ML17277A996

Hanford 2 Power Generation Control Complex (Pgcc) Floor Plate Design.

ML17277A996
Person / Time
Site:	Columbia
Issue date:	09/09/1983
From:	Artigas R, Hart C, Spencer E GENERAL ELECTRIC CO.
To:
Shared Package
ML17277A995	List: ML17277A994 ML17277A996 ML17277A997 ML17277A998 ML17277A999 ML17277B001 ML17277B003 ML17277B005 ML17277B006
References
NUDOCS 8311090364
Download: ML17277A996 (49)

Text

Enclosure; 3 HANFORD 2 POWER GENERATION CONTROL COMPLEX (PGCC)FLOOR PLATE OESIGH C.W.Hart, Manager.Panel and Assembly Oes.'gn Engineering Approved:.E.C.Spen, Manager Product Engineering Approved:ISA'.

Artigas, Manager SWR Projects Licensing September 1983 RUCL-AR"=NGRGY BUSINESS GPER47ICNS

~QEHeRAL=4"c A<CŽC'A~44~=At4'GS 4L P'CRANIA 95'."5 c<N rsaL I<L<cvstc 83ll0903b4 83f028 PDR ADOCK 05000397 I PDR~l+

CONTENTS PAGE IHTROOVCrIOH PGCC LICENSING BASIS OESIGN DESCRIPTION SAFETY EVALUATION CONC LVS ION 10 REFEREHCES FIGURES 12 LF: csc;rm/109081" 9/9/83 2.

INTROOUCTION The HRC"approved GE Licensing Topical Report for PGCC states that the floor plates are composed of steel and aluminum and will sat,isfactorily withstand a design basis fire, as described in the LTR.A small percentage

(~4%)of floor plates at Hanfor d 2 are of a non-metallic material, namely Richlite, in order to accommodate rregulary shapes and close tolerance fitting at the plant site.The NRC ffice of Inspection and Enforcement has requested more information concerning the fire protection capabi;it;y of the Richlite floor plates.This report responds to this NRC request on behalf of the Haniord 2 project.It includes tne accepted licensing design basis, a design description, and a safety evaluat.'cn that demon-st.ates the acceptability o<the Kanford 2 PGCC floor plate design.PGCC LICENSING BASIS General Electric Company documented the safety evaluation of the Powei Generation Control Complex (PGCC)in the Licensing Topical Report (LTR)NEOO"10466<.

This evaluation addressed the fire protection issue.Ihe LTR described how the PGCC design provides defense in deoth by preventing fires from starting, detecting and suppressing any started fire, and safely withstanding a fire that is not detected or suporessed.

The NRC approved the'R in July 19782 as an acceptable reference to be used in specific license applications.

LF: csc: rm/I03081'/9/83 The LTR assessment of'he PGCC capability to safely withstand the unli4~iy'event of an undetected fire in a cableway beneath the control room used a conservative design basis fire (OBF)which was agreed upon by tne HRC, QE, and an independent agency (University of California, Berkeley).

The OBF assumes a localized accumulation of trash in a normally covered cableway.The Hanford 2 plant fire protection program should prevent such a trash accumulation from ever occurring.

In conducting fire tests on the PGCC design, the University of Ca'iifornia found that the PGCC design inherently limits the availability of air in the cableways such that the design basis trash fire would quickly self-extinguish with only a small fraction (~15%)of the trash being burned.I Therefore, in order to get complete combustion of the'.rash for the design basis fire some tests were performed with a control room conf',gu-ration differing from the PGCC design, i.e., with the floor plates ac both ends of a cableway removed in order to allow combustion to zontinue.he PGCC OBF tests conducted by the University ot California met all of the acceptance criteria, e.g., containment of the fire to the fire duct and acceptable smoke production, for each of the three types of cable coverings that were tested.OESIGN OESCRIPTION In he fabrication of the PGCC floor area special consideration must be made for complex shapes and non-standard sizes resulting from construction tolerances iJi the concrete control s.ructure.

Nearly a11 (96K)of>he<F:csc: rm/rggOS'~

9/9/83 0 area in the PGCC is cove~ed with pre-determined s'zes of floor plates 1 fabricated of aluminum honeycomb with steel sheets on the top and bottom ard with aluminum channel closures on the peripheral edges.These assemblies cannot be readily modified in the field to fit special cases or close tolerance fit needs.tn order to accommodate the need for special site fit-up, General Electric Company recognized the need for an alternate floo~plate design which would provide for flexibility of installation wnile meeting the other floor plate requirements.

A material w tn the trade name"Richiite" was selected for this application.

Figures L and 2 show the Hanrord 2 floor plate layout and identify the limited use o'ichlite.

A fiber laminate 1-1/i6" thick RichlIte sheet was selected by General Electric engineers based upon its physical and chemical properties.

Additionally, it was decided to add a coating of intumescent paint (Albi i07A)on the cab!e tray side of the Richlite to r duce the Richlite exposure temperature in case of a 08F.Published data from Underwriters Laboratory testing gave credible evidence that the fire characteristics would improve~ith the Albi 107A coat'.ng.LF: csc: rr./L 0908 i'/9/83

SAFETY EVALUATION RICHLETE FLOOR PLATE TESl INC Verification fire tests were performed to the ASjH E-84 standard test requirements

.This test was selected since it, uses a similar configu-ration and has highe!temoerature and heat releases when compared to the OBF of the LTR.Figure 3 shows the heat release used in ASTN E-84 as compared to t;he Licensing Topical Report OBF.From Figure 3 it can be seen that the temperature near the specimen is much greater for a'longer period of time in the ASTN test as compared to the OBF test.:he ASTM tests utilized an intense flame originated in a furnace made or fire brick which was closed at the top with the Ricnlite sample.All of the energy released by the fire was subjected to the sample.Whereas in the actual floor section design, carbon steel constitutes three sides of the cable ways, thereby conducting much of the energy produced by the fire out of the'cableway duct.Two ASTM E-84 tests were conducted; one on the base Richlite material and the other on Richlite coated on the under side with the intumescent coating, Albi 107A.These tests were performed at the Underwriters Laboratory in Santa Clara, California.

LF: csc: rm/'.09081" 9/9/83 0

The results were:Material Index Value Uncoated Richl i te Flame spread Fuel contribution Smoke developed 40 Intumescent Coated Richlite Flame spread Fuel contrib'ution Smoke developed 80 FLAME SPREAD AHD FUEL COHTRIBUTION The test results show that the coated and uncoated Richlite flame spread and fuel contribution are clearly acceptable, e.g., one common criterion is to have flame spread and fuel contribution ratings not higher than 25;SMOKE PRODUCTION The important factor in the smoke developed is the contribution of this smoke to the total smoke developed during a design basis fire.The smoke index is based upon-smoke developed from red oak as being 100.This index of 100 represents approximately 15K attenuation o-the light source':."sc: rm/I09081" 9/i 83 4 1 t in the fire tunnel.The intumescent coated RichlitMtest showed BOX of the standard smoke generated from red oak, or 12%light attenuation.

The OBF tests of the LTR showed more smoke from the cable insulation than~ould be developed by the Richlite.The fire test results of the University of California tests, as reported in NEOO 10466, characterized the smoke generated from 1)Yulkene/Geoprene, 2)Raychem and 3)Tefzel cable insulation.

The test configuration required that the floor plates be left off the end of the cableways to allow ventilation to support the fire.In all three cases, the smoke generated was determined to be acceptable~

for the PGCC design.A curve, Figure 4, which shows the smoke generated in another Underwriters Laboratory test4 on Tefzel cable and cross-linked polyetl.ylene (Yulkene)V II as compared to the ASTM standard (red oak)and Richlite helps to quantify the smoke contribution from Richlite.As can be seen from Figure 4, the smoke characteristics of coated Richlite are superior to the already acceptable smoke characteristics of the cable insulation materials.

Uncoated Richlite produces even less smoke.Considering that only a small portion of the Richlite would be affected by the OBF (0.125 inch penetration in the uncoated ASTM E-84 test)and the other Richlite characteristics as determined by tests, the total contribution of RichlIte to the smoke from a postulated OBF would be insignificant.

.: csc: r!ne EO908" i/9/83 0

SNOKE CHARACTERIZATION Richlite is composed of 65Ãfiber laminate (cellulose) impregnated with 17.5X phenol.and 17.5Y.formaldehyde, thus for"..~ng a phenolic (C15H1(0~).

Any localized combustion of this compound as a result of the DSF would produce mainly CO<, H<0 and CO.The gases which evolve during the intumescent process were measured at the U.S.Testing Laboratory using mass spectroscopy as the gas analyzer.s The principal gases identified were: N<,'2, A, CO<and H2.Trace quantities (less than 0.05K concen-tration by volume)of methane, benzane and carbonyl sulfide were also identified.

These products of the Albi intumescence and any combustion of Richlite are a minor quantity in comparison to those produced by the cable insulation.

STRUCTURAL INTEGRITY The ASTM E-84 tests on the Richlite samples demonstrated that no warping or dimensional change occurr'ed.

The meehan',cal integrity remained well above the load requirements for the control room loor plates.It has also been demonstrated, by test, that the Richlite floor plates have sufficient mass and are positioned such that the fi re suppr essant, Ha1on, concentration is not changed.LF: csc: rm/I09081'/e/a3 CONCLUSION The PGCC Richlite floor plates used in the special fitting situations are considered as an acceptable alternate for aluminum/steel

'loor plates.This conclusion is valid even when no credit'is taken for the protective Albi 107A coating.

0 REFERENCES 1, H.R.Clay,"Power Generation Control Complex, Oesign Criteria and Safety Evaluation," General Electric Company, NEDQ"10466-A, February 1979 (NRC Approved Revision).

2.Olan O.Parr, NRC letter to Or, G.G.Sherwood, GE;Review of General Electric Topical Report NE00-10466," Power Generation Control Complex Oesign Criteria and Safety Evaluation

'uly 13, 1978.3.ASTM E-84"81a,"Standard Test Method for Surface Burning Characteris-

~'cs of.Building Materials," 1981.4.Flame Tests, A Report on Tests Conducted by Underwriter Laboratories Inc.at Northbrook, Illinois, September 27-28-29, 1976," E.I.duPont de Nemours Co, (E-12952).

5.William Raines,~lice President and Technical Oirector, Stan Chem Company, Albi Oivision;telecon with Charles M.Hart, General Electric Company.LF.'csc: rm/!Q908'/9i83 000000~o 0 000 00 a Ooolo 0 000 00 Ct 00ooo+0000~1~~~CLI~r I PlPOR SECTION pANEl CUP vtEW A-+F)Cute>

I I l 4 I I I'~Q NO C.~SLC.G UhlDER PLA I K 2" STRIP IN FROllT OF BEiICH BONDS=!GU R.': 2 CO~)'O'00t<LAYOU' l652 SOO i I I472 800 1292 700 TEMPERATURE NEAR SPECIMEN HEAT RELEASE;53,i)33 BIO I I I2 600 932 500 T EMPERATURE 752 400 Flic DUCT TEMPERATURE OESIGH 8ASIS FIRE, HEAT RELEASE: 10,588 GATI)572 BOO 392 ZOO 2 I 2 IOO IO Is T I MK-M INUTF S 20 FIGURE 3 TEST FIPE CClMPARISCIN l I.I Sl I@FACE 6 lPN TE M vQ Chal E Dl I(T

'~g~1 ,'~SMOKE PRODUCED (TUNNEL TEST)SMOKE INTENSITY FOR PURPOSE.OF CO45RISOV, THE RED mZ STANDANO IS SHCAVH.NOTE THA QFb MK TEST IS DISCONTIHUEE AFTER IO MINUTES.QAYCHEM FLAMTQOL (ESTIVWTED)

CROSS LIHKED POLYETHYLEHE (vuc~z~a)TEFZEL QFA.Q4K, RICHLITE IO T I}AE-M I NS 20 1 SMV'5 PROVIDED IN U.L.TUNNEL SURFACE BURN TEST FOR DIFFERENT MATERIAI S

~~*y Cg

~~)O Enclosure 4.GENERAL',;ELEGTRICNUCLEAR POWER GFNERAL;FLECTRIC COMPANY 175 CURTIIGFI AVE.SAN JOSE, CALIFORNIA 95125 MC 394, (400)925-3382 SYSTEMS DIVISION Responds to: N/A Hr.L.T.Harrold Assistant Director,'HNP-2 Engineering Washington Public Power Supply System P.O.Box 200 Mail Drop 9060 Richland, Washington 99352 Attention:

B.A.Holmberg Manager, WNP-2 Engineering

Dear Mr.Harrold;SUBJECT:

WG2808 WPPSS HANFORD 2 CONTRACT 2 RICHLITE FLOOR PLATES September 26, 1983 GEWP-2-83" RESPONSE REQUESTED BY: N/A Attached is a letter from C.W.Hart to me which provides the information relative to Richlite material requested by the NRC during the telecon with GE and the Supply System on September 15, 1983.if you feel you need additional information or have any questions, please call Pamela Kingston, Wayne Jones or me.Very truly yours, 48~'.A.MacLean Project Manager Hanford 2 F AM: rm/A09164*

Attachment cc: M.M.Allison H.H.Engelking J.A.Foirest J.J.Verderber J.M.Yatabe File No.H13.1.1 GENERAL', ELECTRICNUCLEAR FNERGY BUSINESS OPERATIONS GEl'/ERAL ELECTRlC COMPANY, 175 CUATNER AVE., SAN SOSE, CAl.lFORNlA 95125 408*925-2595 MC 429 CHH/83/071 September 27, 1983 F.A.MacLean SUSJECTl

Reference:

HANFORD 2 FLOOR PLATE Hanford 2 PLoor Plate Design Report by C,M.Hart dated September 1983 This is to certify the General Electric Company position in regard to Richlite used in the floor section of Hanford 2 PGCC.I.The density of Richlite is 0.0458 lb/in.The density of the>aluminum honeycomb-steel.

covered floor plates is 0.0246 lb/in 2.Underwriters Laboratory Test Results ASTM E84 dared August 2, 1983>August 18, 1983 and September 27, 1983 are attached.These results are for Richlite floor plates uncoated and coated,,..., with Albi 107A.These results arc also documented in General Electric Company Design Record File DRF H13-71 Index 9.3.The PGCC floor plates are primarily honeycomb (aluminum/steel) with a limited quantity of Richlite.The use of Richlite is limited to irregular floor plate shapes and non standard sizes.General Electric Company does not plan to add any more Richlite in the PGCC floor sections.4.The limited use of Richlite is the only variation in the Pire Protection System defined in Licensing Topical Report (LTR)NED0-10466.

5, Figure 2 of the referenced design report has been clar'fied to identify the location of the Richlite floor plates.Hart, anager Panel&Assy Design Engineering

/bn 136 NOTES W Richlite Strips ggjg>>P.i chl i te QQQ-Richl ate ilichl ite Strips Bench Boards 3" x 27'l4 lengths)Floor Plate Over Cables 0 Floor Plate-Ho Cables in Front.of F')CUR.E, Z CONTROL ROOM LAYOUT

~~fl)UNDERSRLTERS LABORATOHIZ INC.taaa%CONC Q Ya'hKTA CLA Ah.CALllOAhlh gNOO an independet<t, not-for-profit organization tes.'ing for pubtic safet y R9752 September 27, 1983 Cene r a 1 Electric Company'75 Curtner Avenue San Jose, CA 95123.Attention:

Subject:

Nr.Ray Mercado-Mail Code 429 Interim Report on Surface Burning Characteristics Tests Conducted on Coated 1.aminated Boards

Dear Mx.Mercado:

In accordance with your request, we are prov'ding information pertaining to the Surface Burning Cha acteristics data developed during the testing of the coated laminated boards, The test work is to lead to the establishment of Classification and Fol]ow-Up Serv'ce by the Laboratoxies, uzder our"Hardboard" category.METHOD The fire testing was conducted in accordance with Underwriters Laboratories

'Inc'."Tests for Suxface Burning Characteristics of Building Materials" (UL 723, ASTN E84,¹PA-225 and ASA-A2-5)as part of the testing planned fox our full scale investigation of the material in question.SAMPLES The samples evaluated consisted of a nom'nal 1 inch thick phenolic fiber laminated boards coated with a UL Labeled Albi 107A coating applied at 20-25 wet.film thickness.

Ea h test consisted of three 24 inch wide by 96 inch'ong c>ated boards butt-jointed end to end to provide a'ontinuous test service.Look For TheListing or Classification Mark On The Product~~~

R9752 September 27, 1983 Page 2 RESUi TS The follow'ng Sur face Burning Cha.acteristics Test results were developed, Test Sample Flame~See ad Fuel Contributed Smoke D~evalo ed Coated Laminated 12.8 Board 2.Coated Laminated 13-7 Board 87.0

74.1 CONCLUSION

The above data indicates that the followirg Surface Burn ng Characteristics will be appLicable for labeled material having a surface coating of UL Labeled Albi 107A.S U R F A C E B U R N I N G C H A R A C T E R T S T I C S Flame Spread Fuel'Contributed Smoke Developed 15 0 80 Lt is understood that, these results apply only to those products produced under the Underwriters Laboratories Follow-Up Service Program and that evidence of this is the appearance of the Follow-Up Service Symbol for Classifi-cation o j Hardboards along With the Surface Burning Characteristics on a label attached to the product.He are presently in the process of conducting additional tests to es"ablish specifications for use in t¹Follow-Up Service Program at the Rainier P'ywood Company plant in Tacoma, Hashing on.He will be in a position to prepare the Follow-Vp Se.vice Procedure and establish the classification in our published advices, as soon as this additional test work is completed.

R9752 September 27, 1983 Page 3 The Classification Barking of Underwriters Laboratories Inc.on the product is the only, method provided by Underwriters Laboratories Inc, to identify products which have been produced under its Classification and Follow-Up Service Program.H t ust that the information provided will satisfy your needs.However, if gvestions arise, please feel free e r to contact us at any time.Very truly yours, Engineering Associate Pire Protection Department HPH:Hd fmltr3 Reviewed by;7 ,~.'LEX.BRIBER Assistant Nanaging Engineer Fre Protection Department UNDER%'RITERS LABORATORIES INC.M4b K(7Tr bNO eA'F 74 CLaRA.r Al lP>H'll t e~~>an independent,not-fur-p~f otr t or'gani:ation tati>>q jar public'c sar'et!I Rg~52, 83SC7541 August 18,.~3 Nuclear Energy Group General Electric Company 175 Curtner Avenue San Jose, CA 95125 At tention: Hr.Ray I!ercado hail Cod-'29 Sub j ect: Tunne'est Results

Dear Ray:

1n response to your request the follow'ng observa=ions were recorded at the time of the Tunnel Tests on the laminated phenolic boards on July 27, 19S3r The one-inch thick boards exposed"o the standard test flame during the UE.723 test procedure were not damaged exten-sively by the flame.After the tests, there was no greax.distortion of the boards, in fact, very slight wa pace and dimensionally the board-did rot change..he area d'ectly exposed to the flame cha red to a depth of about 1/8 inch and further down the test samp'e t"..e charring diminished.

At about seven feet down the zest sa."..pie the char depth was 1/16 inch and from eight feet to twe've eet down the test sample was only on"he su face.Eeyond t..e tn'VQ oot ma}the e was only sur ace di scc'ra-i or-~I onL Fo The>.Listina or ClhSsiticotion Mark On The ProdUc't R9752, 83 SC75-".1 August 18, 1983 Page 2 he boards retained their hea" for rai te a Long period of time after the test but there~as no sub eauent flaming oz=smoldering, Should you halve further questions p'ease Let me jcnov.Very truly yours,.dc ANNEX BREBER Assistant Managing Engineer 2'i re Protection Departm nt AB:vv lcd

~~1 I, I

~~,i'5 III'~DZa~n EaS r~SORAVOaI s L~0UG>O)gg~ac~'g'~~+CLALA.rhLln)4'lA QIX40 an tndepevdent,7!ot-/or-prost organi at!on testis!ger public sf uetv 8'9 August 2, General Electric Company 175 Curtner Avenue'an Jose, CA 95123 Attention:

Mr.J.DeB1'ux

Subject:

Interim Report on Surface Burning Characteristics Tests Conducted.

on Uncoated Laminated Boards

Dear Yr.DeBlieux:

In accordance with your request , we are providing information pertaining to the Surface Burni..g Cha acter-istics data developed during the testing of the uncoated 1am'nated boards.The test work is to lead to the establishment o Classification and sollow-Up Service by the Laboratories, under our"Hardboard" category.METHOD..., II The fire testf ng was conducted in accordance with Underwriters Laboratories Inc."Tests for Surface Burning Characteristics of Building Materials" (UL 723, ASTM E84, i~fr PA-225 and ASA-A2-5)aS part Of the teSting plarued for cur full scale investigation o=the material i n que stion.SAMPLES The samples evat.uated consisted of a nominal 1 inch th'ck phenolic fiber laminated boards.Each test con=issued of three 24 inch wide by 96 inch long boards hutt-jointed end to end to p"ovide a continuous test service.Look For The.~4.Ustinrj or Classy(ication Mark On The Produ'c!

/

General Electric Company Augu t 2, 1983 Page 2 RESULTS The fo13.owing Surface Burning Characteristics Test results vere developed:

Test Sample Flame~Sread Fuel Contribu"ed Smoke l.Uncoated 24.9 Laminated Board 2.0 37.8'.Uncoated 25.0 laminated Board 3.0 39.0 CONGE USiON The above data indicates that the following Surface Burning Characteristics will be applicable for labeled material: SURFACE BURN<NG C H A R A C T E R I S T I C S Flame Spread Fuel Contributed Smoke Developed 25 0 40 is understood that these.esul-s a"ply orly to those"roducts produced under the Undezwrit=s Labora-tories Follow-Up Service Program ard that evidence Gf thi s'the appearance o the Follow-Up Serv".ce Sy-.bol for Classification of Hazdboa-ds along with the Surface Burning Characteristics on a label attached to the oroduct.

General Electric Company August 2, 3.9S3 Page 3 We are presently in the process of conducting additional tests to establ'sn speci.fications for use in the Follow-Up Service Program at the Rainier Plywood Comoany plant in Tacoma, Washington.

We will be i.n a position to prepare the Follow-Up Service Procedure and establish the'lassification in our published advice", as soon as this additional test work is completed.

1 The Classification Nazking of Unde.w"iters Laboratories Inc.on the product.is the only method provided by Underwriters Laboratories Enc.to identify products which have been produced under its Classification and Follow-Up Service Program.We trust that the information provided will satisfy your needs.However, if questions arise, please feel free to contact us at'any time.Very tru" y yours, Reviewed by: H.F, HANS EN Engineering Associate Fire Protection Department HFH:sd fmltr3 LLS A.BR IBEX Assistant llanaging Ergineer Fire Protection Department

~st'.~~g~g<$g&pk>~g.l5.~'.

f~'f'k'y j-.'n+p'@5+~+

4'~

1 FOR INTLRIOR APPI.ICATIONS TMI Panels Systems 440 and 550, with FOI)MICA<9 Iri'ninate dicoralive surfaces, arc in<allcd over any Iii<n ubwall.Thcsc rigid, prcvcncci(.d pancts-in convenient 16" and 24" modular widlhs-are available in a vildc range of solid colors and warinlonc woodgrain dcsfgns.These Panel Syslcms offer an allractive and practical alfcrnalivc al a limo when woods and other"natural" materials are in short supply and often cost.p<ohibilive.

Out most iinporlantty

-the alternative choices{TMI Panel Systems 440 and 550)equal or exceed in beauty, tunction and durability, the tradilional malcrials lhey icplare.Panel Systems 440 and 550 meet a wide scope ol Code it:quirc<ncnls

-fil a range of conslrucliop and design purposes.You can selecl narrow or wide panels, metal or malching decorative laminate moldings, usc splines concealed or wilh oplical reveals-and peri<'.clly malch iuns of standard panel (syslem 440)lo rncel adjacent areas ot fire.rated Class I specification by using Sys)cm 550-and save on cost while mainlaining the same visual<<ftpeaf.5 a phoo()cT n)sc((ir Tfotf Basic Ur,(>: Panel Sy Iein 440 and 2~50 aic m..'ln""'a<u<sys<(yn>s, du:ig<icd for usc as wall surlacing"'.eie goo<I nppca<,>i>cu, duirgbili>y rend rc'islance It;lain and nein<ching and heal Iro<n oidinary soure:cs aie icqniiiid.

Syslcm 440 Is a Ir<ndnrd grarie product ii>d<dc~I toi ncw or t*<nodefing rn aitalions iii (5(irk 5" ho<ncs, hospitals>

I>callh cate ccnlcrs.s.(i Chu<Ctiea, O<fiCC".<CStaurai>IS, inOICIS ar,<5~i.,e<>., b inks and publi-buildings.

Gompnsltton anil Materials:

P;incl Syst(:ml 44".'~,4 550>r(>assc<nllfed and inslalled u I;l.>components I I coml>o>><>nls: (foi I>7/16" panel)Decor<>tfv4 SuifacO:)<0 MICAbrand targ><3ib" thick fire<<>}aidan)~rticleboards

~et~as e iiyiij saute/rs Vlont type 2 glue;i550, U.L.slips)sell ill I rein liana, nil: pt33Gytcir<1 stars Backing skeet, Grace'.r\Ja~'O'~.l 5 I M~~~w'.c r a 5~a Sn r Qr 1975 TMI SYSTEMS DF.SIGN CORPORATIO, O<r;.I(ll'I!i(1', NORTH DAKOTA f>ant I glrc: 7716" ihtck Ii>r Dining>a sizes ot 16 x 96", 24 x g<t)", 16 r 120" aiid 24 x 110", Color(<anr Desfafni I I<ORMII.'./e 9 la Yi<nale sclid'clcrk ant WOO(li)re>I<i deSigrlrs fr in the Slandard Col<iirllon (sweets*<c>', File<I.I4 o).Flnl<kl><.Su<dc Sii,>>ri>jr>utiyly It<winked I xtuie, vrilh lcv, f<ghf<Cite(f<Ve lnlu<>.(5>5>>uy<S aiI n(.ft, arm pleaSing alf<nII Iha'I'ynh'kncs<S beauty (<Ies<4<I her ngaci'<<S<MOI<Ing: in I>>e fr>trn nf Ou Ide COrner t:ip, Celing<>nd (:n.ing ciititi(<<<ntip

<are all~~~Wi><>d COte nn<..i(rrfnced Witt<I Jftht<(.h i<el ammate IO<un<i.h II'!Pnr<O<S.S>aiiied W<rrc(f Sty<<<>C'i t OpliOna<Lsniiriate InSerl Slripa Co<rept(i><r II<fnCk ge, (SCO deal(>it<.n facing page).r lnstalla lon: Panels are applied vgi<h a spline s lem Io Ii>iilng<I(it>e over any subwiill.Iiislallalioni<)str lions hro pai.kriI'li>P3Ch Cei(IO<l Ot pan>:IS.(See

/ar<n page for 7 Ikasic sti>ps~'t Fire ltazs<d classification Rating: (5fp)AsTM 8~4.Flainn S liead-25.SmOI e doxy(<Oped

-<<0 nS le ed on>><e (<<icorative T~>ce.per ASrQ E84~nn liL Tunicl Test and covered under UL C':issiiication arg Foll>w-up Service Agreemcnt R.5035).Api.lic;.!>Io Slandaids:

Con>go>>cn>,<550 mri!I Itic Iollowi(sg pertorrrlnnce standard: Fcr Lain fna<cs-NGM A P<<bi>ration L DI-I&71 A<ncnded", lor Corns-Co<nrrsc<ci,><

Standards CS 236-66.pprovals and Co<nptlances:

(440)t.os Angeles (2351<);City ol ttew YO<k Calendar><28I363SM: City and County ol San Francisco-Gericral Approval 226 F 9,0 I<><eric<Finish Class>8.5!assifica>ion Pile.H5035;B.C.C.A.-su33;I.C.B.C ki italians: p.t el"yst'tns aso nc ssg a e nc, slrucluial ai.d c lnnul be 5<5<id oil (gti.iio>apptica<iors.

~mo>%)s ro Ir I~(1~1 FQAMGLAS'ellll

'(4 I ullar Glass llns r I g 1 ullatlon't*I:'r$I FOAMCiLAS insulation is a rigid in-s(ilauon nia',erial coinposed entirely of (..losed glass cells.It is mani!-factured in n basic block form wi)ich is facto!1/-c(ft and fabrica!ed into blocks nnd boards of varic!is thick-nesses;!()d(taper."..I The glass cells vihich (nake up FGAfvlGLAS insulation are contifi-uous vvith i)o (f)tervening voids...FOAMGLAS ins()iation is all cellular glass.it contains no binders or fillers.lt t)as perlonned successfully for over forty y>.'ifs ii)a wide variety of non-resident(;!I

))(.(if(ling applic;itions.

Tl.e veisntility of FOAMGLAS ins(.)ln-tion pe(mits it to nieet todnys most Cnrnandinq!

>uiidi(ag insulation f)c~<f;.Built.Up or Single-Pfy.

FOA)VIGLAS if sulaiion'5 (otal resistance to n ois-tiire in liquir: or vapor form,<liinen-s(of)at stat)llity 2!nd high co(r)pfs.."a ve strength fr;nkr!it nn ideal iiisulation under built-up roof or single-ply membranes.

improved"R" Fnclor.Since its intro-d!iction, FOAl'/IGLAS insul:ilion lies been cof)lfni(RHy evak)ated thioiirg)3 on-the-job perform;nice.

Ti!rs inojst recent proof!ict d(!veioprnen!

hns)ed to an"R" fac!or ir:Ip(oveni(,nt of 2.Ki p r inch (7.".'Ic fnen()f.Cost Versus Vniuu.Wh(fe FOAM(GLAS if)sulation is i:(;f)s(do(ed to initf;illy l)e n (nore expo()sive ii)sul')tiilg (naieria)than sonic ot))(!r."

it is quite f.:om-, petitive v/i3(!():t is eval()ated on n lif(>>.cycle basis The superior physical prcperties of FOAf).fGLAS i:isiilnlion arr vihat make it cost effective.

I I o$i The Outstanding Physical Characteristics FOAMGLAS~a Cellular Glass Insulation PHYSICAL PROP EBTIES USA i'ETBIC~)I ASf TEST'tll)t, C24J l I I)fp(ive,0nly moisture retained I face re!)5 alter imineision.

Absofp!(onoimoi:!urn; 62"I)!On ab5 I"'.hy volume)'diieiiiig to su (//a(ef.vapor pefmeab(l~((Acid i es(5!ance Combusiibiiiiy Ccl)laos(t(oil Noncombusiibie.

wi))not bufn Puir!qI<i55.tfiiaiiy ii:nfganic.

co(;tams no binde...i;E!X Cofnpfessive stfeng!h)¹psi 7.0kg/cfnr

'689,kPa=t C!65 Average..for'Strong!h vihen surfaces cappe(iviith hot asp)ja!L;I C 24)standard maiefiai" dilfe(ent cappings 9 ill'()(ve diff kent value;.Def)sity avera(;e 8.5lb/It)136!0/in'imensional stability Exceilen(-doesnotshnnk.sviellor via(0 36kg/O'" ,C 303 Linea(coefhcieni ui ihef<nai exp3neioii l 4,6x)O'O'F 83x)0'o/'C II3x10'/K D(u n)/lu iiF kcalf(nd("C The!0)3)Conduciivity O.Zl<>50'F

'0'(7!" O'0.35to>i j F" C.OQc'20'C Certifications I///n)K"C;, C;ii.I)0!71<O'C'58 llQR~.~201 C I Underwriters Laboratorfes e Fire Hazard Classification Flame Spread...G Smoke Developed...0 is (oiafn I U(, fhnirriori hfhlcrlalis orrccrcry!

~Fire Resistance Classification" fhoof Ceiliiig Asseniblies Desigii Nos.P?27.P500-1'/7 Hour DesiCJii No.P801-1 to 2 I-lour/see/an'IS((7(.Frre lfi Sial 1r>C((Oituctoryi e Acceptable for use in Conutniction No.1 and No.2 ((Metal Deck Asseinblies, f.ire Classified) ai!d No.171 (Mete)l Deck Assembly, Wind Uplift Class 30, Fire Clnssi-fir d).f,"uo (1(os((lc Dura(mph(a(urn(ls oirccfory)

Factory Mutual~Steel Deck Canstr<ictio()

Acceptn6!e for Class 1 Construc-tion)~Exposed tntqrioc fnsufntidn Autom!!tic Spr)nkter Protection May Not Be Required: The insula-tion of i!selt has been evaluated for its surface liame spread and fuel.Contribution.

The insulation and its recommended installation inethod rneI.!!the requirements as Class 1;material of Io)v tire hazard in noncombustible occu-panc(es not requiring automatic spnfikler protection when applied to noncombustibte substrates.(SCO/a(CS(Fhi Afrprovhi Ouir!e)Conductance and Resistance of'OAMGLAS':

Insulation r 1 0.00pefni (i).000peim cm g C355~Iincer viou" to'conn)nn ac(ds and (hei(iu(nesoxcepihgufoiiuofic acd Capilianiy None: No(le,~Nofle n(4 Thlcklless of FOA)hGI.AS.

Insulation 75" F (24oc)50'F!10'c)I (inches)2'Vaiue ,350.233 175 1(0 R Value 5.71 7.14 ,C Value 330 i."20 ('65.132 R Va(ue 303 4 55 fi 06 758 34 4 1lr'I$7 I ou 078 07i)8.57 10 00'I 1.43 12 86 I~110 9 09.094 10 61 Io,12.073 t'13 64.,O66 t 15)5 I r (~II 1 psl

-"~mz in~~".g ,c" r Reduced Insurance Rates.Tile I)of)-cot>>bustitiility of FOAfilGLAS in-s(ll ltlon tAiIkcs re(fuced it)su(a(1ce rates and deductibles possible.It can afscr help lo reduce thi(!r('.0<.lire-ments for ()xpensive tire protection eq(!if)>>)e>>I.,Pittsburgh Corning rec-omme>>ds that local fire cit)d buildino codes bc c))eckcxJ c:.irr:fully to deter-'ine coiiditioi!

an(J restrictions on buildin(i>>)ate)ials sell etio 1.Aa>~=".-@~A~~I~,, 1<<4<<<44)'J<iiI','I j,<I<,I-irc S'.;J'i,ii ti;I C:<<,'-,.S'<.:<l Vrittt F()i<r<<<<!AS!AS ilii!i(i: t.l~~><<<in<<i'I:/

i~~(<I l<,<titlt<~,tt I~I ivii I<I"'six<).I(<v!Itc Crt<)rii;!'I ()"",l ii"",i.tf ii I i,!iii'ii ilcc Vermine and Rodent PI'oof FOAfvlGLAS insulation is ii;organic, so it (vill not t)rovirde fod.nts an(l veri>>itl)vith nesting material oi'ood.Olh(r ii)S(tlaliii(J>>)ate<();)IS.

S(irh as pl)ISti(lO(llils CAA bn (Jeslr')yed Oy rodents nii(f (!xpcnsiv~viire tAesh (<i!PI'0('fit)g

)>>list t)e)Astcliled to f)to tect thos<".m:)teri'I!S.FOAI)IGLAS.

unlike o(her in'ul:1<ioAs.

witt not attr:lct or, ust;Itn b;ict<<tid or i))old (J I c.)w lb.FfP%37~~I4,))l Y r)'<.I g)'tgf-" 1 FQA,'<lC<l r)i iit."iiOt) iS Ctl (I(')s:,.)it{i to!<ally itnitco))hi I!<!<;.'c.

In ti)IC!<i,.)(tl I>>>.;ii ja l SS<!A:i)l tO f)rr)V<C,<<

isli<I<'<i)lais)t)

'luhl(h V."Il stand I,'I)<ci,'};.{;u<,'i<<ii ki;)cls, fc)c.,'t<!Iic;<I<I<i<1 II)(i I<)<<<'.I Ill;L~r"!p<'.rk<t<c;(I,'.k.Ati<l li(lr)is v.'t)i"*,'Il

')i tr<s<!Ic/<!<<<t')t(<i<it>>i)

It"I'<1~t)i>>I)c's car, (.'Oi!I.f li.'<tn)(lf)6;i)s{ilatioi) t)i!tie{lit'I i';:I!Ilia'ss<,"..'ti!it<i) tli!Iti('.

IOA(!S The Bi i)(fits.f'hc!r<.w tt'.)i.Ao sli)i';"ii)g cf colilpri.', If)(l o ti'c it)s(il<.tion<

vt))i'A I QAIVIGLtLS ltlsuki-t)OII IS Sf'Ci!IQ{'I atl(l lriripei iy l)1-static.'d, ItI)as the str(:ngth to wi;l)si;)t)(l;i!I I)Or!Anl IOa(IS fnr tt'.()<<rc{)I;h>>llf)pile at>>l1.NOI..COmbuStible Glass cannol burn, so FOAMiGLAS i!is(!Iatioi)is totally i)on{)onlb(istibl(.'.

Ouri>>g storage and coi)struction, insi)l;)(ion IT)atc)iials A!o often ex-posed to sp<!iks arId opetl ffiiines.Many ins)ilation nlaterials are com-buslible aiid A)ust t)e shir)l(ied at all titties.FOAMGLAS insulation does Aoi fcquir(.this costly prot(.etio!).

I)isolation shoiild not b(ir>>nor help to spread fta>>1es if a fire o(;curs in a co!nplett)d structure.

Ii)actual cases.FOAMGLAS.Ai)inorganic ii)sulatioi!

~has cont<iined the fire.pii:.ue;itiiig it frotn spreading.

Oiganic i>>s()tations.

iti cantrast,<sre not only flat)it)lat)IQ.

but oft(!i)help to spr(.'td f!ames and genera(e toxic fut>>es and smoke which compo)ind fice fighting And rescue problems.'<<<ST<'.I I:-I IC,""(!<!<<iv<)<n<I II Stn<als<n<<Vv<iiv<i (i<i<i I'<<<<<'<<'n l<<<'<<0'<)4" FA<i{P 1GLAS<i)scil;Itin<)

"rOvtclcs tt)ax)-i:It)it)ilt;i",(core io t(;clot>ls

<and vein)i;I The 8enefits.Whet)FOAMGLAS in;",ilation is spc!citied and proper!y i rstallcd in aoplicatio>>s such as food and dairy products processing plants.brevieries, rolcl storage ware-ho ses.etc., total resistance to rodents an(l vermin is assured.state ai)d federal health requirements care')tisfied and lossr)s due to infesta-tion are eliminaterJ or dram;)tical!y i<<duccd.

a, S I 4 ATTACHMENT Concern 2.To comply with Section C.6.a of BTP CMEB 9.5.1, the fire detection systems should be designed in accordance with NFPA Standard No.72E.Section 4-3 of the standard stipulates that the location of smoke detectors must anticipate smoke stratification.

We observed that in several fire areas, such as RC III adjoining the control and cable spreading rooms, smoke detectors were not installed at each floor elevation.

The applicant should verify that the location of smoke detectors in areas that encompass more than one floor elevation will not be compromised by smoke stratification.

Additional Clarification Control room procedures require that fire brigade members identify which detector goes into alarm and log it in.In areas where the smoke detectors encompass more than one floor, such as the=cable chase, the detectors alarm per floor in the control room.Therefore, the fire brigade member will know which floor to respond to and can quickly locate the detector which alarmed and find the cause.Concern 8.To comply with Section C.5.a of BTP CMEB 9.5.1, interior wall and structural components and thermal insulation material should be noncombustible.

The applicant should verify that the laminated wall panels in the control room and the urea-formaldehyde foam insulation observed in several areas of the plant are noncombustible.

Additional Clarification Attached are copies'f the catalog cut for the laminated wall panels in the control room and the Foamglas Cellular Glass Insulation used in the MCC rooms (Enclosure 1).Per the fire test data, both materials are satisfactory for the area and use.Automatic sprinklers have been installed in the control room peripheral areas which will provide the needed protection for this area.Concern 12.In Amendment 19, the applicant committed to comply with NFPA Standard No.30,"Flammable and Combustible Liquids Code".In the water filtration building, the diesel fuel day tank is on unstable supports which is contrary to this commitment.

Additional Clarification The supports for the diesel fuel day tank in the water filtration building will be modified to correct this situation prior to fuel load.

Concern 13.In Amendment 19, the applicant coranitted to comply with NFPA Standard No.15 in the design of the water spray fire suppression systems.We observed that the water spray heads and baffle plates protecting the diesel generators and fuel tanks do not provide complete pro-tection (coverage) for the hazards in the area which is not consistent with that commitment.

Additional Clarification Sprinkler System in the Diesel Generator and HPCS Rooms.The rooms have ionization detection in the switchgear room and photoelectric smoke detectors, at the ceiling in the diesel generator areas for early warning of a fire.These detectors will send an alarm to the control room and the fire brigade will be dispatched.

Thermal detectors are located at the ceiling to activate the pre-action sprinkler system.The pre-action sprinkler heads are located 14 feet above the floor with the detectors at the ceiling, approximately 30 feet above the floor.Enclosure 2 provides additional information on the sprinkler head and sprinkler head configuration.

The pre-action sprinkler system js designed as a special hazard system with closed head directional spray sprinklers over the top and sides of the diesel generator.

The attached sketch shows that all areas of the diesel generator are covered except for some floor areas next to the wall.Additions are being made to the system to cover these areas.The glass bulb sprinkler heads have a 12" X 12" heat collector over the head to help collect heat and actuate the sprinkler heads.There are no interconnecting passages or penetrations between the diesel generator rooms.Therefore, the fire brigade actions to locate and/or fight a fire will not expose the redundant generator to fire, heat or smoke.The redundant diesel generator has the capacity to operate the safety related equipment and bring the plant to safe shutdown.In the event a diesel generator is inoperable, plant operation is conducted per the technical specification limiting condition for oper ation (3.8.1.1), effectively verifying redundant diesel generation within one hour of the diesel generator inoperable declaration.

Concern 17.To comply with Section C.7.h of BTP CMEB 9.5.1, fire doors and other penetrations of fire barriers in the Turbine Building should be located so as not to be directly exposed to a turbine oil fire exposure or spill.We observed that there was no curb or floor drain at th'e fire'doors at elevation 441 ft.of the Turbine Gene).ator Building to prevent the possibility of oil leaking from a turbine oil line rupture from passing through the fire doors.

t4I L, 4$V Additional Clarification Per audit item 17, a recommendation was made to install a curb in the doorway between the Turbine Generator Building and the Reactor Building at door T-114.The purpose of the curb was to prevent an oil spill associated with operations in the Turbine Generator Building from entering the corridor.Because of the general traffic through this door, a curb or ramp would introduce a significant industrial safety hazard and, in the Supply System's opinion, increase the probability of significant injury to employees.

On this basis, the door curb or ramp proposed in item 17 will be omitted.The following protection/system design is considered an acceptable substitute o The maximum spill for the area considering a single failure would be 700 gallons.o The Reactor Feedwater Pump Oil Systems are not a major oil system associated with the station turbine generator.

They are associated with the turbine driven Reactor Feedwater Pump Systems.o The piping in the area is schedule 80 welded construction external to the local sample panels.Small diameter stain-less steel tubing is used in the panels.o Design oil temperature and pressure in the piping systems is 160 F at 120 psig.o The floor is sloped away from door T-114 toward 5 floor drains in the Turbine Generator Building.o A hydraulically designed open head deluge system protects the piping and sample panel area.The system is actuated by thermal detectors or a manual pull station.The system is identified as DEL-S2 in the FSAR.The Supply System is in the process of extending the HNP-2 standpipe system to add an additional hose station adjacent to door T-114 in the Reactor Building corridor.This will allow safe access to fire fighting equipment, in the event it is needed, to protect the opening between the Reactor Building and the Turbine Generator Building.PGCC Floor lates An additional concern recognized by the Supply System is the use of substitute materials used as floor plates in the PGCC.In response to a phone conversa-tion (September 15, 1983j between D.Kubicki (NRC)and Messrs.P.Powell, A.Jones, M.Neel (SS)and C.Hart, R.Artigas, F.MacLean and L.Fleischer (GE), the following information and comnitments are provided:

'I,~

Benelex floor plates currently used on the PGCC modules are being replaced with Richlite floor plates.Richlite has a flame spread of 25 and a smoke evolved of 40.which classifies the material as Class A or noncombustible.

The General Electric report entitled"Floor Plate Design", dated September 1983, is attached for your information (Enclosure 2).Additional informa-tion forwarding Underwriters Laboratory Test Reports conducted for General Electric are provided in Enclosure 4.One section of PGCC U-689 uses an insulation"washer" made from Benelex between the PGCC floor frame and the termination cabinet.The termination cabinet and the floor frame enclose approximately 80Ãof the surface area.The majority of the exposed area will be given two coats of Albi 107A, a flame retardant coating.All other Benelex will be removed from the PGCC.A question was raised concerning the possibility of floor plates being dislodged during the discharge of Halon in the underfloor sections of the PGCC's.Ten discharge tests have been conducted to date.The tests have been witnessed by Supply System Industrial Safety&Fire Protection personnel and no floor plates were dislodged in any of the tests.The floor plates are designed to have a flush 1/2 turn locking device installed after Halon discharge and concentration tests are completed.

I 3/4 fnc1osure 2 II l TEMPERATURE RATING.i I55F/STCAM

>7/ie SIZE STAMPED ON REVERSE SIDE OF 3/y 6/4 ORIFICE NOZZLES.TEMPERATURE RATING: 50 I354F/57'C I 75'F/79'C 250 F/I2I C 286'F/I II 4C 325oF/l65'C 360 F/I82OC 400 F/204 C 500'F/260'C, 40 CQ INCLUDED ANGLE OF DISCHARGE PATTERN: 65"'0 l2 5'@0 95 I I Oo I60'804 a: 30 ORIFICE SIZE:/2,/e OR./4.PIPE THREAD CONNECTION.

I/2 NPT 20 MATERIAL S FINISH: BRASS-PLAIN, LEAD COATED OR CHROME PLATED.FURNISHED WITH PLAIN FINISH UNLESS OTHERWISE SPECIFIED.

IO IO 30" 20 G.P.M;II DISCHARGE.'URVE a TO ORDER SPECIFY: GEM (ORIFICE SIZE)(TEMPERATURE RATING)(SPECIFY FINISH IF OTHER THAN PLAIN BRASS IS REQUIRED)PROTECTOSPRAY NOZZLE, TYPE EA" I (ANGLE OF DISCHARGE PATTERN).(QUANTITY)

GEM (N-4l)Pt llttltt It)U.S.A,<<1t 6-I-73 g N 7-'~5'-'3"]~yg AXED Gt=ND2-2/E Pf~DiEC778AZ.

CA~~0<ATlBAlS

/.l 8 77K DF PZO)'C 7PD Sl.)ZFACZ X'P].7)JE'FP JVG"..RES-.Q=(C-8 fl C>~l ie.B.-D)X a-'~l"=/tD, SC,Q maC-W~~l--(Z'='-'O-.l'-~e-'O)X a'-e'l~S',C 8 C 0 l z.i c ru er-rz"-..:~;r=u w PZFAQE Y!JJ)'2:'r rV DF 5 UZ.~AC>Z..AEZcV=<-D~~=0 Y 7-':" f wb.-t'44 7YP/CAL&f2~5 eazS A Ll,;,-/),';-z hZ)!75 DA Y MA A.l/C/I/e CJ'f/bl pV/G=M QZATS: ez-~r-~l@AC L=PA 4ZES 8 Al'='.-.-.LE PM7/c..T'BA/

CA LCM A77Ml&I AREA DF Fat/8 DS 7AAlk77 D-93/88 2~AZ~DF SVfPDZ7 4>DSZ<CIMZ PPl.Y)=Z'-Z" X Z-'za"=B$14 z nrxx or-ex~)z mxsc rely.rz-os zc,ru=m~s=zu.>>zw.-4.AREA i EDFEQTi."9 BP AfD2>E 4-9'Z'KD,4.2+4 x 4-'8"=Z>.EZ a,szsz"'"--ez.=z3 sz 4+v~sashes Bsv~)=zs,7r, d.PCS)">>~St"=Eh/Dig QIID~kx ZO,/Z)w Szuuze (5,Ã)=BI.88~7(%~A"'."C-'P/'Eplp-.'<(/-/u>/~~Et cz-.'Z).NAF4 Z-fZM7EIASYST~

N---'=='.-'>~s'~WW~A770 EFRMl&l erC'P a.