which power circuits are typically contact with multiple coaduits (which were necessary routed. to assess tho resultant dead'air spaces'bctweea conduits) enclosures constructed of panels mounted oa For such enclosures,'surface area ii low'est for low a UnJstrtN ihaao poteadally result in: a-" lirge gap numbers of small conduits: A single con'duit ~ould. between ihe'conduit swhcc aad: the inner wall of the.
provide.for the lowest suiYace arka bht'oufd'nos- paneh. Fice air exchange willexist between that larger'.
inclu'do any intetnal air gip-sinor TVA's m'eth'o'd of space aad thtr'smaH conduit~duit gaps, rendering appiicatioa is to utilize preformed sections'pre-butfered' the latter insignificant Tests'te asiesa" the effect of the withI trowel grado'ater'iaL" WMle a::twcr conduk- large'ap aad'tho. Thermo-Lag 330-1 ERFBS were encaaemcntwouid result in aa caclosaf air gap; each '- conducted using a single conduit;-.thus avoiding tho condliit'as an adjs~end wa0" Gem'which to" concern for "interfereaco temperature risc" mentioned radiate. Thus, a eel of ee coatkkl"ia'a row wai' above.'-
qg judg jk~jxinrlM~RaAMKci'%84.whilestill-includmg a coaduit4hfc59&it idjaccnl'to'one of tNe:. Those tesat were coaducted 'itsing-I-inch'oaduits, i
ead'aBhj'Sich'atr arrangcm@t'using.h-inch conduits,'wiihC5i& the OK testatg;. TVA'-
recogntzed that ~5QQSt'ae baakad m mulaplo."
'acldsure givctr that its thermal resistance ta the surrounding air is higher ihart that of a 4-inch (duo to the smaller sur&cc area of the foitacr); A'i tlted cariicr, this is the mabel atr addltioaal 'alt" gap Ia taloctod bettveea tho'. smallest size: conduit--ia which@neet: circuit. are 2,a3 rawbabas; ht onhrta asaaaa thatstlsa'ri'asaembly typ re~
htchdat 4hkh'hittl gdoubQ~5f throiH I'-ihcM 'aa coathhtL While'this arraagctaent leaves each r'ow with '- For a given size caaduit; the tnhthnum box size. (and adjaccat fb a radladng:surhice, it wis. fcl!'that'ho .
likelihood'of'encounterin" thror rows of ERFBS 1 'I yl ~
by the. diameter of the. conduit; tho th~
therefore thc minimum air gap thickness) is established of the t
1
~ ~ ~ ~ .ap
's .1 'q. ' <~
0 0
0
Ca 1 cul ati on PTN-8FJM-96-028'evision 0
Attachment 1 Page 14 of 23 Unistruc and the gap between the conduit and the of cables above the sufecatls could sometunes preclude Unistcut. For a single 1-inch conduit with Thermo-Lag the encasement,of individual trays (necessitating the mounted over 1.CLS" P1000 Unistcut. the smallesc enclosure of several trays in vertical stack). ln some Possible box would be approximate.ly 4.75-inch x 4.75- cases, the mounding could be addressed by raising the inch (as measured a . thc Unistrut). tn contrast. thc siderails, though the effect of increasing the thickness larjest box (regacdl of conduit size or numbers of of the air space above the cable mass would have to be conduits) which TVA ticipated was be approximately evaluacccL Finally, inadequate space between adjacenr 30~inch x 30-inch, TIIough ic was believed that the trays could necessiate their common enclosure.
(and therefore the to ravidc thc needed onficmation.
~
.lee box would factl It tc better convective cooling ACF), both-boxes were built As with thc non-standard conduit configuradons,'here is na specific guidance for thc performance of such tests in. P-848.'hus, TVA identified thc major Th) final ACF for box canduits on a Unistcut frame varianans and evaluated each ta detenninc ifadditional wapld thea be mast co nsccvative of the two determined tests werc required., For thbse configurations which, above tests,a cl adjusnnent for the appropriate requited additional testmg, it was noted that only minor ing factor C tsten t with 1EEB P.-848, bodr excension of the P-848 logic was necessary. Those mblies.were 12 ee) in length. issues were addressed. by TVA's selccdan in the following nuumcr, with the results given in Table 16; I
Th results of the non-standard conduit testing are given in Table 1S. S/dd-by Side- Upon review, it was determined that the I I results of the TU tray casts cotdd bc usccL to represent kaMe ts thc common cnclosuce of trays which are horixontally Cotrcctloe Noa4taadard Coodotco adjacent (ie cun side-'by-side). This acrangemeat is Foots ~ t, consistent. with the.Stolpo modeL (an. which tray ampacities given. in ICEA P-$ 4-440 are derived).which 0.'N 4~ considers that heat is dissipated out the top.and bottomc surfaces.only. " - ..:
~
SOU Aisec lac. aa.l 4aaiia QCS
',4rasuaenc bee.tet Ofiecc~e1Ll e I OIIIcc exwal o I 8 t
I ~ coeLw's 0.01 perfocmcd on ladder type thotfe. roatlts'ould'also 4e
~~
Solid:Bottom frccyr I- Zsough;the TU: testa. were-i@was determined that to represent solid boaotn. trays. This extension is conservative in that tttto solid bottom uaysdo not have an air gap bcovecn.
~
thc bottoim of thc cable mass and the ERFBS barrier as.
dard Three-Hour 'ta ray a result of the presence of the tray rungs.:.
The Itest pcognun at OPL,also included trays. wrapped" Coviemt1'rayr <<The test pcogcacn at OPL addressed the withja composite Therm j;Lag 330-1/770-1. three-hour enclosure of ladder type trays over which a sheet steel upgtIsdc system as was IprcsIiously. described. Those cover was applied pria'rto tho'applicadoa of any barrier tests'/erc configured ictqmpUanco widt Dry 14 of P- materiaL During application of the ERFBS, it had bcea The ccsttita are gi in Table.15. TVA's objecdve to" ensure that no alrgap existed IO ~
between the upper surface of the cover and the tandarO Toy C Qgttradona underside of the barrie. This vras accomplished by applying trowel grado Thermo-Lag 330-1 to tho top of Con tructabtTity at 1VBN had determined thc cover prior to ittstalling the panoLL Unfoctut tatcly, that not aB. 'bo-'individually wrapped when thc masan was applied, thc dpi sheet stccl cover t witLt.tho.~ gg~aipy outlined In,P- sagged and moctcmasdo was ccquhed ta.obtain.a level, walkdo had showi that elccirical .
surface sa that no~ wouldaxlst. Thls yraccss was.
s acaci005'
~ ~ i ' rc peated = undl ..tho'umtdadve. thickness;. greatly incluIsioa.of~~- 1-top cover. undemeathc the exceeded tho nominaL 5/8-inch; Thus, cho <<elm depth o-Lag'33&1 or, wrapping of solid bottom which accumulated'during its installation resulted in a Tho same. coYMwf shown that the maunding iw 14
~ s
0 a
Calcula6on " PTH-8FJH=96-028 Revision' Attachment 1:
Page, 15. of .23 final multiplier of 0.60, well below that which had been region's 'and modelled using conventional heat trutsfcr predicted. prir)ciples. Their or)alysis suggested corrcctio(t factors ranging from 0;7$ to 0.78 for onc-hour. Thcrtno-Lag Vcr)ical Stack- Thctests also included a vertical stack 330-1 systems applied to tray. These values were consisting of ooe control and two power trays within a 'conservative compared 'with the.vendor's. original common Thett)N-Lag 330-1 ERF BS; with'th'e control published'ntimbeta (Tabio 1). However, data published tray on the bottom. Though a measured baseline was 'ihce release of'their paper, appeared to indicate that detgrmined, for this multiple tray arrangement, the th'e mo'del was non~onsetvative. Thus, in the absence actual ACF utilized by TVA'was based on thc morc of confirmatory tests, TVA's eQorts tumed back to co(tservative equilibrium cunent for a single tray . identification of acceptable test methodologies.
Thg ACF developed for'thc stacked triple tray arrangement will also be used for cases where the
" After completion of'the'TU,and TVA tray tests, the common enclosure includes two power trays without 'nowledge that'dditional. tray configurations were the. control tray: The 'applicatroit of this ACF is liki:ly tt) be enc'ouantcred as thc fire barrier installation conservative since: the two'ray arrangement affords procesos pro'gressed at WBN,:lcd TVA to;revisit the more direct heat dissipation'or the l'ower power tray Save-Engmana.analysis... With a cable depth ef 2.95-
' 'inch, the 'ACF established by the TU tests can readily (hau ttte'tested three tray enciosut'e'.
be compared to the 3.0-inch computation in the Save-
~
.. ' Talc Engmann paper. The inital review, the results of fatrosi-TroyI l6'orrccaoa which ate given in Table 17, indicated that there was a significant discrepancy. bctweca thc inodel and the TU Caariyvasaee results.
Soleil Ooaaaa ~ Iboar 049(sew I)
Sba)a waI ooo ave I boar ,. PrcHafaary Coaoparfooa of hCFa YorOao) gaa a I)NMr 4)9
~a 0.4! (RW I)'
" Cob)o Oa)ab (4) ~
Trav Oa9OI(4) ., ACf.
5a)ebs'de
~
I boar SwaaO Iroo -1 boar 4$ f= S~oww.. '.',.'+, ! ) . O.ra a'ore.).
Those. values are by aukyati. no )pectN test. T(r ' ' }.99 pcrfonnocL Note) . 'r 'Ai tokco 6eiII Tobfe'1 of tt(o rebrer)ceo RaLfaf Sickrai/s - During the course o 'preparaiion f for'he non-standard fray tcsthg, good conctation between The value of Thermo-Lag 330-1 thermal conductivity the KU'results and a mathematicat model~as.note(L -'"'ince used by, Sivc and Engmann may hive been the cause of thc model had'the ability to address'the effects'of"" much of thfsvariance; When the paper was prepared; raiscct sideraiis, no special testing was'erfotme(L they. liad obtained a value".of thcimaf conductivity of Results of the analysis performed are ghcn bclow.'"- 0.430'Waits/{meter-'C) from TSh'ubsequently,'TSl a
'l has refuied: their estimate oE this parameter 'a'nd have Savh-'En'gmina Modil-' id(a)tificd 'that it"h)fs )trfairly st'rarig temperature" dependency:. 'ased od the Thcmfo-Lag'30-1 As TVA identified tha pcef to.obtaia pew correction temperatuie.measured during the tcstitig at(OH- a i
factors for wtmppefftlya) sutvcy h)d been conducted correspo'ndihg conductiv)ty of 0232 watts/(meter-'C) has been used'ia4e foHowing rc-analysis:.:-
whirrh co'uQboueekc4c(s a aa of molesting.
I The only other modiflcatioa to the Save-Engmana A litcratute scatch revealed that such a model had bcctt model necessary for, direct coin jolson to the TU pro+ised'y PhB- Save" and" Gary "Engmantt i the additioa of sunple"c(tuatlotr to 'evrahiate the cffectrofthe silicablanket= .Th~ee dlNiact refyona.
data's ofthe other appDcable. litetaa)re for..: ...'ollowing'a'review f)re;it)rotected trays, Save and Engmann suggested that. of the model were reducecf by Save aiid Engmann tr) a ttay/Qrc barrier systcta couldbe divided into seven system ofhvo cquitfoiis witt tv'l(r)o%rS. Thi two
'he 4 i a4 J +0 ' ';RC vAV'"ra Q a I a ~ ~ rI I p ran o
a
' a V '
0
~ ~ ~ ~ v~r ~ ~ uur v ciliif ~, a 4'v ca. ~ 013052466231:a) 7
~ -" '= "-. Calculation Revision.Q.......
PTN-BFJM-96-028..
Attachment 1 Page 16 of 23 equations, Noc. 5 and'12 from their paper, have been thc TV tests, the ACFsshown in Table l ofthe Save-rcptoduccd below (defmiuons are given in Appendix Eiigmann paper ca'n be recalculated using the modified t). I Thc system cao. be solved'teratively by first parameters. For these standard configurations (ie no g uc Png i a value fee th two unknowns'a and s , th e silica blanket) 'equation 2 is not required. The results asymmetry of the le mass and heat intensity, arc shown in Table 19.
res ppctfvcfy. Thc ncw CIluation determines an cFcctive conducdvity, km3, of c combined mass (cables and One diAiculty in establishing a w'orkable'model for bi+ ct}. As before, thk sys tern is solved by guessing protected raceways'has been the P-848 requirement that valup for the unknow/ an d iterating until the system the bascTine'used'in dctctminatioa 'of the ACF be conge rgcs; With these bh anger; and using a baseline cmpIrically cstabHshed. Such a requirement works a valub' ofs which cortespt)n ds to the actual baseline test particular'hardship on the dcvc:lopment of a consistent CufrC nt (rather than a num inal value from P-54-440), model for use with cable tnys, given the conscrv:uistns the'ab odel correlates very well with the.TU tests (see inhcrint in the Stolpe modeP. Stolpe assumes no air 18). flow through the cable'mass. and his original tests l included thc use of a plastic sheet across the bottom of Ta bk ta the tray to block such flow"."Th'e usc"'of'a'measured 0m paT(eon or MortUI Save-Knarnaan <<uh Tlt Testa. baseline current as imposed by'-'848 results in a loss of rcpcatibility since the efficiency'f cable packing Trav Drorrr all Cnrle Daonr iirr) (and thus air flow duough the mass) will vary from tert O.0T to test (reae 1) a00 4 0.4l i As an example of that conservatistn (and thus the I (rro<<()
potential variability induced by the use of a mecLsurod reer vr0llNialre '-- l29$ .' 4' ~ ~ 0 ~
baseline current), the normalized current in the TVh single open top tray test was approximately 11%0 above.
o(c t. Vati>>a us)nod>> actnatZU.base(tnc heat. the. adjusted. 1CEA cunent:fl)r: the same.diameter and depth4
~
Ir g.shown thit thcmpfel favorably. co(it pares with I ~
Sav ngmann.equation((5)'.; '...- " . ....: ..:;;;;,
1 iianf 'v((1)rr- - )-1'c)4, +sbviob((5)ir- ; ) -1Xt })"'
(.1)
I I uation use@to evaluate the effect of tho silica blanket.
~
Ncw ~
- II 16
~ W I ~
' ~ ~
V DI
0 0
vaI ~ I V ~
s~-~v o i d: ~ vnrM s UVKQ angl . 5 SVCS. 813052488231:F18 Ca 1 cul ation PTN-SFJM-96-028 Revision 0 Attachment 1 Page 11 of 23 Save/Enynann cctuation (12)
)- 1.911 a(r -rc- s' W
(t o) d t 4 d (
-cern-OAQ td(t-O) ae IM~
~2 TKIF ltd q J~
'-(t-r~h~rr kg i )
~
~'
(3)'...;... 3 Table l9 Seve/Eaauseaa Theres~ ACFO
., (+ ). 'ear Sa~aySIdaet a JLS'd Ma&A(IAs&ale
~5weCseaaee" 12$ 'd MddCnde Aaaheis 34odr S~+~
0.1$ 0.41 ass I O.s1 1
0.1$ a01 0.10 ase ! 0s1 5 0.11 041 0.10 1.5 a10 0.11 a14 0.11 0.11 2 1 0.14 4 ate ~ ~ > ~ an . a$ 4 5 a10-" ..... a00..... ".'..9u.........j OAF as4 0.10 0.11 05$
asl 0.55 0.$ $
$ 0 4 on on oc Note 1. These foctots ere based on the nombril 0.30'lthica'ofhw'baricr and orisiaal Ttxnn~a conducdvtty.
Nose +,. Theec Ssclors ~ conscrvsclvcgbescdonlbamaahaen %chiesa(0625 Lof Qw 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> berrlccen4 tba revised Thcrrno4aa conductklflty.
Note 3. Thew factors enrbescdunttw&ainsH4ourdcsiasu-. ~ '1 Note 4. These tbctore are based'onew'nominal thickncse oMe 34owupgredesynenc, Notch:,-:AlbcncskebescdonnomhaHCRh;vstucs otbcet bucnslsyds:...r.:.".;= --.: ~-.:-:-: --." --".--'- ~-=-.-. ~,
Tltc ~eQt of snc)~odd is that cable sizing for a- - - - only simple modification of the input parameters.
d&ld I, Idd-,N time and expensi'of- testing per P-84$ once.i -'
. d Inthecascofanupgradesystem,thethicitncssofthe con aroty testa liive beeii 'p'eBoimccC "'Such EK:BS;ib'and sr;inust b'e adjusted," L'Qcewise, where conifg'uratient$ $2ight hclude various upgrade- -- ---raised sidcraiis exist, the paiametcr dmust be adjusted.
traZL~gggsei.'Qyqno-~330-I side rsHs Table 19 showa that such extensions typically result ln (to c car undcriying cables which axtcnd above then "-- -'an additional raUs) and trays with"removable'sficct metal top or.-; ' '" dcrating of Bo or less.
""' -" - . - -" "i botto covers under the ERFBS."Each castFrcquircs --* - Application.-of-a..tap cover 'or '.battotn..respires h 0 4t V
0
~ f 0
II ~
~'
Cat'cul idion PTH-SFJH-96-028
~~
Revision 0 Attachment 1 Page 18 of 23 adjusunent of the parameter err. though in the case of. design of Thermo-Lag ERFBS.
the TVA testing; the sagging of the top cover fumhcr I rcq 'rcd cho odjustmens for bmricr thickhcss.
SUMMARY
e FUTURE WOE The 'HA tes! program has demonstrated Thermo-Lag materials, when properly designed, tested and installed At t is writing TV% is plannmg onc more fire.'test in are'c'apablc of providing rated one and three-hour late )995. This testis necessitated by diffcrcncesdound< ERFBS.
cika 1985 Thermo-f.ag 330-1 installed at SQN and
'n the @erma-Lag 330-1 materials used in thc testing and TVA ampacity test program and results'<<have been
-installations at %BN BFN,. The anomaly eviewed"itang with a rationale for tests af multiple while pe reforming Thermogravimetric was'iscovered raceways. within .a .comtlton enclosure. Recent Ana)ysis (TG'A) and I frared (IR) Spectroscopy'n ampaci(y tests conducted an,cables in tray by TU have r esp pnsc to the N s request for., additional, " been'a'nafyicd and compared with an existing information". Thc full scale fire test v/ill attempt to Modifications .to that model have'een thermal'model.
demI)nstrate tliat the minor, difference. =. in the proposed which. provide good carrchtion with the TU composition of the rent vintagcs of Thermo-Lag "'results.""
willnot adverse] effect the perfonnance of the diQ'30-tf ERI'8 S. e ACKNOWLEDGEMENTS A ca cuhtion methodoloIDp based an the mass and size The authors would like to acknowledge tho of lectrical raccwaysl including'he cable fil] contributions made ta the Thermo-Lag project by the abtte) protected wiSP ercno-Lag materials should other TVAteam members, Thermal Science Inc., Texas apcci An appro ach'imilar to the work Utilities and Omega Paine Lalsarataries.
pcfffed by the'Amcricru) Iron and Steel Institute sauclura] stec] protect]a'j w v'e'ty uscfu] to fire Special thanks are extended to Gary Engmann of B]ack for'uldbc, pro tion and electrictII en gineers involved in the and Veatch for his time and patience in the discussion
'I ~ ~
ofhcs work...":
~ e ca 'PP NDIX I - SYMB
" LS'avet Engmann a distance from cbc b'ocrori othe packed'cable mass'to i'pl~ Nufacc rINs ls'a'ihe maximum'npctaturc'n tbe pack'cd cable mass
~ .:,,, ...,, .. -
(meters) ." ~ ~ ' ~
s ~
. hcatgcclctaticus le' oiume of packed qsbtc macs-()sects/merce') '" .. - .: ~ ~ ~
cm cob cmissiricy ofcbc cmissiricy of the ~ide cabtc mass surtbccs (dbncaskultcss)
OfibCWrari'maceriatcri'iheseiiOmOfrbetray(dintCnciaaieSS)" '-.'-..a '- ' ". S'i" cmisciricy of ibc outside ofrbc wrap nascsiai orf cb'e lap of lbc tray (dlmcas(onlcss) "..: " -:,":" 'v'.",
cu I
- f cmis&ky of cbe iaswdc o cbemasaial on cop of sbe slay aide rslts - lray curst; wrap naccriak ccc. (dbrlcaslonlcss), ~ .
ke -'. thermal cooducciviiy of alt. '~tmrrct~C)3..., p sb Scc~ '
j,y4j e-ti)'[w'aej~~ t%
K')
c siuckiiesaof rbc cable i(miner}'t a ~ ~
Tc temperature oftbe (degroes C)
Tw lelnperanuc of cbc cable mesa (degrees C) km lbcimN " "" .of de. pecked cable mms Iwau/(~grec C)3 ksb chasmal 'ot'chc macci9fon sbe boccorn of cbe tray Iwaa/(meccr4k'gicc kss 'bcrnud bf wrap cop of ibc tray (wacs/(mc~cgrcc C)]: "." C)]'namrfironcbe ~ ~ ~
" " .~
Cab]e 'ruy'(mcccra) d TKAI .
toadiag'de'psbHC
.cernperacurc of cb ~ 4eemm (dclcca *...
- ~ ~ ~ . << ~"
~
a, TKw lcmpcracum of cbe (degrees,K) 'ass sb chicknesi of cbc file mscendal as sbc boaom ot'rbc riiy (merers)
' ' '<<r"3:. e'.le r
~ ch)ckncs'a of cbc tire craacsctfat ac rfrc sopor ibc tray(mcccrs) ~ a ., 'r.'ray W . loading.wkiib of cable (meters) I " . " ~
~
~ r ~
Ra Rayibigb atunbct(d mens ~ t'oatcss)
Nu Nussels numbct (duSIcasloai ess) 1 e
e der s w rent. (
\, ~
us ~
,e
~
<<r e ~
0 0
0
Calculation P7N-HAH-96-028 Revision 0 Attachment 1 Page 19,of 23 hcrwcc33 ihc iop fshe packed cd i of ihe iray side raiis Addltioaal Parasactcm kc cffccuvts 3er333aicotd33csiviq ofihesiiica cloih fwaal(33cicr-degree C)l cah(omahas zc u3ldc33eee of shc nlica doth (333cicisS i,ml errccshttadheisiial oot3dttciivisy ofihe33ppcr potsiotsofihe packed consider(nit ihc <<(fees of the siiica doth (waa/(mcicrwsrcc c))
n ioiai th ckntsss of ihe cabie mass and silica cloih (o3cicis) r APPENDIX 2- REFERENCED TESTS",
ONE HOUR FIRE VESTS 6.1.1 "Fire Endurance Test.of a Thermo-Lag.330-1 Fiie Protective Envelope (1-in. and 5-in. conduit configurations3 and 2-in. air drop coaflguratIoa)"; Omega Point Laboratories Project No, 11210-94S54c.
6.LQ "Fire Endurance Test of'a Thermo-Lag 330-1 Fire Protective Envelope (1-in. and S>>ia, conduit configurations I and 2-in. air drop configuration)", Omega Point Laboratories Project No. 11210-'94554m 6.1.3 "Fire Endurance Test of a Thermo-Lag 330-1 Fire Protective Envelope (3<<inee 2-in., 1-into and 4-in. conduit configurations).", Omega Point Laboratories Project No. 11210-94943m 6.l. "Fire Endurance Test ofa Thermo-Lag 330-1 Fire Protective Envelope (3-in. steel, 3-in. aluminum, and 1 1/2-in. steel configurations and generic 2-in. and 4-in. tube steel support members)", Omega Point Laboratories Project No, l l210-94943b.
6,1. "Fire Endurance Test of a Thermo-Lag 330-1, Fire ProtcctiveSnvciope (l-.in 2-In 3-in., and 5-.ia,.conduit and five junction boxes of varying sizes)",.Omega Joint Laboratones Project No, 11210-94943cL 6.1. "Fire Endurance Test of a Ihcrmo-Lag 330-1 Fire Protective Envelope (large junction box and three 4-in.
- -- conduit'sections), Omega Point Laboratories Project No. 1 1210-94943e.
6.1'., "Fire Endurance. Test of'a Thermo-Lag 330'-F Fire Pietcctive Envelope(Tltree 18-in. Cable Trays and a 3-in.
conduit)", Omega Point Ldaborato'ries Project No. 1 1960;9718$ ;
"Fire Endurance Test ofa Thermo-Lag 330-1 Fire Protective Envelope(Special Tray Fitting Pith Two 18-in.
6.1.
Cabte Tray Sections)"; Omega. Point Laboratories Project No.l 1960-97186.
6.1... "Fire Endurance Test ofa.Thermo.Lag 330-,1.Eire.protccttve Envelope (Common.Enclosure with Three 18-in..
Cable Trays and Covered 18-ia. Tray with 1 sin. and S-in. Air Drops)", Omega Point Laboratories Project No.
11960'-92187'.-'.t.lf
'Fire Endurenee Tert or n Thermo'tutg 33ntt Fire Protective Enveiope (Two Sided Mutdpte Conduit Enclosures and Cable Tray Support System)"; Omega Pomt Laboratories Project No. 11960-97257.
6.1.1 "Fire Endurance. Test of a Thermo-Lag 330-1 Rue Ptotcctive Envelope(Three Sided Box Enclosure Encasing'Grou ps-opf Horimntal'Conduits and'a Large Junction Box)"";Omega Point Laboratories Project No. 1'1960 97258!
6.1.15 "Fire Endurance Tm'ofa Tlfc'rmo-Lag 330-1 Fire Protective"Envelope(ThfetiFour Sided Box Enclosures.
Eaca'sing Qsstups ofVertical Coaduits and an Enclosure Eacasing a4 ia Conduit and a function Box)", Omega Point Eaboratarics Projec! No; I 1960-972593"'.1.1 "Fire Eashtraaal Tcstofhlltcrmo-Lig330-I: Fire Protective Envclo'pe'(Box'Enclosure Encasing a Gang of Sevea 4 iss'.Ssisal Coaduits and Iadividuat Enclosures on 3/4-in Alufninuin and Steel Coaduits)"; Omega Point
%11960-'972M;. '
'". w
'"We~wc -" '" a':-': VHREF HOUR FIRETKSTS- - '>>-= '-:, --,t..
62.1 'FireEisduiatQSTc'sio'fi'ihunisktg 330-1.Fire Protective Envelope(12ciis aad 24-in..Cablss Toys,'and a 12-ia.'x:12M:x 60'-in Jusictiott'3oss)<<",=Om'egst'.Poh)t Laboratories:ProjcctBo. 11960-97555.
622 "Fire Ehduriuicsfrest afallseHA~330-1'Fusr Protective Eavclope(12 ia'.ands-ia Cable Trays,aui5-ia.-2-'in:; and'1-:hr Steel Coadtiia)",'.Omega Poiit LaboratorIcs Project No. 11960-97553; r
19 e
r
~,
r t.l W ~ ~
~ ~
~ '~in'~e *
' 't',~
0
~ !
II,
Calcul'ation PTN-8FJH-96-'028 0 'evision Attachment '.1:
Page 20 of 23 .
R EFERENCES
- l. Cade of Federal Regulatians (CFR), Title 10 Part 50, Appendix R. Paragraph lll.G.2: '
~ ~ ~
- 2. A detailed discussion on thc mechanics of Thermo-Lag 330-1'an be found in thec NRCJNRR Og Ncial 'al Transcript of P roceedings IIMeeting with Thermal Science, Inc.. io Discuss.issues tnvolving Thermo-Lag 330", dated October 17 l 991. N RC Public Document Room Accession Number 920205305.
- 3. NRC Information Notice 91<7; "Failure of Thermo-Lag Fire Barrier Material to Pass ass Fire
.,lre Enduutance Test"q T Augiist I I
~ ~ I ~ ~
I
- 4. NRC Information Notice 91-79 "Deficiencies in the Procedure for installing Thcnno-Lag Materials", December 6, 199 . NRC lnfarmation Notice 92M, "Thermo-Lag Fire Barrier Materiah Special Review Team Rc F d C ent Fue Endurance Testmg, and Ampacity Chlculatioii'Eirors~ Sune 23, 1992. NRC Information Notice 92-5$
usrent Fue Endurance Test Results for Thermo-Lag Fue Barner Materia, July 27, 1992. NRC Information Notice 92 8I2, "Results of Thermo-Lag 330-'1'Combustibility Testing". NRC Information Notice 93~;"Results of Thcnno-30-1 Combustibility Testing", December f$;.:1992;" * '
Bulletin No. 92@i,."Failure of Thermo-'Lag330 F'uc Barrier Systems to Maintain CabHnggin in Wide Cablee Trays rays mall,Conduits Free from Fire Damage", Sune 24, 1992.
C Bulletin No. 9241,~ Supplement I,"Failure of Thermo-Lag 330 Fue Bamcr Systems to Perform its Speci6ed
~
~
durance Function", August 28, l 992.
~ t 7.N C Generic Letter 9248, "Thermo-Lag 330-1 Fire Bamers", December.17, 1992.
~ ~
- 8. Subject 1712, "Outline of Invcsngation for Tests for Ampacity of IitsuLucd Electrical Conductors Installed in Fire tive Systems", luly, 1984... ~
- 9. IREE P-848, Draft lEES 'Standard.entitle d, "Procedure for the Determination of the Ampacity. Derating of Fire
~ .+10.
Cables"; under preparation hy Task Force 12<5 ofthc Tests and Mcasuriments Subcomnitte'e af dt'e hiuiatcd Commiaee of the IEEE, -,,;
Mcrmo-Lag Elcctncal Raceway Fire Bamer Systems. Testing and QualiQcation":,
BTOTITI, delivered at the February 1993 WATTEC conference in Knoxville, TN; " Mark, H. Salley. and Kent %;
's iT.~ 'l"l": tt I ~
pacity af Wrapped Cables", Keith A.gcttysiIEEE yapcr 86 SM 398-2, IEEE TrattsaTctlons on Power Delivcryi voiuke 3, number 1, January 1988 pp 3$ -38:; ......
- 12. P 'ters ll' 13.N C Request for Additio'nal fnfonnition Regarding Gen'eric Letter 9248; hsucd Pursuint i'.NCHVM4(0dated aber 22,1994.. -"~- -.
- 14. N C Gcncric Letter 86-10, "irriplemeniation of Pire Piotection.Requirements",'datedAprt -241986..,
- 15. A TM E-119-~'.Standard Test Method for Fiie Tests of Building Construction and Materials".
- 16. IIMAERPRAQifdalincs for Fuestop and Crap Systems ofNuclear Ficilitici;Article B, Revisioii 0, 1987'
- 17. derwritm;Labeaary'Subject 1724, "Outline of Investigation for, Fire Tests for Electrical CIreuit Protective Systc ", Issue nutnber 2, August 1991.
- 18. C.Standard. Review Plan, NUREG<860, Ication'9.$ .1,'Fi're Prate'ction Proyim".'" "
19.
to Se CGeneric Letter 86-10 Supplcmcnt 1 "Fire Endurance Test Acceptance Cnteria for Fire Bamer Systems
'edundan't'S'afe Sliutdown'-Tiaini'within the'Samo Fire Area" dated" Mirchl25<! 1994.
~
used 20
'I I ~
~ t ~ l
0 II
Cal cul ati on PTN-BFJH-96-028 Revision 0
."* '" 'ttachment"1 .
Page 21 of 23 2p, TVA refers ta tho niatcrial as nominal 5/8-inch or 3/8-.inch rather than minimum 1/2-inch or l/4-inch -ol '
o TMs is duo ta TVA QA acceptance requirements. The taleranccs used are 5/&-inch + 1/8-inch <<nd 3/8-inch + 1/8 Inch.
ments taken durin the construcuan of ihc test assemblies indicated the material was typicallyy -inc or 3/8-5/8-'nch.
The ambient starting temperature of 83'F must bc used for thc equations to be valid. Figure 2 has been simplified (i.e. the ambient.temperature subtracted} ta graphically show tile:allowable tcmperatu're rise (i.e; BT 250'F),
23..The Calcuhtian of thc Tem pcrarure.Rise and E.oad Capability of Cable Systems",, L H; Neher and M. H. McGrath, AIEE Transactians, volume 76, October 1957, pp 752-773. i ~
- 24. ICEA P46-426, 1962 edition, entitled, "Pawer. Cable. Ampacities". '
Published by thc Insulated Cable ~ Engineers 1 ', e ~
Association'.
- 25. IEEE Standard 835-1994, entitled "Power Am pacity Cable. Tables", dated, December 1994, section 33.5.
'i
- 26. For a single open top tray, an equilibrium current of 29.48 am ps was determined. With tho triple tray. vertical stack, an equilibrium'cur'rent of 26.19 was'ineasurcd. This'I I~to difference is the result of a mutual heating effec. This effect is'not presently addressed by any standard. It is iho author's experienco that some companies apply a SN detate to account for this effect whereas others ignore the influence of other trays in close proximity because of thc signi Geant diversity that exists in nuclear station power trays.
- 27. "Fire Pratecdan Wrapped Cable Tray Am pacity", Phil Save and Gary Engmann, IEEE Traiisactians on Energy Conversion, volume 4, number 4, Decanber 1989, pp 575<<584.
- 28. The seven regions o'F the thermal mixfeI proposed by Save and Ehgmann were:. I) 'who kiwerjunction of the cable.
mass; 2) Thd Qiewrapbelaw the bottom af the tray." 3) Theolitside mr&cc oHiewrsp bchnv th'e 'bottatn'af thc'tray.
- 4) TItc upper portian of the cable masL 5) The ah'space between the top of the cable mass and tho fire wrap, 6) TItc firo wrap over the top of the tray. 7).Thc outsidyaurhcc of the fire. wrap over the top ofthe tray,
- 29. "$mpacitics for Cables in R@uhmly Filled,Trays", S,Stolpe, IEEE Trat tsacttons on. Power Apptiratus and Systems,.
Volutne 90, Piit I, 197), pp. 962-9)4 .. .
r
- 30. SUPE frst cditian, 1)88 National FireProtection Association, Quincy, MassaIchuseas 02269, Chapter 34...
~~ ~ i
~
li
\ V I
-. ~
~
~
A
0 0
1
~ ~,
jS c l
~1" dia condut I -1 taye<, 516" (TVATost 6.1.1) ~ 4 50'. AT MhXNUMALLOWED
~ i'>; l 'i, 1 dta cooduit - 2 tayefs, 516)
+ 3Itt" (TVATest 6.1.3) 4
~ ~
~
~3 die collduit-1 taye<,5/6" (TVATest 6.1 4)
~ f,r, ~3'dia conduit - 2 lay e4. 31tt
+ 3IS" gVA Tesl6.1.3)
~ 5" dia co educ 1 taye<,
Test 6.1.1) 5fQ'TVA
~ T '
~~'AA gy ~ID of I
jw
~.4
~at%~
gg gg~O O MP o~g~~ -~
~~OO FO I O
I CP Tl
+I lO Ol NKNR RATED THERMOUS 33M CONDUIT ERF SS
~ ~ AVERAGE NTERNALTEMPERATURE RISE (5T} co NO CABLE FILL
0 Cl
~O
~
't I
~
C
~ r < ~ '.
250'T MAXNUMAllOWED a',
I.'-HOUR i'
r I
~" i i i ~ ~ ~
RATED T HERMO4AG 330-1 CABLE TRAY ER1:BS AVERAGE NTERHAL TEMPERATURE RlSE (hT)
~18 Cabhlray-Nocabhhtt gVATeal 6.1.7) 18'able cabh tray - 6.24 hs/ft 68 (TVATost 6.1.7) 1
~18" Cabh tray - 69.36 Ibs/ft cable Q (TVATeal 6.1.7)
~+NA i't<
as <<+ iD
~BOA
~A (ii gy Vl g1 oa$ 5 8
~~WO
~l A
~
C D
C I
n LO I
FO CQ