ML20037C780

Summary of 810202 Meeting W/Util in Bethesda,Md Re Hydrogen Control Measures for Facility.List of Attendees & Draft of Section 6.0 to Analysis of Hydrogen Control Measures Encl

ML20037C780
Person / Time
Site:	McGuire, Mcguire
Issue date:	02/03/1981
From:	Birkel R Office of Nuclear Reactor Regulation
To:	Office of Nuclear Reactor Regulation
References
NUDOCS 8102200843
Download: ML20037C780 (33)

Text

• **

pn nouq(o, 4

UNITED STATES eg i

NUCLEAR REGULATORY COMMISSION 3

E WASHINGTON, D. C. 20565 o

/

FEB 3 1981 Docket.los.: 50-369 and 50-370 APPLICANT:

Duke Power Company FACILITY:

McGuire Nuclear Station, Units 1 and 2

SUMMARY

OF MEETING HELD ON FEBRUARf 2, 1981 A meeting was held with the applicant on February 2,1981 in Bethesda, Maryland, to discuss hydrogen control measures for the McGuiro Nuclear Station. A list of attendees is shown in Enclosure No. 1.

The applicant presented a draf t of a proposed section 6.0 to its report "An Analysis of Hydrogen Control Measures at McGuire Nuclear Station".

This section considers the effect on equipment inside containment subsequent to a hydrogen combu, tion. The draf t of section 6.0 is,' resented in Enclosure No. 2.

The staff discussed and connented on the draft.

The applicant indicated that a final section 6.0 would be prepared and filed as a ptrt of the McGuire application.

Ralph A. Birkel, Project Manager Licensing Branch No. 1 Division of Licensing Encic sures:

(1) Attendees List

~

(2) Dre ft Section 6.0 McG. tire Mydrogen Control Measures Report cc:

See next page 0M 81 0220 r%

l Mr. William O. Parker, Jr.

Vice President, Steam Preduction Duke Power Company l

P. O. Box 2178 422 South Church Street Char 1otte, North Carolina 28242 cc: Mr. W. L. Porter Dr. Cadei. H. Hand, Jr., Director Duke Power Company Bodega Marine Lab of California P. O. Box 2178 P. Q, Box 247 422 South Church Street Bodega Bay, California 94923 Charlotte, North Carolina 28242 Richard P. Wilson, Esq.

l Mr. R. S. Howard Assistant Attorney General Power Systems Division State of South Carolina Westinghouse Electric Corporation 2600 Bull Street P. O. Box 355 Columbia, South Carolina 29201 Pittsburgh, Pennsylvania 15230 Office of Intergovernmental Relations Mr. E. J. Keith 116 West Jones Street EDS Nuclear Incorporated Raleigh, North Carolina 27603 220 Montgomery Street San Francisco, California 94104 County Manager of Mecklenburg County 720 East Fourth Street Mr. J. E. Houghtaling Charlotte, North Carolina 28202 NUS Corporation 2536 Countryside Boulevard U. S. Environmental Protection Agency i

Clearwater, Florida 33515 ATTN:

EIS Coordinator l

Region IV Office Mr. Jesse L. Riley, President 345 Courtland Street, N. W.

i The Carolina Environmental Study Group Atlanta, Georgia 30303 854 Henley Place i

l Charlotte, North Carolina 28207 Mr. Tom Donat Resident Inspector McGuire NPS i

J. Michael McGarry, III, Esq.

c/o USNRC Debevoise & Liberman Post Office Box 216 1200 Seventeenth Street, N. W.

Cornelius, North Carolina 28031 Washington, D. C.

20036 Shelly Blum, Esquire Rober t M. Lazo, Esq., Chairman 1402 Vickers Avenue Atomic Safety and Licensing Board Durham, North Carolina 27707 U. S. Nuclear Regulatory Comission Washington, D. C.

20555 Dr. Emmeth A. Luebke Atomic Safety and Licensing Board l

U. S. Nuclear Regulatory Commission l

Washington, D. C.

20555

Enclosure No. 1 Attendance List McGuire Nuclear Station, Units Nos.1 & 2 Feb. 2, 1981 Duke Power Co.

W. H. Rasin R. B. Priory N. G. Awadalla Tom Heitman l

Neal Rutherford l

NRC Staff R. Birkel P. S. Check T. P. Speis l

E. Ketchen J. Scinto L. Chandler l

P. T. Kuo J. W. Shaapaker C. G. Tinkler W. R. Butler C. P. Tan J. F. Meyer L. S. Rubenstein H. Polk R. Tedesco J. C. Pulsipher G. R. Mazetis H. C. Garg V. Benaroya Z. R. Rosztoczy P. R. Matthews i

K. I. Parczewski

)

ENCLOSURE 2.

t DRAFT FOR INFORM.2 TON 5.1 Introduction ANOlOR REvig,y -.,,,

The accfdent at TMI.2 demonstrated that hydrogen can be generated in amounts greater than required to be considered for design purposes by ICCFR 50.44 and that hydrogen ccmcustion can occur inside containment. The major concern over hydrogen combustion inside a nuclear reactor containment is that the resultant pressures may cause a breach of containment with subsequent release of radio-activity to the environment. The structural analysis in Chapter 4 demonstrates that the McGuire containment can withstand the pressures generated by a TMI-type (See Chapter 2 for an analysis of temperatures and pressures resulting acctdent.

from the burning of hycrogen generated by a TMI-tyce. accident at McGuire.)

A secondary c:ncern over hydrogen c:mbustion inside containment is the effect the resultant temceratures may have on vital equipment located inside contairment.

Results from the Fenwal t'ests indicate that the thermal effects on vital equip-ment ex:osed to hydrogen burns are not overly severe. Cuke Pewer Comcany has also conducted an analysis of the environmental effects of hydrogen burns on i

vttal equi: ment. This eget: ment evalutation is provided in this chapter.

5.2 Necessary E:uf ement A review was ;erformed to determine the equipnent !ccated in Cantainment that c:uld be requirec to function after a hydrogen burn resulting fr m a small break LOCA. This equipnent is listed in Table 1.

All supcort scui: ment necessary for the listed ::mponents to function, i.e. cable, function boxes, et:., are

nsidered, for the ;ur:ose of this analysis, to be 1 part of the listed ccm-The location anc function of the ::mponents listed in Table ' are ponent.

briefly described below.

rans:

Canta tnment M r tetur9 Two 20,0C0 :fm fans are ;rovidec :: assure the rapid return :f air :: :ne

'cwer : meartnent felicwing tne initia: bi:wcewr resul ting ' 'r:m a '.:CA.

This

~.

1 function not only tends to dilute any hydrogen that may be generated, but also allows hydrogen combustion to occur in the lower comcartnent by supplying a source of oxygen. Burning hydrogen in the lower compartnent results in reduced pressures because the ice condenser is available as a heat sink. The air return fans are located in the upper compartment and discharge into the lower compartment.

A more detailed description of the air return fans fs provided in Section 6.6 of the McGuire FSAR.

Hydecqen Skimmer Fans:

Two 3,C00 cfm fans are provided to prevent the accumulation of hydrogen in restrteted areas of the 1cwer compartment by continually drawing air out of :hese areas and discharging it to the upper ccmcartment. The hydrogen skimmer fans are located in the upper ::mpartment. A more detailed description is provided in Section 5.6 of the McGuire F5AR.

Hydrocen !:nfters:

See Chapter 3 of this report for a description and 1 cation of the hydrogen igniters.

Steam Generator Water Level Transmitters:

The steam generator water level transmitters provide indication of the availa-blitty of the primary heat sink. Level indication is provided for each steam i

The level transmitters for each steam generator are located in the generator.

icwer : mcartment in a separata accumulator recm, that is the steam generator

• A* transmitters are located in accumulator recm A* ; "S" in rocm "3",

etc.

A more detailed discussion is provided in Chapter 7 of tne McGuire FIAR.

3ressurt:er Water Lavel Transmitters:

Indication of the water inventory within the reactor coolant system (RC3) is 3rovided by the ;ressurf:er =ater level transmitters in ::njuncti:n with RC5 6-2

,[

\\

These transmitters are located in the lower ccmpartment in two pressure.

separate accumulator recms. A more detailed discussion is provided in Chapter 7 of the McGutre FSAR.

RCS Tencerature and Pressure:

The core exit thermccouples, the RCS resistence tempers' are detectors (RTD's),

and the RCS pressure transattters provide information for determining the degree of sub-cooling in the :tCS. Cables for the core exit thermoccuples are routed through the lower containment to a junction box located in the instrumentation room and then to the electrical penetration. The RTD's are routed through the 1:wer c:meartment directly from their loop loca: ion to their electrical ;enetra-tions. The RCS wide range pressure transmitters are located in tne annulus and are therefore not affected by any hydrogen' burns that may Oc:ur inside con-tai rment.

1 Conta f nment '4ater Level Transmitters:

i The containment water level transmitters provide indication of the water level i

in the energency recirculatton sump. The upper range is equivalent to approxi-l mately 10 gallons of water in containment. These transmitters are located in 5

the annulus so they are not affected by any hydrogen burns that may occur inside ccntaimnent.

5.3 Ect.mnent Thermal testonse The ccmcuter ::de CLASIX was used to investigate the res:onse of the McGuire :en-tatnment o hydrogen burn transients.

(See Section 2.3)

Inout parameters include nass and energy releases frem the postulated break, containment ::ncitions prior to nycrogen generation, containment gecmetry, and varicus c:ntaimnent related system ;arameters.

lA detailed discussion f the McGuire OLAS*X analysis is resented in Iection 2.2.)

The results of the base case 3rs sncwn in Figures '.-3.

DRAFT 5-2 FCR \\NFC R..V ~1C':e ANC/CR RE'.1EW @,i v

i l

in the 1:wer ::m-These figures show a series of ten hydrogen burns occur ing partment with one burn occurring in the ice condenser. The analytical models i

I cresently in CLASIX are very conservative in that all heat sinks with the Preliminary exceptions of contairment spray and the ice condenser are neglected.

CLASIX results indicate that in the event hydrogen burns do occur, air tempera-tures inside containment could exceed levels presently considered in the design basis accident (08A) analysis. In view of this, Duke Pcwer Company has performed an analysis on the thermal response of the ccmcenents discussed in the previcus csed to the thermal conditions resulting frem hycrogen burns.

section when ear Compared to the temperature profiles shown in Figures 11 the pressure transients l

shewn in Figures 5-8 are very mild. TDe effects of these ;ressure transients on the c:mponents of interest are discussed in tr.a nex: section. Conservative l

analytical models have been developed to investigate the thermal response of In addition to the internal con-containnent equi; ment to hydrogen burns.

servatisms, these analytical medels are dependent u;cn an input temperature 4

CLASIX is profile wnich is presently obtained frcm CLASIX calculations.

presently being modtfied to provide more realistic results of hydrogen burning I

scenarios. Mcwever, until results are obtained frem a modified CLASIX, :ne tamcerature responses calculated.using our analytical model are not considered realtstic. These results are supplied here for information only. The assults of :he heat risk calculations presented are considered to be uocer bcund values for actual equipment res:ense.

In order to effectively evaluate equipment res:onse to a hign temperature i

1 All inree metnods gas the tencerature rise of the ccmconent must be known.

transfer (:cnduction, convection, and radiation) were considered in the of nea calculation of equi: ment temcerature and are discussec :elcw.

<i 5s

. 7

.f',0.i a.% ;;C R W :.;..

+:

e w e y

Unru i FOR INFCRMAT: ors AND/OR REVIEW CNLY Conduction:

i i

Two very simplifying assumptions were made: 1) the component consists of a solid steel cylinder and 2) temperature thrcughout the cylinder is uniform.

These two assumptions allow conduction to be neglected in this analysis.

Neglecting conduction simplifies the heat transfer calculation and adds some cons ervati sm. The geometry of the equtpment of interest is easily modated as a cyltnder.

f Convection:

An empirical formula was used to obtain the c nvective heat transfer coefficient for a hort: ental cylinder in air. Since the effect of the location of the listed ccm:enents with regard to the amount of cooling previded by the air return fan discharge is difficult to determine, no forced air c cling was c:nsidered i

Only natural convective heat transfer was c:nsidered. This analysis present.

conservattsm results in an overestimate of the peak tenperature.

Radiation:

There are two sources of radiative heat transfer to be considered: 1) the flame front and 2) the residuai atmosphere-combustion product mixture after the flame Since the flame temperature is kncwn to be dependent on initisi i

front has passed.

hycr gen concentration, the flame tem erature is cotainac frem Figure 3.

Assuming no burning occurs untti 10 percent hydrogen is obtained results in a flame tempera-

~ perimental data shows that partial c:meustion will Occur belcw ture of 13700F.

x 10 percent hydrogen. By neglecting this fact, a conservatively high flame tem: era-

ure is obtained. The flame front was assumed to :e a :erfect enitter. The a mos:here temperature profile was cotained from CLAS*X.

(See Figures 14.) As j

te discussed earlier in this sec: don, these temperature profiles are known ::

lign :ue to the ::nservatisms of -he ana'ytical 7.ccels in OL.23:X. Since nite gen inc 3xygen are essentially :rans:arent to radiation, :nly One water va:ce ;resen vas issumec :: raciate. 3asec an the ;arttai :ressure Of :ne water va:ce : resent, 5-5

4 an atmosphere emnittance was calculated.

By neglecttng conduction, as discussed above, the heat transfer calculation is reduced to a coupled pair of differential equations with non-linear coeffi-The computer code C5MP III, Continucus System Medeling Program III, was cients.

used to calculate the time-dependent surface tenperature of typical components.

(CSMP III is a'widely used IBM program for solving non-linear coupled differential equations.) The results of these calculations are discussed by compartment.

Ice Condenser:

Since none of the equipnent listed in Table 1 is located in the ice endenser, no heat transfer calculations were performed for this c:mpartment.

Uccer Contatnment:

The atr return fans, hydrogen skimmer fans, and associated cabling are located Section 6.2.1 of the McGuire FIAR pr:vides a discussion in the upper compart.ent.

m of the upper compartment thermal response to a 08A. A comparison of the upper compartment CBA temperature ;rofile to the CLASIX upper ccmcartment temcerature proffle, Figure 3, shows that the DBA analysis bounds the calculated upper compartment tenperatures during hydrogen transient: resulting from TMI-type accidents. Equipment surytvability had been demonstrated in this envirer. ment.

Therefore, no heat transfer calculations were performed in the upcer ::meartment.

Cead-Ended Volumes:

A majority of the equipment listed in Table 1 is located in the accumulator tank These rooms, along with other etuipment areas lccatad below the ice : n-rooms.

enser, are modeled in CLASIX as dead-ended solumes.

(See Section 2.3.)

The

~~

Barton transmitter was selected as representing the ?,ost tencerature sensitive c:mponent in the dead-ended ::mpartment. As statec earlier, the transmittar nas I. 0. ",4 I

'e 3~*

)sc ^*:;, - C hw

modeled as a horizontal cylinder of solid steel. The location of these transmttters ts typically about two feet above the recm ficer. The hydrogen burn was assumed to begin in the lower c:mpartment and pe pagate into the accumulater room through a ficor hatch. Que to the gecmetry of the recm, the The flame front was modeled as a disc with a diameter one half the recm width.

flame was assumed to transverse the entire height of the rocm at a velocity of This process was assumed to occur for each 6 feet per second and then disacpear.

of the ten lower ecmpartment hydrogen burns.

(Figure 4 shcws the atncspheric temperature profile.) The transmitter was assumed to be a perfect enitter and the heat capacity of the atmosphere was neglected.

Figure 10 shcws the calculated temperature profile for the transmitter surface. Accreximately 10 percent of The the calculated temperature rise is due to radiation frem One flame front.

rematning 90 percent is due to heat transfer fr m the recm atmosphere, with the This breakdown convective and radiative contributions being acproximataly ecual.

of heat transfer contributions indicates that an im rovenent in CLASIX analytical models will have a significant impact en the resulting c:mponent temcerature profiles.

In an attempt to gauge the validity of Figure 10, calculations were performed to I

As an estimate the tencerature rise of the dead-ended com:artment heat sinks.

Results of these added conservatism, only steel was considered as a heat sink.

calculations indicate that the temcerature rise of the heat sink will te :n the order of ;200F. Therefore, the resulting peak component temceraturt would :e no greater than 2CC0F. The :i:e of the temperature difference betneen Figure 10 and OLASIX

)

i is a definite indication of the conservatism of neglecting structural heat sinks Results frem an analysis af ta a modified OLASIX code are expected to in CLAS:X.

temperature rises similar to One results Of the neat sink :al uistion sacw coces:

nen toned accve.

$~?

< ;(

,Ts: ' -

~

Lower C mear*_ ment:

The only equipment of interest located in the icwer comparment are the core exit thermocouples and RTD cables. Cable routing varies but is generally The cable was modeled as within the upper regions of the icwer compartment.

The a hort: ental cylinder extending from the primary shield to the crane wall.

flame was assumed to begin at the ce11tng of the icwer compart-ent and transverse A

the entire height of the C0mpartment, dissipating upon reaching tne floor.

The flame was velocity of 6 feet per second was assumed for the flame front.

modeled as a disc whose diameter equals the distance f em the primary shield to the crane wall. Figure i shows the atmospheric temcerature preffle of the lower :cmpartment. The cable wa: assumed to be a perfect emitter and the heat capacity of the atmosabers was neglectec. Figure 11 shcws the calculated temperature profile for the cable surface. Althcugh the flame c:ntribution to the calculated temperature rise nas large-than in the dead-end comparment case,

However, its effect was still small compared to the atmos neric contribution.

the atmosphere's convective and radiative cont *ibu*. ions were significantly When heat transfer is fecm different t.han in the dead-ended c mpartment case.

the atmosphare to the cable, the radiative contribution is almost 31/2 times the convective contribution. When heat transfer is from the caole to the than atmospnere the radiative contribution is approximately a factor of 2 larger The fact that the cable was more sensitive to the convective contribution.

temcerature than the transmitter nas not unexpected since the amosphere temperatures and the surface to volume ratio were mucn higher than in the dead-ended ccmcartment tase.

As in the previous case, esiculations aere perfarned to gauge the validity af Figure 11 by estimating the tem erature rise of Icwer c:mearment heat sinks.

As ex:ected. the talculated temcerature rise was significantly less than :igure

]O A 'Q iw t.

-3 FCR %.

MCCR W

Results of these heat sink calculations show a calculated 11 indicates.

temperature rise of approximately 20 EOF. Therefore, the peak ccmconent temperature would be on the order of ACC F.

As before, results of a modified CLASIX are 0

expected to show medest tenperature rises similar to those obtained from the heat sink calculattons.

5.4 E;utoment Surifvability of Three environmental concerns must. be addressed in evaluationg the ability These concerns are: 1) the necessary ecutpment to survive hydrogen burns.

flame front itself or secondary fires initiated by the flame front, 2) pressure effects due to the hydrogen burn, and 31 thermal effects due to the hydrogen The effects of these three environmental considerations are discussed burns.

below.

Fire:

The concerns regarding fire are twofold. The najor concern is tne actual burning of any tart of a necessary ccmponent is a result of either the hydrogen flame i

The sec:nd concern itself or secondary fires resulting from the hycrogen flame.

is the added thermal affects on necessary ecuipment that may result frcm secondary fires.

In addresstr.g these two concerns, Ouke Power Ocmpany has reytewed the combustibility of the necessary equisnent listed in Table 1 and has conducted a preliminary review of the likelihcod of secondary fires resulting The first review has shcwn that all necessary equipment from hydrogen burns.

is protected from fire. The transmitters and cable junctions are protected by All steel casings. All necessary cable is armored and therefore protected.

sole interfaces with transmitter casings or junction boxes are sealed, thus The seconc review has preventing a flame frem propagating to the inside.

The ice indicated that secondary fires will not result from hyctogen burns.

concenser foam insulation was considered in this review. Therefore, the thermal g4

,)

e l'. :

analysis presented in See:fon 5.3 is still consicered to be conservative, pressure:

With all necessary equipment being *:losed up as a protective measure against fire, the concern arises regarding the capability of this equipment to with-A comparison of the pressure stand the pressures generated by hydrogen burns.

profiles in Figures 5-3 wi:h the contaimnent design basis analysis provided in Section 6.2.1.3 of the McGuire FSAR shows that the pressures are quite comp The fact that the equipment is ;ualified to pressures much higher than shown in Figures 5-3 adds further assurance that survivability is achieved.

One additional concern that arises fr.a these pressure transients is the effect The of the upper-lower c mpartment pressure differential on the air return fans.

11 whether.7e fans trip off due to overspeed of the specific concerns are:

A review of this situation fan blades, and 2} whether th* fan blades deform.

by the fan manufacturer resulted in the conclusion that there saculd be no (Mote that one of :he proolems with the fans operating in their intenced use.

At present, CLASIX mcdifications underway deals with the air return fan acdel.

Tead/ flow characteristics of the fans are not included in the analytical model.

These characteristics are currently 'seing incorporated.)

Temeer s ture :

As discus:ed in Cection 5.3, the calculated surface temoerature :r0 files in Ffgures 10 and 11 are the result of a series of very conserva:ive calculations The major and are not intanced for use in evaluating ecutanent survivacility.

conservatism is, of : curse, the exclusion of structural nea; sinks from :he

n an 3::emot to account for some of ines f analytical models of CLAS:X.

Results of nis conserva:1sms, a second series of calculations was :erformed.

seconc series of :alculations snew tem:ers:ures relatively close :o those :ta:

• ,I U.1 ?'t e.3 gg, virl Ahey y' : 7.,

n= -..

have been previously considered in nuclear station designs.

s Using the more realistic results of the heat sink calculations sufficient information was available for all necessary ecutoment located inside the con-tainment to make the judgment that all components should function in their e

intended use. Sources of information were in some cases analyses based on test

~

reports and in other cases discussions with manufacturers.

tt is Duke Power Company's conclusion that the survivability of essential.

equipment has been reascnably demonstrated for TMI-sype accidents. Efforts ~

are continuing on further refining the calculations discussed. As an example, extensive efforts are underway to modify CLASIX to include structural neat sinks and return fan head / flow characteristics.

If, as a result of these or other calculation refinements, the surytvability of any Title i c:mponent is cuestfonable, appropriate measures will be taken to ensure survivability.

l i

• ^

• *a*,(-

T*1E e*/ ()..g r

'~

l 5-11 l

syy

.a,q w.

s.

. ~-

• eq j

[

N, s

- m s.

n

.s

• , ~,

.,s s

5 n

7

• ~

e

,w,,

a.

s;

',, a

/ /'

s

\\

1

,g V

E i

6.5 FD%'M45sul ts

.; w...,;:. b y%.h.

~

f,_,jThe surecs'e' of' the Fenwal-tests was basically to deternin'e the ignf tien F

6

erfernanc'e characteristics of'a gicw ;1tq, f1ydrogen ignier in a series of' 7 '-

~

n environments representative of various areas within c njaintrqnt during a TMI-s s

(These tests are discussed in ~ greater detail 'in Ch'acter 5. ) Mcwever, i-type event.

A

nis is not to say that some useful data cannot be extracted frem these tests.

ccmcarison of the~ test data in Chapter 5 Gith the discus icn in secticn 5.4 shews

^ that:

I) The cressures achieved in many of the tests are comoarable 'to w'

or exceed those shcwn in Figures 5-8.

\\'

s s

2) ~Temceratures achieved during tne hign hydrecen cc.. centration tests

.c are crAcaracle to these snewn in Figures 1-4 This secticn will discuss the effects c# t'hese environments en the ccmconents utili:ed in the Fenwat tests and hcw these effec,ts relate to. essential ecuiement inside the McGuire Centainment.

Table 2 provides a list of c:mcenents that'wtere used in :ne Fenwal tests to s

characteri:e igni'.er :erfermnce. After ex:csure to a wide variety of enviren-ments and hydrogen burn'secuences, :nly bree c:mecnent failures nere Observed.

Witn :ne axceptien of the checonents discussed belcw, no signs of degracation were exnitited by any of the items listed in Tacle 2.

• /a Mn':er Assemelv:

1early all internals shewed scme si;ns of surface scorching.

This is attributec Oc small ::ening in the ignitor box near the tcs lid.

The :cening was such that 1

• das 1ct covered ty the lid gasket.

is ;cstulated that not ccmcusticn cr: duct

ases erte ed :hrougn ne :cening anc causec :Me sc:rening.

Crecence is Ten: o

nis 9yectnesis by ne :stn Of scor:n marts On :te internal wall Of :ne bcx eminating '- m ne :ceni9g.

Inctner :c:ential :ath :f 9c ;ases ints :ne bcx vas througn ne :cx : ening for ne trans'OrMr :cwer : tale.

~his 0:ening was 0._

I f

not sealed.

Other evident effects frem ex:asure to the varicus tests was a hardening of the lid gasket and a slign: c:rrosion of ne box exterior surf ace.

At the time of its replacement with the Duke igniter assemoly, the TVA igniter still functioned well; showing no adverse affects en its coerating performance.

Cuke Ieniter Assembly:

The only v'sible effect of its ex;osure to hydrogen burn environments was a This accears to 52 due to

" melting" of a small cortien of the box lid gasket.

hot gases ficwing into the space between the box lid overlac and the box Inis distance in the area of the " melt" is en the Orcer of 3/32",

exterior wall.

wnere the distance in the unaffected areas is accroximately 1/32".

The gasket material, Neccrene, has a listed maximum service tem 0erature of accroximately Its reaction w' en ex:csed to the het c:mcustien gases is not unex:ected.

3 n

200..

Mcwever, in scite of the " melt, a seal was maintained as evidenced oy no scorch I

marks en any of the assemoly's internal surfaces.

(Note that the ccening for the transformer :cwer cable was sealed by wraccing aluminun feil around ne c nduit protrusien frcm the box and the cable.) :or both the Cuke and the 77A igniters,

cwer was sucolied by cacle urapced in a steel wire mesh armer with a plastic covering.

No visible effects On these cables aere Observed.

The cable per-formance das not affected by :ne varicus envirencents c whic? :: was subjected.

Wecd 31cck:

Used as a mcunting for the gicw clug fan, Ont; bicek ex:erienced a thin :rewning over mus Of the bicck surface.

Althcugn tnis result is to be ex:ected, it has no imcac: On Mc3uire since wcod is not : resent inside Containment.

.?

Therecc:ucie '_ead W' res:

After :ne :es:s of Phase I:, Part 3, i uas Observec :na: the Tefien insulatien DRAFT 1

6-13

Cfi :NFORMAT*CM s NDrCR REVIEW CHL' a

sco:P.t Th -

.'72J.

s3.... a.. t/ s EN ON -

had burned" off a portion of the wires.

This obviousiy resulted in an Whether this was the result of cne burn or its ex csure to electrical short.

hcwever, with a maximum ecerating temcerature of about 30 burns is unkncwn.

500CF for Teflen and a melting tamcerature of aceut 550CF, the effect appears This s eculaticn is based to be a result of chronic and not acute excesure.

en the fact that the average air tercerature nas much higner in the Phase I tests tnan in the Phase II tests.

If an acture excesure was the failure mech-anism, tne failure should have Occurred in Phase I.

In any event, this failure mecnanism cannot occur in McGuire since all necessary cacling is artered.

Fan *ctors:

Two failures in the three fan motors used aere cbserved -hroughcut the Phase I In both cases the failure mecnanism was detachment of a and Phase II tests.

soldered wire connection. *ibe:her the fsiivre is a result of chrenic ex:csure, as may be evidenced by fan me:Or 11, or an acute exccsure, as may be evidenced by fan motor 13, is unkncwn.

Arguments could be mace for both ;cssibilities.

Mcwever, the ;cint is mcc: since solder is not used for electrical c nnecticas inside of the McGuire Contairnent.

Electrical connections inside of the McGuire Centainment are made by wraccing the wire ar0und a lug anc tigntening with a As an added ;'etective measure, electrical c:nnecticns are enclosed

' 9 nut.

in a Orc: active heuting, wnether it :e a junction box or an instrument asing.

The cnly 0:ner cbservation regarding the fan mot:rs was a slign: cxidation film

n f am me:ce fl.

This film cid not affect the f an. meter :erfccmance.

In ?hase II, Part 1, of :ne Fenwai tests, several : mcenents tycical of ccmcenents These found inside of a reac cr containment wnere : laced inside One test vessel.

mcenents are listed in Tacle 3.

Adci icnally, tre transmitter casing, tne limit sat ten, and :re sciencid valve were in::rumentec ui n :nerecc:ucles to :rovice Also 9c:a :na: the ~7A f; niter in eri:r air anc ex ar'Or sur ace tem:eratures.

in uni:n i: xas used.

iA : eta'lec eas 'ns rumen:ac :ne same day f:r all -he tes :

5-11

discussion of the Phase III, Part 4 results is presented in Cha:ter 5.) 4s expected, the data shows that ai varies frem about 2CCF at Su/o hycragen Oc about 500F at 12'J/o hydrogen. ' A ai of acoroximately 50 F accears to also be Since no a good average for the additienal instrumented ccmcenents in Dart 4.

attempt was made to prevent hot gases frem entering any of the instrumented cemcenents, it is reasonable to expect theses aT's to rise if the comocnents were " closed uo* to prevent hot gases from entering.

This "c1csed uo* si:uatien describes the design of necessary equipment in the McGuire Cen:ainment.

Thus, this data demonstrates one of the conservatisms menticned in section 5.3:

that of uniform temcerature thr0ughout the ;cmocnent. Of the other ccmcenents olacec in the test vessel, the only visible effects of exposure to hydrogen burns were a very light oxidation film on the :aint samoles and two "scoren" marks en :ne unarmored cable. The oxication film cn :ne paint samples is not unex:ected.

Since the coatings used inside of the McGuire Containment are not cemeus:ible, there is no concern over a prolcnged exposure to hydregen burns.

The two

" scorch

• marks are prcbably better described as scrace marks.

There were no.

indications of insulation melting.

It is possible that the arks were un-observed en t$e cable prior to its placement in :ne test vessel.

In any event, even if the marks resulted frem ex:csure to the hydrogen burn, this has no imcact en McGuire since all necessary cabling insice the McGuire Cen:ainment is armored.

5.5 Ccnclusions As a result of the analyses and tests described in this section, Cuke 3cwer Comoany draws the felicwing conclusiens:

1) Heat transfer calculaticns using the CLASIX temcerature :refiles result in conservatively nign calculated c mpenent surface temceratures tnat are not representative of the ex:ected temcerature res:ense and snould not be used to evaluate ecui: men; survivacility.

DRAFT FOR INFCRt. tai:CiN

• * "e ANDoCP RE'tW'n M m

-?-e,

2) Incorporating heat sinks into the analyses provides more realistic results for ccmcenent temceratures and indicates that these temc-eratures are close to valves already considered in the design of I

McGuire Nuclear Station.

3) The Fenwal tests demenstrate that various ccmcenents can withstand and operate during hydrogen burns.

4) Two functienal failure mechanisms were observed during the Fenwal tests.

These failure mechanisms are not acciicacle to McGuire due to the staticn design.

5) Sufficient analytical and test data exists to conclude with reasonacle assurance that all necessary equi: ment insice McGuire Containment will 4

function as intended during hydrogen transients.

Efforts are continuing to refine and re-evaluate the analyses described.

If in the future, new information calls into question the survivability of a ne-cessat/ c0ccenent, Duke Pcwer Ccmpany will take accropriate measures to ensure that comcenent's ability to function as intended.

s DRAFT FOR INFCRMATICH ANC/CR REVIEW CNLY 5-15

J TABLE 1 Necessary Ecuiement Centainment Air Return Fans (1) (2)

Hydrogen Skimmer Fans (1)

Hydrogen Ignitors (1)

J Steam Generator Water Level Transmitters (2) 4 I

Pressuricer Water Level Transmitters (2) i RCS Loco Temceratures (2)

Core Exit Thereccoucles (2)

RCS Pressure Transmitters * (2)

Centainment Water Level Transmitters * (2)

I

• not located in Centainment 4

l (1) Required fer contair. ment integ-ity.

(2) Required for shutcewn and coo' :cwn.

J DRAFT FOR swegAfATICM 4%# REVIEW ogg t

4 4

l 1

\\

TABLE 2 i

COMPONENTS IN FENWAL VE33EL FOR FHASE I AND PHASE II TE5TS i

EQUI. MENT NUMBER OF TEST EXPOSURES, D

30 TVA igniter 7

Ocke igniter 20 wecd block (4" x 4" x E ')

10

nermccouples tnerecc:uole lead wires 30 a) uncrotected b) wracced in aluminum foil 0

1 s: ray no::le 4

i fan mot:rs 20 a) di 20 b) #2 1

c) #3 i

i J

I4 r,

c.

ll#.J F' *J l

4 w

4 Table 3 Cemconents in Fenwal 'tessel for-l Phase III, Part 4 Tests Ecuiament Numcer of Test Ex:osures Paint samoles a) en concrete biccks i

b) on metal slabs 1

SX-ty:e cable i

Unarmored, insulated cacle i

Namco limit switch 3

Asco solenoid valve 3

Barton transmitter casing 5

Miscellaneous wiring i

Fiscner regulator 1

i

{

i DRAFT

'%_Wo Tervirg qq m marm t

r

t

.~ w.s u R.

--. c e.v t.ip. e.

c. uc r.ve..i s i.e.,s e n

.s n

i a

i: i

' i e v

. r,

W 4 i i i i i a e i

j i

i !,,

i(

! riiiiiei iii'.I 1 4 ie i ! ' ' 4 i i i

l l5 i i i.

Jie i i

! t 61 i 1 eeiiii! I I i i iii i i t i i

• i 4

i t

. 4 i.I g

4 I i i i t 4 i ti* I i i i i 6 I i II iiiiiie i,

ii a

i i i i t t

! I I i i t t l i i f f I + !

I i

s i

i i i i 6

s

• 6 4 i

i l I I 4 i i ! ! ! I i i l I i i i e f

? i i I

i

}!it !'iI ! t i I I I i 1 i i

• I 6 I tiI i 1 1 II I I i i !e i

' i i 1 i t ' i 4

' i i f f i i f

t i

t i !

' ' i 1

f f i

' ' !I i e

Il 4 1 ii111 1 i i !!

_f f i 6 '

t ! e t i i i t a '

e i

f 3mJ

)!iiit t'I i i i li'i1 16 !

.I 'ti

• i

(

j i i t i Ii+1 Iiii! I i 4 l I.I i 4 is li 111 t'

t i i (

I i i !

t i

liiifI Ii! i i l i i It i i :t 1 l '!! !l'.

-l' I

t' t

I t

t e f

Ii tiiilit 1Ii1 i I 'I t i. l i ll lp t i I ll ;

1 i

e l

ti l I il l 't

{ i et f (; 1 jf i (a i i l s t # ( s i W

eiit i!! t1 4 i '

=

i

(

i lt

].

liiitt i ! i ! Ii all I

.3 1 1: : l! i lt i l i i i

l ll h

l*it I lI l t 1 i l ' t ll1 l I ; y ' t. ! !i

(-

t i i 2

It i i iti i i If;ie ii :19 l t 4 i i: i ll f-ii ei, l

11 i ji. i

,4

!6 8 I i It t : it i i.lt

<ll t {t (

f I i i i Iii11 :ii i 111

.Ii.I t.; iit l-i ae t

l W liiI

t i,,

I liiI6 I II i i liI i hti h !*t ti i i vil ):6 l

.ti,

f l1 1 1 I i 6 t Ii!ii F I lt] I { t i I 4 : 6 :#l t Ii'I ii4 i

i l

i i t i ! It t l t tf 16 I t 'l ( lid :l i l!q i (l*

19,

a l1

,I l l 1 l 1iiit;ItiI il,

l't'I'6:ta

! l \\: i i i l1 I i i i 1 i I I11I i il \\

s i 'll N I ij l ili s 6 ' 4 1 :(

• 4 l

lt 6 1It i It i !6 I i *l I ' I I iit I ! 6i l l 1ii1) i N *iii

• i t i i, i.

linet i I

'A_* I l t e i i i i i t i e t i t #

il i eie i i i ;i! r i t i t i i ei

: > i i :

lii

• /i

(

e i. e eie i, ;,

,ri t i i i e i i ! ' i t i !

i e i i i

(

I si i'i l

i i 4 1 6 i !

a 1

(,

l}

i 6 i M

.6 351J 33.4 m

• i.T ( SE.ve5 )

.e-i t

I

=b

4. *.\\

g

\\

w

23CRE E

CE C3CE35EP TE'PERAME 4

illllil!!'tIi!lll1!IlliililIiilIilill11i I

l1lllll1ii!lIIlllllllilll4I!lIlllll!lll

!Ilit !!!;llllllilllll!lll!!.illllll!llll' l

llill!!Illli'illllllllilliiilllli!!ilill I

Il!!!iiililiillll!!Illlillllilllilillllii I

!Il!!'!!!!!!!Illil lilillil:lill!llilill l

li.

.!iilii!!i!!l ll!!l}iilllll!,Illlll l

liiill iiii!,illill ll!!ll111111ll!!lll1 l

c Ill!!!!!il!illllll lilll!!illlll!!! lilt i

litii!Illittiililllllllllllllllllllilli t,liliillilll!!!llill E

lillllillllli!illili i

E lIllllillllililllll 1111111 l l l l l l 11 l l l l l l l!IIIil!Iil!lll!III lllllllIIIIIlllllllll

!ililllilillll!llli lillllllllililllilli i

lilllli!il!illllli lill!Illlillil!llili' I

lililillilil!Illl!

lilllillllllillillill Illiilllil!!Illlli lillilillililllillill Illlfillllllilliti, Illllllllllililll!ill llIl!IilIllilI!lIl$llllllilllll[l!!!llll

, G IIIIll N ITiTI l'!!llilil l.i

..i y

u au an.a maa en.4

?:m isa::yes DRAFT EOC INFCRMAT.CN ANClCR 4EVIEW ONL'O

~~' ~ ~~

' ~

..z-

=. -

t

• t.w. o.:.,

1 u.pe r.o. e. sp ame.n.,. me r.'v* ". +::.

n A

4 i

i l

ra.a bI as t

J N

n.a i

1

e.a j

ws g

v

_2.m 2.4 i

t 1

i c

.:s.4 I-__

l ma i

an.a m..

(

?:s <st::mm l

.1.ht FCR '.WCR.yAgn NCiCR REv:gw gp g

d a

g a

me-

,W.

6 4

=-'w

a e-I O

9 f

J p9 h

i s

l 1

. 'GURE 4 CEAC-E?:CEO CP:C'E37 TE'dPEWJRE 2

i i

4 l

1 l

1 a

66 e

s 1

i b

4 a

3 4

l i

h-4

}

E I

a 6-i l

e 0

47

. i 4

i i

e I

i

.~

SE 5

1 I

.)

I.

y e

Y f.

5 m

6 0

W 5

EM i

i i

I f

g i

J I

\\.

e I

f M

I t

i I

J t

~

i W

4 t

ame -

1 La m

3md 324 4md i

?I8 TIE N 3 j

4 i

a i

s q

.g s

m-ea e

. ;.. s s

d p

1 i

i k

d o

--rww.--

c

-.%-----,y.-----,,

- - - - - - --i-,m%en,,r2w----.,-,-%,.,-w-

%g-

,ev.

.=.

\\

~'GURE 5 Lt.;c.o.. t.~uc. 1:m.. :.st.

c:.c.p. :

~t.

l i

l M,

c.a a

r e

i a

4 I

g t

e i I {

e s.a i-

-1 c

6 4

f a.

pg.

1 g

~

I i.

e y'

If

_d a

ga

,p.

6

.~~

g,3 v

i I

a4.

I.t -

p.

2.a '

M N

N 35.J 434 1

r:.c im:ress w-DRAri TION

.l ~R INFOp,u e...np RE/tCN CNL e6y 6

w Sese

r 5%- r e,.C : :

s' e.: e.r:t C E'tS =.=.. = ===..=U ' c.

e

+

a r

=

l 1

i 1

k i

f f

i i

i

&J.

+

o I

&J e

6 y,

i t

I i

i s

a s s

a

. 4, i

y4 3

4 i

i g

i f

pj e

I e

pe I

t l

6 3

i a

i

.c.

I 2 4.#

i t

s i

i

/

g, 'i 6

I

&J 1

f u

m ama M

Ma 1

?!.T ILW a 1

n,.."r-

E s; s,

.--c=

y.si

~.

__,,--..,,-,.,,--r.,,

,,.,,,-%,,m.-

,--.,---._v.,,,-m,y.-7,y_.

w

,9 i

i

.IGURE 7 URCER C~M;aRTAEN~ ;;E55URE ll1IIiiilIlll}ll!!

lI'I!!it ' IllllI!!!i'I

]

lll}lIi!IIIlllllll lllllIilIIl!lll!lll lllliillllllll!lll llI!!llllIllllllllI!! iilIiiliiilI(11(hl3Il!Iiliii!IlliliiiilI f I ! I 11 l ! ' I l l i I iil li:!i'l i l ) i,! I i i ! I I l i i i I ! I i l i l l i l l l l i l l l l l l ilil i t'islMlll d I Ai l l t i l i l l ! I l l l ! i l l ! I l l l iff i l i ! I l i'l W '.l!1 ii l l l\\l ! l l i ; ! ! l l i l i l l i l l l)1 t i l l l i l ! I l l l i l il l NJ N i l i l i IllIB y fl!!illlilllill!Illl!!illin ti! a E Pfil!lillllilllilllllll!'l!!Ilililillill l j " )'llllilllililllilllilli!Ilillilllii!Il!!! i lil!I!llililllillill.illlill!llliiitlilii i j lilllil!Illlllifillll!!I!illill!ii!!!'!i l l l l l l ! i l l l~l l l l l l l l l l l l 1111 i ! ! ! i i ! ! ! : l l ! !Illillllitillllitilllilili!!!!f!ll!!iii tilllllllilllllllilllill!!!!!llillililll lillllillllillllllllllllllililillil!lli! lillllli!!illllifillll!!iilillllil!il!ll I lillllllllllllillllitIlli!!illlilil!I!!! llIliIIilllllllIlIlIil!l;i!! IIi i:!i !!!l " e.a an.a an.4 m.a e4 I ?!?t t K:"NOS 1 1 E, i FOR INFORMATION i ANoiOR REvtEW ONLT i D es e g.mm., e., en ee i L

., ~ q Q 4 e <e l me,Uno ..: MU 1

====.. 4 4. a g. j e, - 'd,,f Y

• U; U'E

'emel F ** R I.. U l #$U)w 4 4 4 3 1 I i 4 I i l, b 1' I 4 4 l I 1 e 1 o t e i t j l 0 ( l l 0 a s j 1 5 M. a e s -j 1 I i' e - e e f a e i 3 i ee 3e.d. 4 e I I g e i e e i e 9 i j b gg' e 2 gr a j a A 6 e 7 j i e .O e i.. g* e l r - me l } t 5 I I I V i 6 ed. i 6 2 + e 1 i e i I i k u m mu m e J 1 ?!M (E M ) l 4 h O h l a - j .,,., } *. es A g. ,,ag g ?E U,,,..'J w-i. t -N f i f ~ g i. l

l J. i 3 i g i l i i 8 i l 2SfMI / lillTIAL C O N DiflOt4S / y P - 100 kPa a 1 alm / I - 298 K - 25C Allt SAll11(AIED Willi WATER VAPOR [ / 6 2<MM) / 2 'M 5 9 1500 6 -t l n' N 4

• • 8 IEMPEll AltlilE /

a '~ i,xm N/ 5 u / 38, u d l>llESStillE ,/ e s<W, rs a B 7i - 2$ "5O s$ 3 ~-h ,v, p si I -i % - 1 4_ __.1 L 1 . _ _1 g oN O 4 8 12 16 20 24 28 }2 ,,,,,,gt,,y,3,,ggg,, gg,,gg,j y,,,j,gg gj g l'itl';Silitt ANll llMI'lltAltllti. AfIlit ilYl;ltI4fH-Allt UEllllSilllN, UINSI ANI VillilML ANil AlllAllAllC l i 16 tilt! 9 i l

4 S e e' 9 -1 8 I i ! Q i 1 m I umS 8 -. ~ N $4 N i + Z N } 9 ~ N$ 3 5 d 4 '1 e 1 ) M lt N = I E= me E

a. =*

l .Z. E -3 4 C es

==. %I

k. -

~ 23O A T a g mI Y I -= Q =4

== b 4= 6 M Bim E DRAFT ,,Nmanen M 6 jt c,9 "*N L e TNO r e i e i e = a ~e n e o 45 w 3 2 e 3 .N m g M Jg

• Eed)$N e

e

1. h 3 S

p Me 6 a

enop G - erE i l i 9* e j 9 + l s - t 4 N i 8 N d k N d e .E. uma / = gr m $7 7

== a mg d e M m wNse a w

1. tuer E:-

I P <z E== '.d y

== Z.

• Se a -

+ g ,e w at W =b w ",fg g, ' x I,'l /, a g e .e s

• ww e

e- .G. e v, .s ..a* se

1. n.*

es. wE E* -E e-

== + U w i E e== s

• C W

W e i l I I O N N N N N N N O =* m 2 P p m N p m

== 4 i 3HS11fE3arG1 n =

o, s.' ' MEETING

SUMMARY

DISTRIBUTION g Docket File G. Lear f d( T }/ *. J .,) [ h 'i NRC POR V. Noonan Local POR UT 1991 S. Pawlicki TIC /NSIC/ Tera V. Benaroya p FEB 061981

/

NRR Reading Z. Rosztoczy LB#1 Reading W. Haass ? v 5 g g m== J. A H. Denton D. Muller \\ M E. Case R. Ballard D. Eisenhut W. Regan q _:. \\ J- ~~ R. Purple D. Ross B. J. Youngblood P. Check A. Schwencer R. Satterfield F. Miraglia

0. Parr J. Miller F. Rosa G. Lainas W. Butler R. Vollmer W. Kreger J. P. Knight R. Houston R. Bosnak T. Murphy F. Schauer L. Rubenstein R. E. Jackson T. Speis Project Manager RBirkel W. Johnston Attorney, OELD J. Stolz M. Rushbrook S. Hanauer OIE (3)

W. Gammill ACRS (16) T. Murley R. Tedesco . Schroeder N. Hughes D. Skovholt M. Ernst NRC Participants _: R. Baer C. Berlinger i See Enclosure No. 1 G.' ton A. Thadani D. Tondi J, Kramer D. Vassallo P. Collins D. Ziemann 4 i bec: Applicant & Service List .}}