ML20058A840

From kanterella
Jump to navigation Jump to search
Forwards Slides Presented at 931116 Meeting W/Nrc in Rockville,Md to Present Response to NRC Findings on Gist
ML20058A840
Person / Time
Site: 05200004
Issue date: 11/18/1993
From: Leatherman J
GENERAL ELECTRIC CO.
To: Borchardt R
NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), Office of Nuclear Reactor Regulation
References
MFN-200-93, NUDOCS 9312010264
Download: ML20058A840 (77)
v • d • e


Text

,

)

GE NucIcar Energy Genera lDettac Compary 1% Cu#er Avenue. San Juse. CA 95125 November 181993 MFN No. 200-93 Docket STN 52-004 Document Control Desk U.S. Nuclear Regulatory Commission Washington DC 20555 Attention:

Richard Borchardt, Director Standardization Project Directorate

Subject:

Response to NRC Findings on GIST Enclosed is a copy of the slides presented at the November 16,1993 meeting with the NRC at the GE offices in Rockville, MD to present the GE respo 4e to NRC findings on GIST.

Sincerely,

[.

7 Ju@v Jr fS on J. E. Leatherman SBWR Licensin ' Manager MC-781,(408)9 5-2023 9312010264 931118 PDR ADDCK 05200004 A

PDR unx.3.,o j

c

0 9

GENuclearEnergy SBWR-GISTDifferences and Systems Interactions T.R. McIntyre SBWR Performance and Technology Manager November 16,1993

. p.

g.

?

SBWR-GISTDifferences

. GISTis an adequate simulation of SBWR to:

- Confirm the theoretical feasibility of GDCS i

- Provide additional data for TRACG qualification Differences existdue to:

- SBWR design changes 1987-1993

- Facility design requirements and compromises Differences must be identified and reconciled to assure that:

v

- Important phenomena are not effected

- Non-representative system interactions are notpresent oh.

_ - _ _ _ _ _ _ - _ _,._._.;.~_._.....--.__..._.:....._-,

SBWR - GIST Difference Identification Cateaory S8WR GIST Vessel & laternals Downcommergeometry Continuous Anulus 2 separate pipes Separators / dryers includedin design Notmodeled Chimney configuration Shroud extension Modeled with standpipes Core geometry TypicalBWR Heaters in concentric circles I

Deoressurization System ADS configuration 8 SRVpiped to pool AII ADS flow to 6 DPVto drywell suppression pool Gravity Driven Coolina System Watersource Separate GDCSpools Elevatedsuppressing.

in drywell pool Piping Geometry 3 lines, each branch 4-4" lines, orificed into 26"-lines Inletlocation.

1 Mabove TAF

, 3m above heaters Decar Heat RemovalSvstems Isolation Condenser 3ICHX

.Notmodeled.

Passive Containment Cooling 3 PCC HXDWto GOCS.

. Not modeled' pool

SBWR - GIST Difference Reconciliation - GDCS Performance Cateaory Reconciliation from Reconciliation PIRT Data Scalina Vessel & Internals Downcommergeometry X

X Unimportant-Does not effect depressurization rate, heightpreserved Separators / dryers X

X Unimportant-Low steam flow, low OP. Does not effect depressurization rate.

Chimney Configuration X

X Unimportant-Chinney height correct, bouyancyproperly modeled Core Geometry X

X X

Unimportant-DPproperly.

scaled, would only become

/

importantif core uncovery occurs. Core heatup not a test objective.-

Oenressurization Svstem ADS configuration X

X Unimportant-RPV depressurization insensitive to flow destination, containmentpressure a test variable, drywelland wetwellpressure dependent variables

SBWR - GIST Difference Reconciliation - GDCS Performance (Cont.)

1 Reconciliation from Cateaorv PIRT Data Scalina Reconciliation i

GravitvDriven Coolina Svstem WaterSource X

X X

Conservative - DWathigher pressure than wetwell will enhance GDCSflowin SBWR relative to GDCS. DWpressure always higher than wetwell, but NOTindependent. GDCS L

sensitivity to containment pressureminimal. Absolute RPVpressure effected; GDCS driving DPrepresentative to conservative.

Piping Geometry X

X X

Test Variable - Results sensitive to flowarea. Testmatrix bounds SBWRrange.

l Inletlocation X

X Unimportant - GDCS driving pressure function ofsurface elevation.

Decar HeatRemovalSvstems isolation Condenser.

X X

Conservative -Presence ofIC wouldremove energy and increase RPVdepressurization rate.

Passive Containment Cooling X

X Unimportant - Containmentpressure dominated by air mass location, not heat transferprocesses l

.. ~.

~

SBWR - GIST Difference Reconciliation - Systems Interaction 1

Category PostulatedInteractions Reconciliation 1

Vessel & Internals Downcommergeometry Asymetric effects observed Unimportant-does not effect RPV depressurization rate, or GDCSperformance Separators / dryers Nonpostulated Chinney configuration None postulated Core geometry None postulated Deoressurization Svstems ADS Configuration Potentialloss ofIC function Unimportant-No credit taken i

due toICsteam source forICheatremoval. Willnot l

proximity to DPV.

effect depressurizationrate.

1 No mechanism exists for-degredation of DPVflowrate.

i GravitvDriven Coolino Svstems WaterSource None postulated

. Piping Geometry None postulated I

[

Inletlocation ~

None postulate'd

--....~.....,_.m_.=.

L SBWR - GIST Difference Reconciliation - Systems Interaction (Cont.)

l Cateaory PostulatedInteractions Reconciliation Oecav Heat RemovalSystems isolation Condenser -

- See ADS-Passive Containment

- None postulated Cooling Non-Safetv Svstems Non-safety Coolant RPVDepressurization Mostlimiting system (CRD)

Systems due to non-safetyinjection simulatedin GISTand effects is non-conservative quantified V

~.. _,

Conclusions i

SBWR and GISTdifferences do exit Differences do not effect important phenomena Postulated systems interactions do not effect GIST results w

5 GIST Facility Design is satisfactory to

. Meet test objectives 4

,w+-e&-a m

v-e^

g*-w m

+

tinim -s im

=-w--

v-

+

vm

-e n4

+ +

gw--

-~

ww e

h e-w

- wm g w

---w ev%,,v---

er e-te r4e-T

-w'*CT'

-TW T-t vr me-we tMe

4 9

GE Nuclear Energy GISTScaling F.J. Moody Consulting Engineer November 16,1993

.. - ~. ~.

. ~.,

Topics Purpose forscaling GDCS Summary ofscaling analysis Scaling concepts employed by GIST Response time Specific frequency Top down scaling Bottom up scaling Parameterspreservedin GIST Distortedparameters in GIST Scalingconclusions 1

2 i

___.______.__.__....,_.__m

___....-._.-..__.m...._,-

Purpose for Scalina GDCS To show that the performance of a reduced scale test (GIST) provides a representation of dominant phenomena and system response, which is sufficient:

To confirm the technical feasibility of the GDCS To provide additional data for confirmation of the TRACG code usedin SBWR AccidentAnalysis f

O-I

- w.

.mm i__m_--.,_.,mr.

..w.<m.

,ws.

s

.w am...,

e..

+,%

e.-

3, w.

.,,w,.

L i

Summary of Scaling Analysis i

t

  • The GISTfacilitypreserves the dominantphenomena which occur during operation of the S8WR GDCS
  • The volume scale of GISTis 1/508 of 1987 reference SBWR design, corresponding to 1:1 vertical elevations, and 1/508 horizontal area scaling of the RPV, drywelland wetweII
  • Fullscale SBWR fluidproperties were employedin GIST
  • SBWR and GISTtime scales are 1:1
  • Representative Scaling of GDCS

- Top down (control volume) phenomena (system pressures, flows)

  • Non representative with minimum impact

- RPV upperplenum percolation

- Stored heat simulation in lowerplenum

,w-,-.

,,.,.,w>.w

.,.+ -

,w-

-.w e-we.----

e w

.w.-*.

,e, w

.---i.s--

e

+ cw w

,+, ',,

.r y,

1 Scaling Concepts Employed by GIST Response Time, t*

Sometimes called:

- Time constant

- Volume filltime i

- Residence time

- Reference time f,)

ll T o

Q

=

Reference Volume

$* =

Reference Volume Flow Rate.

V.

T*=

-r.

Scaling Concepts Employed by GIST (cont.)

ResponseTimes Associated with, E.G.,

- Filling / Emptying Volumes

- VesselDecompression

- Valve Opening / Closing FlowAcceleration

- Pipe Flow Transit

- Heat Transfer l

l

.e-

- m.i m.--..

f..

Specific Frequency, Q 1

Described as TransferIntenshv Capacity (A)

=

f

= Capacity of a physical quantity

" = Transferintensity of f cap

=

p.

.r.

PictorialSignificance of%"and M Ip System pressure, or otherproperty (top down phenomena) kNon representative high frequencyprocess (bottom up phenomena)

Waterflow or other transfer rate

=

Time I

Not necessary to exactly scale transfer rate start-up since

&o M

$p e

Not necessary to preserve high frequency phenomena to scale representative low frequency phenomena

Top Down Scaling (Coarse Structure)

ForAverage Properties in a System Example:

Mass conservation d"=N

.i cR W W.

Energy conservation y

E T) 4.E e <Ut + 4 ha at State MsTY E=Me i

e e = e.(PsT) =

g-t }

Non Dimensionalize m

+

t e s.,PERS. G A 5 f+

  • g q; p, p_ p; g+, g.

-gg Ap

)

a Mass of Energy Conservation Become s

+

a p *'

(F-t)ft.

ha ho(W'll ap g to 4

cit

  • vee Me?

_ yp 4

.-2. -

- u u 1

. o m

s u am m a

v

.-L

<_s--,-__

m-r mer-

-w+=

=

  • ,e

-i- - w e e-v ww e-w rie---

-w-+-wi o

asw i- -+ w re+=

eww

+

w

< e se mwr

+m

-rfr ywe--~w

+w*v-

~

=

s For GISTFacility, Top Down Scaling Yields

/s T*h

. (b t* 3, W 4g

/ $ - (v TgssewR

./Tx-0fi t*1

{(x-Ogt' 3;% '( V4P-)58%R

.y 4 9 f( k-t N h t * '

(((g_qskot*3 *(

V AP

/sawR V AP J3 Prototypical fluids, properties, Q**1,q*ewR s

A

-5 HS NS fas N9 M

_g I'***

3 ts awe.

(H A )sawa

_~

Nsswa

$sgwa, t

= R Fullsize height, Y

= N$%wK y

4.-

dP

% 0E 6WK 3

5 69 t. K 2% {

}

t i

Flow resistance K g K ss wg (/ Mag-gt g,,

j

-fS

Heat TransferEffects Nu 4

o T 4k'

~

Qo Heat Capacity of Receiving Volume oE' s 4* g*eA*

HeatAddition Frequency W e. '

d e

,b a

33.,

ygw EnthalpyAddition Frequency h*ok*

  • h*db'_

s Nok W-Q*g*Ak' T ' pao j

I Phase Change Number Q.

T eg s C4 $o Au M g*

p When Eps# t, the specific heat transfer process can be neglected i

Bottom Up Scaling (Fine Structure)

. v.

q1 + s v-v = o V' C N

ot

" O h 9.y.Sye +3 h %g.e -) b_I-bl{KYr+c5f-V ot ter f

us t

Non Dimensionalization may yield j

Fluidpropertygroups Vse V

Groups with geometric dimensions TT s l

T pp Groups preserved in prototypic fluids l

TT6 Groups may suffer distortion and bottom up phenomena may not be preserved However, it is not necessary to preserve bottom up effects in GIST in order to obtain representative top down processes t - l ~.

E-l--__._=____.____________

Parameters Preservedin GIST Core Heat Pool Volume PoolArea RPV Volume RPV Vo!ume RPVArea i

Volume (Mass) Flow Rate.

Elevations 1:1 RPV Volume Distorted Parameters in GIST Parameter Reconciliation RPVSensible Heat initial waterlevel was varied to RPVvolume account for lower plenum void collapse RPVHeat Loss less than 8 percent of CORE RPV Volume heatgeneration GDCS Pine Flow Area GDCS flow area was a testedparameter RPV Volume RPVpressure, inventory, GDCS 1

Core geometry and flow rate not affected

~

flowpattern Downcomergeometry

Scaling Conclusions No parameter distortions in GIST have a significant effect on scaled GOCS flow rates orinitialstates 1

t The GIST test is scaled such that important phenomena are simulated

}

1 Therefore,

- Technical feasibility of the GDCS is proved

- GIST data for qualification of TRACG is appropriate for SBWR 4

__.-_.._._.-.,._.____._..m__.

__________._m_

O GE NuclearEnergy Response to NRC Findings on GIST l

1 l

November 16,1993

. f.

Agenda Introduction T.R. McIntyre

- Meeting Goals

- SBWR Design Philosophy

- GISTOverview

- Acceptance Criteria GISTas a SBWR Simulation

- Phenomena Identification and Ranking JGMAndersen Table (PIRT) t

- Scaling F.J. Moody 8

- Configuration Differences & Systems T.R. McIntyre

?

Interaction

- Heat TransferEffects JGMAndersen i

  • Instrumentation & Quality Assurance P.F. Billig

^

Summary & Conclusions T.R. McIntyre GE Responses to 10/4/33 NRC GISTAudit P.F. Billig

.t.

L Meeting Goals l

  • Resolve NRC Concerns on GIST

- TechnicalConcerns

  • Physical differences between GIST and SBWR designs
  • Facilityscaling approach
  • Effect of heat losses on test results
  • Test Matrix andinitial Conditions

- Quality Assurance Concerns

  • Design Record File (DRF) deficiencies from August

'93 audit

  • Reach Concensus prior to 12/17)93 ACRS meeting 2

.I

SBWR Design Philosophy Core Cooling Design

- Very Simple ECCS

- Natural Circulation /No Core Uncovery Inventoryis the oniv variable No core heat up

- ECCS Issues / Appendix K not applicable Depressurize RPV and Flood Using Gravity Simple, weIIunderstoodphenomena Systems Design

- Minimize Use of EmpiricalData Qualification of firstprincipals Engineering ComputerPrograms

- Maximize use of Existing test data BWR-1 through BWR-6 and ABWR experience

  • Simple Design.
  • Experience from BWR-1 on u.,...

y m

---r__-i.e-#w-*__._-.~,.,mi_

.w....,,

,-_e

,er.

m u.r.-

z.-, - _ - = =

c..,i.,s.

ew m

__________m.____,____

.m.. m

GISTOverview i

Test Goals

- Confirm the theoretical feasibility of the Gravity Driven Cooling System (GDCS)

- Provide additional data for TRACG qualification

  • GDCS Flow rate
  • GDCSInjection Time Test must be scaled and designed such that importantphenomena are simulated

Criteria for Technical Acceptability Phenomena Simulation

- Phenomena Identification and Ranking Table (PIRT)

- AnalyticalorExperimentalConfirmation

  • TestFacilitySimilarity

- Scaling Studies

  • Identification of attributes
  • Reconciliation of distorted parameters

- GeometricalDifferences 1

  • Reconciliation of differences Test Program Quality Assurance GE willdemonstrate thatallof the above criteria are met

~

. -..,,.. - -,..... - ~

1 Reconciliation of NRC Concerns NRC Concern GEAcoroach Acceptance Criteria PhysicalDifferences List differences and Differences lead to conservative j

between GIST & S8WR reconcile results orare unimportant Facilityscaling identifyimportant Reconcile alldistorted attributes through top-parameters as unimportant to down and bottom-up key test results studies Identifyimportant Importantphenomena are phenomena through simulatedin GISTexperiments PIRTAnalysis Heat Transfer Effects Utilize scaling study Parameteris smalland may be to identifyimportant neglected attributes Use test data to quantify Benchmark against TRACG RPVheatloss Perform parametric input changes do not appreciably studies with TRACG effect results Test Matrix andinitial Compare testmatrix importantand/orbounding Conditions with SBWR range conditions tested QualityAssurance Demonstrate deficien-DRFavailable forreview cies are in documenta-tion, notin test'itself

---,m-

-.vs.

,,w--tw,--w-

=

c-

-wr+

,4,,-e-w

--ere e--

. i m.

--,....+-v',--e...

,. ~ >-

..--e#-.',

s.-.

O GE Nuclear Energy Instrumentation & Quality Assurance for the GIST Test Program Presentation to the NRC November 16,1993 PF Billig SBWR Test Programs

=%

SBWR W

-,,a

e A

SBWR QA Deficiencies from NRC Audit W

I Test classification as non-safety-related developmental test Design Record File (DRF) Deficiencies

- Missing documents:

Final Test Report Instrument calibration records As-built design drawings

- Reference to original data tapes I

- Verified input from TRACG analyses

- Non-conformance reports

- Documentation of analytical or experimental verification of engineering calculations, specifically facility heat losses l

All have been resolved l PFB-2

A SBWR Test Classification tg Tests were classified as " developmental" for new reactor designs GE procedures (EOP 35-3.00).do not distinguish between

" safety-related" or "non-safety-related" All QA procedures were followed for engineering tests

- EOP 35-3.00 does not implicitly call for different levels of QA for development and design basis tests

- Procedures followed for GIST satisfy QA requirements for

" design-basis" tests i

PFB-3

.,.. ~.,. _

A

~

SBWR Missing Documents in DRF t?

Original DRF has been supplemented

- Reference to Final Test Report.(NEDO-31680) included Calibration records for all instruments included

- NBS traceable-As-built drawings are either included or reference given if they are i

retrievable elsewhere Reference to location of original data tapes included Non-conformance reports. generated and included

- Disposition of all " invalid" tests were settled by the Responsible Test Engineer and the Test Requestor prior to the next test.

PFB-4

...__._______.___....,_u._.

A SBWR TRACG Verification t?

TRACG analyses for GIST program do not require verification

- Facility design was verified by Design Review TRACG qualification analyses are engineering design calculations and verified

- Qualification analyses have undergone design review and verification

- Verification records for qualification analyses kept in separate DRF from GIST DRF PFB-5

M SBWR Calculation of Facility Heat Losses ti RPV heat losses are measurable from test data

- Power to keep RPV @140 psig averaged 6.7 kW Heat loss estimated by TRACG comparable to test data

- TRACG calculation of RPV heat losses @ 140 psig.~ 5 kW Test results are not sensitive to uncertainty in facility heat losses

- TRACG runs were made with heat transfer coefficient varied to cause corresponding heat losses of ~2.5 and 10 kW

- No appreciable change in vessel response Heat loss drops as RPV pressure / temperature drop

- RPV heat loss at GDCS injection ~_3.6 kW

- RPV heat input at GDCS injection ~_75 kW PFB-6 l

=%

SBWR Conclusions W

Deficiencies from GIST audit were mainly the result of

. misplacement of test records

- All missing records have been compiled into a supplement to the original DRF

- Heat loss measurements comparable to TRACG estimates Other deficiencies related to GIST were the result of misunderstandings of the GE test process

- Test classification had no impact on quality assurance requirements

- TRACG analyses were not design calculations requiring verification Remaining deficiencies related to errors in TRACG deck for GIST do not impact use of GIST data for design work and are resolved separately QA deficiencies noted by NRC audit team have been resolved for the-GIST program PFB-7

b

~

w GENuclearEnergy Response to NRCFindings on GIST l

November 16,1993 r >

S v.

.s

~. -

......en,.+..e

-ew.,

y s~

m

=.-.,.

r-w--.v,w

..--w14-m

,,w,c.-

-,,s.

r --,, - -

-+-.e-.

y 3.-.

O GE Nuclear Energy Response to NRCFindings on GIST Presentation to the NRC November 16,1993 PF Billig SBWR Test Programs

=%

SBWR

$F

a SBWR M

Response to NRC Findings on GIST Q1 GIST is the only experiment performed related to performance of SBWR Gravity-Driven injection system, and' appears to have serious flaws The following pages show that:

I 1.The flaws identified by the NRC are unimportant or resolved.

2. GIST remains as an acceptable database for computer code qualification.

l.

I.

PfB 2

a SBWR Response to NRC Findings on G/ST W

Q.1.1 Technical adequacy of experiments questioned even before QA review, due to.significant differences in current SBWR GDCS design compared to that used as GIST reference configuration and. concerns about instrumentation and test procedures.

Differences in current SBWR design from GIST referenced design

.do not invalidate the usefulness of GIST data for code qualification.

- All design differences have been reviewed Sensitivity runs at GIST capture effect of significant differences

- Response to Q.5 addre'sses specific design differences All GIST instruments were calibrated and records exist to demonstrate; calibration and accuracy

- GIST Test Program followed established procedures for Engineering Tests.

- Deficiencies identified by audit team were result of misplaced files, not a departure from test-procedures.

PFB 3 o

.m

.m...

.m.

m=.

-m..

2

-.. ~.

.e,~

++ - --

<--*~e--

w-v-

s SBWR W

Response to NRC Findings on GIST Q.1.2 Inspection results indicate significant uncertainties exist in experimental results due to failure to measure / document facility heat losses Heat losses have been determined by GIST data analyses and engineering calculations

-- They agree GIST results insensitive to uncertainty in heat loss calculations Detailed discussion given as response to O.7 PFB-4 m

u..

a-th m

e-e-

m a

a ir ee*--eer r

aewwy gg--

ee m

eer.=

Mese-y an e

oc e

e e

sbtien e we

  1. s um s-9

a SBWR Response to NRC Findings on GIST W

Q.2 Qualification of TRACG for prediction of GDCS performance rests solely on comparison to GIST results All major phenomena for SBWR LOCA covered by at least one test.

- Major new phenomena for SBWR:

- Coupled RPV - containment response

- Gravity -Driven Cooling System

- Passive Containment Cooling System Only phenomena covered by GIST alone

- Condensation by GDCS flow in RPV downcomer Following depressurization with DPV's open and steam flow given by decay heat level, RPV and DW pressures about equal GDCS flow given by elevation head and frictional pressure drop TRACG pressure drop calculations covered by other tests (ATLAS, FR!GG, GIRAFFE, plant data)

PCCS performance confirmed by GIRAFFE Independent qualification of major phenomena for SBWR LOCA PFB 5

a sawn W

Response to NRC Findings on GIST Q.2.1 Uncertainties in GIST data call into question usefulness of comparisons of TRACG to experimental results All significant test parameters were measured in GIST

- All GIST instruments were calibrated and records exist to demonstrate calibration and accuracy Heat loss effects reconciled by measurement and analyses

- Heat losses derived from analyses of pre-test initiation data and heat transfer calculations of facility geometry and materials

- Results are in agreement

- Calculations show test results insensitivity to heat loss

- Detailed response on heat losses given in response to O.7 PFB 6

t a

SBWR W

Response to NRC Findings on GIST Q.2.2.

Errors in TRACG'Modeling of GIST Facility also compromise usefulness of results TRACG deck for GIST facility did contain errors in the elevations Inconsistencies in the TRACG deck affected the calculation, not the data Inconsistencies were not carried over to the SBWR deck Inconsistencies in the.TRACG deck for the GIST facility have been corrected and verified

- Elevations of piping

  • Initial water level *

- Minor adjustments to volumes

- As-built bottom drain line orifice

- Decay heat versus time

- Heat loss Only corrections with significant. impact on calculation PFB-7

~

.... _. ~... - - _,,, _ _... -. _. -

g SBWR W

Response to NRC Findings on GIST Q.2.2.1 Correction of elevation errors resulted in significant reduction (ca. 20%) in GDCS injection flow Previous calculations with errors overpredicted GDCS injection flow for GIST Correction of errors.resulted in improved TRACG predictions.

PFB-8

~ _,.

a SBWR Response to NRC Findings on GIST W

Q.2.2.1 Correction of elevation errors resulted in significant reduction (ca. 20%) in GDCS injection flow Main Steam Line Break Test B01 with elevation errors m E-Oi

.........i.........i.........,..................

Q h

3 00E-ol x

~

CD

~

~

f' 3

.J.........T 0

2 00E-01

~

o

_O 1.50E-D I l

u.

1.00E-01 o

IE

~

~

TRACG g

- 5.00E-02 DATA o

Q O

0 00E + 00 O.00E + 00 1.00E+ 02 2.00E + 02 3.00E + 02 4.00E + 02 5.00E+02 Time (seconds)

GE Proprietary information PFB-9 e

e v

4 m,,*

-+r w

w 7

w v

w e e= w rw'

As SBWR Response to NRC Findings on GIST

  • ?

Q.2.2.1 Correction of elevation errors resulted in significant reduction (ca. 20%) in GDCS injection flow Main Steam Line Break Test B01 without elevation errors s

3 50E-On g,,,,,,,,ig,,,,,,,,,g,,,,,,,,,

,,,,,,,,,g n

E E

3 00E-01

{-

j m

\\

E E

D E

E b

2.50E-01

[--

'.......1--------

g 2-

.s O

2.00E-01 2

5

/

5 go 1.50E-01 E-l a.

=

=

E E

1.00E-01 l

-5

~

E o

E 2

j TRACG j

5.00E-02 E-DATA

-j g

U h

E z

O 000E+00

''''I"!' '''I''''I''''

O.00E +00 1.00E + 02 2.00E + 02 3.00E + 02 4.00E + 02 5.00E + 02 4-NOV-93 T.ime (seconds) 17:08:11 HRS GE Proprietary Information pre.io l

l

A SBWR t?

Response to NRC Findings on GIST Q.2.2.2

Section in code qualification document on GIST

. qualification currently contains same elevation errors

  • Code qualification document issued before detection of errors TRACG deck for GIST facility has been corrected and verified I

l 1

PFB-11 I

l

t a

SBWR W

Response to NRC Findings on GIST Q.3 Since SSAR calculations are based on version of code qualified with incorrect input for GIST, and since calculations did not undergo required verification, their usefulness is questionable GIST data is a minor part of the overail TRACG SBWR qua!!fication GIST. data are not directly or indirectly used to predict SBWR performance GIST provides database to qualify TRACG for SBWR applications TRACG remains qualified for SBWR LOCA application Errors in TRACG deck for GIST have been corrected and verified TRACG runs for GIST are better than before TRACG deck for SBWR completely independent from GIST Elevation errors in GIST deck were not present in SBWR deck 1

PFB 12 Y2 74 W@ --'

TuW78'a r=-'rka~*Aheit gy-e F =t + m-v?-34+1y-h4t-9tN*T

+7m1-T'9 E1M a f7 W w" t*W9hPTtv?9'-WM'4t'"---*s-eqg4gg-g,-

t' d

t'

4w--+E

% rw-W T

4

  • --M e--

T At'1-%* * - * *

"Id

--79

-e we

  • Mw*et4Wd++m--

u-a-

e

  • '*Wie-uN

a SBWR Response to NRC Findings on GIST M

Q.4 No scaling analysis for the GIST facility to demonstrate GIST covered scaled parametric range similar to SBWR An independent scaling analysis has been conducted for all GE SBWR test programs (NEDC-32288)

Conclusions from scaling analysis supports GE's initial scaling assessment

- GIST was a full-scale sector simulation of GDCS performance

- GIST data covers all essential phenomena

- The phenomena and processes present in the tests do not significantly differ from what is expected to take place in SBWR.

- The overall behavior of GIST does not diverge significantly from ey.pected SBWR behavior.

- GIST tests provide sufficiently detailed information to provide an adequate and sufficient data basis for qualifying TRACG PFB-13

a SBWR W

Response to NRC Findings on GIST Q.4.1 NRC assessment found significant scaling distortions An independent scaling analysis conducted for GE found no significant scaling distortions

- Design of GIST is in agreement with a general top-down scaling criteria.

- Overall top-down system effects of pressures and RPV water inventory were captured by GIST Geometric distortions in GIST were due to SBWR design changes and the one-dimensional character of the facility

- Impact of those distortions were judged by GE to be insignificant (Appendix of GEFR-00850)

PFB-14

a sewn W

Response to NRC Findings on GIST Q.5 Facility does not represent adequately (or at all) components and interaction paths in SBWR GISTwas an integrated systems test that captured all significant interaction paths in SBWR

- RPV-to-containment

- RPV internal volumes (upper & lower plenums, top guide, standpipe, etc.)

- Intra-containment (main suppression pool vents, upper drywell-to-lower drywell, wetwell-to-drywell vacuum breakers)

Impact of systems not represented bounded by wide variety of test conditions PFB 15

~

a SBWR W

Response to NRC Findings on GIST i

Q.5.1 RPV internals not represented accurately GIST modelled all significant RPV internals (core, downcomer/ annulus, plenums, etc.)

Internals not impacting RPV performance over range of GIST study were deleted (steam separators and dryers)

RPV core not geometrically identical to SBWR

- GIST core (as well as SBWR core) remains covered for all DBAs

- GIST core need only capture the low pressure blowdown flow characteristics of SBWR GIST RPV simulation supported by independent scaling study PFB-16 i

a SBWR Response to NRC Findings on GIST W

Q.5.2 ICS, PCCS not represented Impact of ICS and PCCS on SBWR performance easily understood and covered by sensitivity studies L

No. lCS ICS aids RPV depressurization w/o loss of inventory =>

increased margin for GDCS

- Sensitivity studies confirm it's conservative to assume no ICS

- Same logic used in current plants to ignore RCIC in LOCA analyses l

No PCCS

- DPVs ensure most steam goes to drywell during LOCA f

- DW pressure controlled by main vents during initial GDCS phase

- Sensitivity studies show small impact on RPV inventory I

response over broad range of containment pressures i

1 L

PFB-17 1

. ~.

. ~.

~

=%

SBWR W

J Response to NRC Findings on GIST Q.5.3 Containment not properly scaled Containment was properly scaled for SBWR at time of GIST design Design change in SBWR containment did not change the key phenomena 1

- SBWR retains pressure suppression type containment with drywell and wetwell regions

- GDCS 3001 moved from wetwell to drywell results in increased availab e driving head for GDCS

- Independent scaling study confirms earlier GE assessment that

. containment design changes did not impact usefulness of GIST-1 data Sensitivity studies at GIST and using TRACG for SBWR show only a small change in minimum water level FFB-18

r?

g SBWR Response to NRC Findings on GIST W

Q.5.3 Containment not properly scaled Containment Pressure Sensitivity Results for GIST (No break tests)

DW Pressure (psig) 0 14.7 GDCS Initiation (sec) 343 244 Downcomer Min. WL (mm) 6400 6830 Containment Pressure Sensitivity Results for SBWR (Main steam line break)

DW Pressure (psig) 25 35 55 GDCS Initiation (sec) 535 510 460 Downcomer Min. WL (mm) 5120 5180 5690

Conclusion:

Core remains covered over broad range of containment pressures PFB 19 -

...._.--.,___...-~.,,...-..-.--:..,-,.__.._;_.-_..-_.._

t a

SBWR W

Response to NRC Findings on GIST Q.5.4 ADS not.modeled as it exists in current SBWR Impact of ADS design change on GIST either negligible or conservative

- GIST SBWR design had 24 SRVs piped to suppression pool

- Current design has 8 SRVs to S/P and 6 DPVs to DW

- DPVs ensure high containment pressure for all LOCAs as well as RPV pressure -> DW pressure

- GIST main steam line break simulations closely resemble all vapor breaks

- GIST-tests included runs with 0 psig in containment; impossible in current design l

- GIST and TRACG sensitivity studies show relative insensitivity to containment pressures I

I

)

PFB-20

...~....

~, _ -. -. _ -.. _. _ _. ~.,... _

a SBWR W

Response to NRC Findings on GIST Q.5.5 Only one non-safety system represented (CRD)

CRD flow' represented the most probable and most limiting non-safety system that would be operating during the LOCA phase studied by GIST

- On low level L2, CRD pump flow would be boosted to maximum rate and inject water in the bottom of the vessel

- GIST confirmed that the inventory increase offsets the sooner L

- collapse of voids in the lower vessel resulting in a higher minimum water level

- Other-injection systems coming in higher in the RPV are less likely to collapse voids below the core-Impact of other non-safety system operation (e.g., LPCI) is well understood PFB 21

.._..u_._.,..

. ~. _.., _

.. ~. -

=.

=w SBWR W

Response to NRC Findings on GIST Q.6 Test matrix did not include feedwater line break, and covered only a very limited range of conditions Impact of feedwater line break bounded by other breaks

- Phenomena in vessel straight-forward and well understood; similar to previous BWR designs

- Water level quickly drops below sparger

- Two-phase level swell caused by depressurization does not reach feedwater sparger FW break behaves like main steam line break Broad range of test conditions were studied by GIST

- Main steam, GDCS, and bottom drain line breaks simulated, as well as no break (loss of feedwater) cases

- Parameters varied included; low pool level, increased &

decreased GDCS flow, low RPV water levels, increased &

decreased DPV area, high & low core power, high & low containment pressures, GDCS injection elevation PFB-22

a SBWR W

Response to NRC Findings on GIST Q.6.1 Test initiated at 100 psig Nothing phenomenologically unique occurs as RPV depressurizes to 100 psig

- Break flow is choked for pressures higher than 100 psig and no feedback from containment exists

- Depressurization is covered by other tests (TLTA, FIST)

- Same range of break sizes and locations

- Initial conditions for tests and depressurization to 100 psig easily modeled by TRACG Tests are adequate to qualify low-pressure depressurization and GDCS performance PFB 23

is

~

sewa W

Response to NRC Findings on GIST Q.6.2 RPV inventory " adjusted" per TRACG calculations to correct for scaling-induced heat input distortions RPV inventory heating by vessel metal mass different in GIST from SBWR

- Low metal mass and low initic; metal temperature for lower plenum lead to small stored heat release for GIST compared to SBWR

- Large metal mass and high initial metal temperatures exist in other test facilities used for TRACG qualification (TLTA and FIST)

- Parametric studies on RPV water level were conducted in GIST tests and the tests were used for the TRACG qualification TRACG simulates the facilities the way the tests were run and the test data are useful for the qualification of TRACG PFB 24

a SBWR t?

Response to NRC Findings on GIST Q.6.3 Test ended when GDCS had recovered core; no data taken in " transition" phase between GDCS injection and containment /RCS recirculation Core never uncovered GIRAFFE captures " transition" phase

- it takes 2-3 hours to drain the GDCS pool

- GIRAFFE started at 1 hr Following depressurization with the DPV's open and steam flow given by decay heal level, RPV and DW pressures are about equal

- GDCS flow given by elevation head and frictional pressure drop

- TRACG pressure drop calculations covered by extensive test data base: ATLAS, FRIGG, GIRAFFE, plant data PCCS operation has very little effect on the GDCS performance if the water level in the GDCS pool is higher than the water level in the RPV, GDCS will flow into the RPV PFB-25

sw sawa W

Response to NRC Findings on GIST Q.6.4 Existing / planned loops (GIRAFFE, PANDA) not designed to model accurately injection or transition phases Injection phase well covered by GIST; no need for other facilities to modelit Transition phase from GDCS injection of initial pool inventory to injection of water supplied by PCCS drainage is covered by GIRAFFE

- GIRAFFE tests start at 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> into LOCA; draining of GDCS initial inventory takes 2-3 hours Time aeriod between conclusion of GIST tests and start of 1

GIRA:FE tests does not represent a " transition" phase

- RPV slowly filling with water as GDCS pool leve19nd RPV level reach equilibrium

- TRACG shows small (>6 psi) depressurization as GDCS flow l

l increases the core inlet subcooling

- DPVs cause RPV pressure to stay equal to DW pressure

- TRACG & GIST sensitivity runs bound expected containment pressures

m SBWR Response to NRC Findings on GIST C

Q.7 Failure to establish heat losses means uncertainties in facility power input can never be quantified Heat losses have been established Heat losses are measurable from test data

- Power to keep RPV @140 psig averaged 6.7 kW Heat loss estimated by TRACG comparable to test data

- TRACG calculation of RPV heat losses @ 140 psig ~ 4.9 kW Test results are not sensitive to uncertainty in facility heat losses

- TRACG runs were made with heat transfer coefficient varied to cause corresponding heat losses of ~2.5 and 9.8 kW

- No appreciable change in vessel response Heat loss drops as RPV pressure / temperature drop

- RPV heat loss at GDCS injection ~_3.6 kW

- RPV heat input at GDCS injection ~_75 kW Pi B-27

t l'

is SBWR Response to NRC Findings on GIST W

Q.7 Failure to establish heat losses means uncertainties in facility power input can never be quantified Facility performance is unaffected by small uncertainties in heat loss

- Heat loss given by conduction.through insulation, free convection heat transfer and ambient temperature

- donduction:

hc= k/d = 0.5/2 = 0.25 Btu /ft hr-F condJctioni-cYnvection 2

- Free convection:

Y 2

h r = 1 Blu/ft hr-F

- Net heat transfer coefficient:

N insulation heatloss

  • 1 1

Vessel wall hc hf 2

= 0.2 Btu /ft -hr-F

- Heat loss dominated by conductivity for-insulation PFB-28 1

=%

SBWR W

Response to NRC Findings on GIST Q.7.1 -

Usefulness of data for quantitative best-estimate code assessment without means to estimate errors / uncertainties is questionable Heat loss has been quantified PFB-29 i

...r-c.-.

-n-

_g SBWR W

Response to NRC Findings on GIST Q.8 GIST program does not appear to constitute an adequate basis, as required by 10CFR52.47, for

- Demonstration of performance of GDCS in SBWR

- Establishment of sufficient data base to quality codes to assess GDCS performance in SBWR These responses qualify the GIST test program 10CFR52.47 (b)(2)(i)(A)(1) states the following:

"The performance of each safety feature of the design has been demonstrated through either analysis, appropriate test programs, experience, or a combination thereof;"

- GDCS performance is a simple analysis validated by tests PFB-30

a SBWR W

Response to NRC Findings on GIST Q.9 Code qualification and SSAR analyses partly based on GIST results must therefore be considered suspect GE has demonstrated that:

- GIST test program followed established procedures for engineering tests

- GIST met its design objectives to demonstrate the performance of a gravity fed cooling system for ECCS and establish a database for code qualification of the phenomena

- Errors found in the TRACG deck for GIST have been corrected resulting in improved performance predictions

- TRACG coc'e qualification based on extension database of which GIST is only a small part SSAR analyses for 98WR performance are not suspect PFB-31

t n

SBWR Response to NRC Findings on GIST W

Q.10 Technical quality of data from GIST is compromised by-QA deficiencies and other design and operational factors Technical quality of data is not compromised by "QA deficiencies"

- Deficiencies uncovered by NRC audit were the re'sult of misplacement of test records.

I

- All misplaced records have been recovered and filed

.GE has demonstrated in responses to earlier findings that the-technical quality of the datajs not compromised by design and operational factors of GIST i

Design changes in SBWR since GIST do not alter the key SBWR phenomena studied by GIST I

l t

PFB-32

~

q

~ _ - _.,. _ _ _. _ _ _

.........a

. ~, _,

r

...-..,.-..._..u_-._

_-3 sewa Response to NRC Findings on GIST

  • ?

Q.10.1 QA deficiencies stem mainly from misclassification of test and GE's failure to comply with its established QA procedures Classification of the test in no way diminished the QA necessary for engineering tests GE did in fact follow its own Engineering Operating Procedure in performing this test. EOP 35-3.00 does not implicitly call for different levels of QA for development and design basis tests. A statement exists that may be interpreted that for development tests, less OA is permissible. However, declaration of a given test as

" developmental" does not preclude that full QA be performed. This was the case with GIST.

GE is reviewing the EOP for potential revision to clarify this situation.

PFB-33

g SBWR W

Response to NRC Findings on GIST Q.10.1

-QA deficiencies stem mainly from misclassification of test and GE's failure to comply with its established QA r

procedures "QA deficiencies" cited by the NRC are the result of m sfiling of i

information

- All GIST design records have been retrieved and properly filed GIST tests followed GE's procedure for engineering tests

- Design reviews were conducted prior to the tests to establish test objectives and approve the GIST design to meet those objectives

- All test documentation was prepared and approved prior to testi.ng to specify the test requirements and the test procedures to follow to meet those requirements

- All test instruments were calibrated to NBS traceable standards prior to testing

- All test data were recorded in accordance with the test procedures with appropriate verification PFB-34

.wo-

,.......mw.

-e<..-*

w,....-.%,*-,,w erw....<,wi m..wwm.c-w.v.m-w,...-,,--,m...,,,-

+.

4w.

-ww m,

,e-

..+-,,.r,.-,..

v-

.4%--

m.-.*

a sews W

Response to NRC Findings on GIST Q.10.2 Deficiencies cannot be rectified by "backfitting" QA on DRF or by rerunning analyses, as GE has committed to do, nor by additional tests in other facilities Established QA procedures were followed during the GIST program, and no "backfitting" of QA is required

- Those records which had been misplaced have been added to the Design Record File for GIST GE has rerun the TRACG analyses with an updated GIST deck and results show better agreement to data than before The uncertainty in the facility heat loss has been eliminated GIST tests remain valid and additional tests in other facilities are not necessary PFB-35

s a

SBWR W

Response to NRCFindings on GIST Q 11 In the future, GE should follow its own internal QA procedures for performing safety-related, design application calculations using developmental (Level 1) computer codes GE agrees PFB-36 1

e r-

+-ysn, eg-,.-

%+--og-.y.,,e, w-

%c,-

i,-

qq-ww g

p-g--w-g y

er

f y

T-p

.y-,

in7-

l.

=4 SliWR i

Response to NRC Findings on GIST t?

r Additional GDCS testing by GE in an integral facility is Q.12 likely necessary to support design certification of SBWR Testing should not be done in GIST facility as currently configured

- - Facility should represent current SBWR design, and include potential interaction paths not contained in GIST, such as isolation condenser Facility should represent current SBWR design, and include potential interaction paths not contained in GIST, such as isolation condenser' GIST is a well conceived and conducted experiment which is one element of the overall program to support SBWR certification No additional testing is required PFB-37 m

m m

m m.

me-em mm

=

e.w

,e.e..e e

~w---

ee e.

=e%- +..

-..t-,

=

.--e

.e=-e

t O

GENuclearEnergy Response to NRC Findings on GIST November 16,1993 l

l l

l

-E-

-.m.e

+.

m

,...,j,..+.e-.

gag--

,,.,w%s.

_am...,,,..,, -,,%h-mee ar

='--d w

we am N-e m

m u

e-m

.a u- +

=

e w-m--.

~

Suminary and Conclusions l

l GISTis a well conceived and conducted experiment which is

{

one element of the overallprogram to support SBWR e

V certification 7

l TRACG is fully qualified to predict SBWR performance.

l I

Point-by-point discussions of each of the items from the 10/4/93 i

NRC meeting are included in the handout Technical backup for the point-by-point discussions has been l

presented this morning i

l All of the'above information supports the conclusion that no additional GIST testing is

\\

required for TRACG qualification or validation of L

the GDCS concept a

.-.-.__,.-m

,,,,..,-~m..

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

.~....m.-

. ~

=

.

Retrieved from "https://kanterella.com/w/index.php?title=ML20058A840&oldid=4476725"