ML19345B788

From kanterella
Jump to navigation Jump to search
Demo Pretest Predictions for Fftf Transient Natural Circulation Tests. Demo Pretest Predictions of Reactor Inlet Flows & Temps During Fftf Transient Natural Circualtion Test, Encl
ML19345B788
Person / Time
Site: Clinch River
Issue date: 03/31/1980
From: Calvo R, Laster W, Planchon H
WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML19345B780 List:
References
WARD-94000-0032, WARD-94000-00321, WARD-94000-32, WARD-94000-321, NUDOCS 8012020510
Download: ML19345B788 (137)


Text

-

3 f,

(

WARD-94000-00321 1

l DEMO PRETEST PREDICTIONS FOR THE FFTF TRANSIENT NATURAL CIRCULATION TESTS H. P. PLANCH0N W. R. LASTER R. CALVO APPROVED 3Y:

y L-j W. J. 56verson, Manrger Performance Analysis b

6k E. ScMmidt, Manager Reactor Plant Analysis a

Prepared for the U.S. Department of Energy Assistant Secretary for Nuclear Energy Division of Reactor Research and Technology Contract: DE-AM02-76CH94000 Task: DE-AT02-80CH94047 Submitted to DOE /CH March 1980 1

WESTINGHOUSE ELECTRIC CORPORATION Advanced Reactors Division P. O. Box 158 Madison, Pennsylvania 15663-a i

((/

6015A-602A-(S1478) i i

TABLE OF CONTENTS Page d

1.0 Ih?RODUCTION i 2.0 TEST DESCRIPTION 3.0 MEASUREMENTS,

4.0 DEMO PLANT MODEL 4.1 Reactor Model 4.2 Reactor Vessel Upper Plenum 4.3 Sodium Pumps and Pressure Drop 4.4 IHX Model !

4.5 IliX Model 4.6 Piping and Pump Thermal Model 5.0 BOUNDARY CONDITIONS 6.0 PRE-TEST PREDICTIONS-BEST ESTIMATE

- 3 7-7.0 DESIGN CASE PREDICTIONS 8.0 ACCEPTANCE CRITERIA 8.1 Post Test Analysis 8.2 Bottom Line Criterion 9.0 RFFERENCES APPENDIX A CALCULATIONS FOR THE TRANSITION TO NATURAL CIRCULATION FROM 100%, 75% AND 35% POWER (BEST ESTIMATE CASES 7

APPENDIX B FFTF UPPER PLENUM MODEL

-102-APPENDIX C THE REACTOR MODEL

-120-1 6015A-602A-(S1478) ii

LIST OF FIGURES Figure No.

Title Page 4-1 Fuel Assembly Group Dynamic Pressure Drop Correlations 4-2 Non-Fuel Assembly Dynamic Pressure Drop Correlations 4-3 Bypass Dynamic Pressure Drop Correlations 4-4 Plenum-2A Model of the Reactor Vessel Upper Plenum 4-5 Pump and Flow Dynamic Simulation 4-6 Secondary Loop Pressure Drop Versus Flow 4-7 Reactor Vessel to Pump Pressure Drop 'lersus Flow 4-8 Pump to Reactor Vessel Pressure Drop Versus Flow 4-9 Check Valve Pressure Drop Versus Flow 4-10 Main Coolant Pump Stopped Rotor Pressure Drop Versus Flow 5-1 Heat Sink Boundary Condition for the 100% Best Estimate Case - Sodium Temperature at the DHX Outlet Nozzle Thermocouple 6-1 DEMO Predicted Primary Flow at the Permanent Magnet Flow Meter (100% Power) 6-2 DEMO Predicted Intermediate Flow at the Permanent Magnet Flow Meter (100% Power) 6-3 DEMO Predicted Primary Hot Leg RTD Temperature (100% Power) 6-4 DEM0 Predicted Primary Cold Leg RTD Temperature (100% Power) 6-5 DEMO Predicted Secondary Hot Leg RTD Temperature (100% Power) 6-6 DEMO Predicted Secondary Cold Leg RTD Temperature (100% Power) 6-7 DEMO Predicted Primary Flow at the Percanent Magnet Flow Meter (75% Power) 6-8 DEMO Predicted Secondary Loop Flow at the Permanent Magnet Flow Meter (75% Power) 6015A-602A-(S1478) 11i

LIST OF FIGURES (CONT'D)

Figure No.

Title Page 6-9 DEMO Predicted Primary Hot Leg RTD Temperature

]

(75% Power) 6-10 DEMO Predicted Primary Cold Leg RTD Temperature (75% Power) 6-11 DEMO Predicted Secondary Hot Leg RTD Temperature (75% Power) 6-12 DEMO Predicted Secondary Cold Leg RTD Temperature (75% Power) 6-13 DEMO Predicted Primary Loop Flow at the Permanent Magnet Flow Meter (35% Power) 6-14 DEMO Predicted Intermediate Flow at the Permanent Magnet Flow Meter (35% Power) 6-15 DEMO Predicted Primary Hot leg RTD Temperature (35% Power) 6-16 DEMO Predicted Primary Cold leg RTD Temperature (35% Power) 6-17 DEMO Predicted Secondary Hot Leg RTD Temperature (35% Power) 6-18 DEMO Predicted Secondary Cold Leg RTO Temperature (35% Power) 6-19 Primary Pump Coastdown from 100% Power and Flow 6-20 Secondary Pump Coastdown From 100% Power and Flow 6-21 Primary Pump Coastdown from 75% Power and Flow 6-22 Secondary Pump Coastdown from 75% Power and Flow 8-1 Acceptance Limit on Flow for the 100% Power Case 8-2 Acceptance Limit on Flow for the 75% Power Case 8-3 Acceptance Limit on Flow for the 35% Power Case.

6015A-602A-(S1478) iv

LIST OF TABLES Tacle No.

Title Page 3.1 Temperature and Flow Measurements for the FFTF Steady State Natural Circulation Tests 4.1 Reactor Vessel Upper Plenum Model Dimensions -

5.1 Fuel Assembly Decay Heats 5.2 Non-F!'el Assembly Decay Heats,

l V

6015A-602A-(S1478)

ACKNOWLEDGEMENT i

The pretest predictions in this report are the result of a long tena task to gather data for and codel FFTF on the DEM0 Code. A large number of people have conributed significantly to this effort. The advice from Mr. W. J.

Severson ( ARD) and support from Mr. S. L. Additon and his staff at HEDL was particularly significant. We also wish to acknowledge Mr. J. C. Reese for calculations of the Reactor Plenum Temperatures with VARR II and TEMPEST, Dr.

l Paul Howard (ANL) for providing the PLENUM 2A model, Dr. Y. S. Tang, Dr. R. D.

1 Coffield, Dr. T. L. George (PNL) and Dr. R. Racco for assistance with the DEMO core model, Mr. C. V. Walling for assisting with the computer programming and providing the plots for the report, and the Word Processing Staff, and Mrs. D.

Oshnock for typing the report.

i i

l l

l 6071A--602A Vi

1.0 INTRODUCTION

Tnis report contains DEMO (Reference 1-1) pre-test predictions for the transient natural circu.ation tests to be conducted in the Fast Flux Text Fa cil i ty.

Calculations of transient flows and transient temperatures in the reactor and the primary and secondary heat transport loops are presented.

Tr.ese pretest predictions were made as part of the Breeder Reactor Natural Circulation Verification Fragram. The pcrpose of the progran is to detemine the validity of DEMO for whole plant, breeder reactor, natural circulation analysis.

Post test cocparisons of the f1ows and temperatures measured in FFTF to the pre-test predictions in this report are an important part of the DEMO vtr if ication.

Subsequent to completion of the FFTF Natural Circulation Transient Tests, it is planned to issue a report making comparisons of these pre-test predictions with the FFTF test data. The logic to be used in the post test comparisons and the criteria with which the adequacy of the DEMO models and the data used for the predictions will be judged are given in Section 8 of this report.

The DEMO verification is part of a coordinated effort to verify a series of codes for breeder reactor natural circulation de ;y heat renoval analysis.

DEMO will be verified f or whole plant analysis, COBRA f or whole reactor analysis, and FORE-2-M f or fuel pin analysi s.

The principal variables of interest from the DEMO analysis are the total primary heat transport er+em fim available for cooling the reactor and the reactor inlet temperature. The reactor flow and inlet temperature are input variables to COBRA. COBRA is used f or an overall Reactor Thermal-Hydraulic analysis. COBRA provides thermal hydraulic boundary conditions to FORE-II-M.

DEMO also predicts temperatures for lumped regions of the core (fuel, non-fuel, and bypass regions are calculated for FFTF) and hot-spot tenperatures for selected fuel-coolant locations; however, COBRA and FORE-II-M are relied upon for the final values of these temperatures.

The FFTF Transient Natural Circulation tests that will provide data fer the code verification are planned as part of FFTF's Acceptance Test Program.

In these tests the reactor will be scrammed (causing the sodium pumps to be

-I-6071A-- 602A

tripped) with the pony motors on all six pumps turned off. The plant will then undergo a transition to natural circulation flow.

Measurments of flows and sodium ter:peratures will be used to monitor the test and verify the The capability of FFTF to rmove decay heat with natural circulation cooling.

measurerrents will also be used to verify computer codes for natural circulation analysis. A detailed discussion of the tes.s and the instrtsnentation that has been incorporated into FFTF to provide the necessary data and accuracy for the natural circulation conditions are contained in Sections 2 and 3 of this report.

I I

6071 A--602A __-____

2. 0 TEST DESCRIPTION A series of transient natural circulation tests will be conducted as part of the FFTF accept ance tests.

The transient natural circulation tests are described in Hsiford Engineering Developnent Laboratory (HEDL) Test Spec ification TS S A008, Revision 2 (Ref erence 2-1).

Excerpts from this test specification, relating to and necessary for the DEMO code verification are reiterated below.

The test specification covers tests of the transition to natural circulation cooling f ran a series of steady state at-power initital conditions. All the tests will be initiated with a reactor scram and mair coolant pump trip. The tests as described will include:

A) A transition to natural circulation in the primary loop f rom 5%

reactor power, 7S% primary loop flow. The secondary loop pumps will coastdown to 10% speed at which time pony motors engage and provide 10% secondary loop f orced flow.

B) A transition to natural circulation in both primary and secondary loops from 35% reactor power and 75% flow.

C) A transition to natural circulation in both primary and secondary loops from 75% reactor power and flow f ollowing test subset B, and D) A trcnsition to natural circulation in both primary and secondary loops from 100% reactor power and flow.

The parts of this testing that are of direct interest for DEMO code verification are the transients run f rom the three nonnal power points, i.e.,

100%, 75%, and 35% reactor power--test subsets, 8, C and D above.

The plant conditons at the beginning of subset 8 of the natural circulation tests are specified as f ollows: 6071 Ar-602A

1) Reactor Power at 35% +1% of 400 Mw.
2) Primary loop flow at 75% +1% of 13,443 GPM.

0

3) Primary cold leg temerature at 630 +5 F.

0

4) Primary hot leg terperature at 750 +5 F.
5) Secondary loop flow at 75% +1% of 13,200 GPM.

O

6) Secondary cold leg temperature at 6T F.
7) Decay power between 1.4 and 1.7% of 400 Mw.
8) All six heat transport system (HTS) pony motors de-energized.

The transient will be initiated by scrarming the reactor, which causes a rapid control rod insertion and a trip of all six HTS sodium purps. The pumps will coast down to zero speed, and the plant will undergo a transition to natural circulation flow. The peak core temperatures will be moni tored by neasuring Fuel (ben Test Assembly (F0TA) temperatures. Eight tenperature channels are displayed in the control room. The test will be continued until tre primary 0

hot leg and secondary cold leg temperatures zre within 50 F, at which point the primary and secondary pony motors will be started.

Subset C (f rom 75% power) of the tests will be conducted after subset B of tne testing has been successfully completed.

The plant conditions specified for tre beginning of subset C of the tests in cl ude:

1)

Reactor pover at 75% +1% of 400 Mw.

2) Primary loop ficw at 75% +1% of 13,443 GPM.

0

3) Primary cold leg temperature at 662 _+5 F.

l

4) ertmary tot leg temperature at 920 +5 F.
5) Secondary loop flow at 75% +1% of 13,200 GPM.
6) Secondary cold leg temperature at 602 F.
7) Reactor decay power to be 3 to 3.5% of 400 Pw.
8) All six heat transport system (HTS) pony motors de-energized.

l As with the subset B tests the transient will be initiated by scrcrming the reactor, c1usi ng a trip of all six HTS sod ium punps. The pumps will coast i

- 4' 6071A-- 602A l

down to zero speed, and the plant will undergo a transition to natural circulation flow.

The peak FOTA temperatures will be monitored by means of the eight tempmature channels displayed in the control roon. The test will be continued until tte primary hot leg and secondary cold leg temperatures are within 50 F, at wnich point the primary and secondary pony motors will be st arted.

The subset 0 (from 100% power) tests will be conducted after successful completion of the subset C tests.

The plant conditions at the beginning of subset D of the tests include:

1) Reactor power at 100% +1% of 400 Mw.
2) Primary loop flow at 100% +1% of 13,443 GR4.

0

3) Primary cold leg tenperature at 680 +5 F.
4) Primary hot leg temperature at 938 +5 F.
5) Secondary loop flow at 100% +1% of 13,200 GPM.
6) Secondary cold leg temperature at 602 F.

7)

Reactor decqy power to be 4 to 4.5% of 400 Mw.

8) All six heat transport system (HTS) pony motor de-energized.

Here again, the transient will be initiated by scranining the reactor, causing a trip of all si x HTS sod ium pumps. The punps will coast down to zero speed, and the plant will underp a transition to natural circulation flow. The peak F0TA tempratures wil' be monitored by means of the eight tenperature channels displayed in the control roon. The test will be continued until the primary 0

hot leg and secondary cold leg temperatures are within 50 F, at which point the primary and secondary Dony motors will be started. 6071A--602A

3.0 MEASWEMENTS The test data most significant to DEF0 verification are measurements of primary and secondary loop flow and primary and secondary hot and cold leg sodium temperatures. These measurements will be compared directly to the DEMO pre-test predictions. The instruments to be used for these measurements, the maximum uncertainty for each instrument, and the expected instrument time constants are listed in Table 3-1.

The accuracies shown in the table for the

. 'sistance themometers correspond to test specification TS-51-4A009 Rev.1 (Ref. 3-1) requirements for precision measuring equipment with +5 F maximum uncertainty. The FFTF Instrumentation System

.;gn Description, SDD-93 (Reference 3-2) requirements for the RTD's specify repeatability and 1000 hour0.0116 days <br />0.278 hours <br />0.00165 weeks <br />3.805e-4 months <br /> drift of less than 3.5 F for the instruments. Measurements ta<en at isothemal conditions prior to the steady-state tests will be used to quantify measurement bias between the hot to cold leg RTD's, thus reducing errors in the inferred AT.

Therefore, the maximum error in reactor 6T is estimated to be les s th an 5 F.

The +2% r.aximum flow error is based on the errors in recent pulsed neutron act'vation measurements of flow in the FFTF primary and secondary loops. The cvvs of magnetic flow meter's voltage response versus measured flow, constructeu during these calibrations, will be used to obtain flow data during the natural circulation tests. The values for the instrument time constants are the comb:ned values for the sensors (estimated by FFTF designers) and the instrument train through the data handling system (specified in SDD 93). Powr history prior to the test is required to calculate the post-trip decay power.

The following data are to be acquired by the DDH&DS for the duration of each test. The required scan interval for each parameter is specified following the parameter. The feature of the computer that increases the scan interval for selected parameters for 30 seconds following a scram is not to be used.

o Reactor fission power (as provided by all three ranges of the Wide Ran e Nuclear Power indication and eventually by the LLFM

- 1 sec. 6071A--602A

TABLE 3.1 TEMPERATURE AND FLOW MEASUREENTS FOR THE FFTF STEADY STATE MTURAL CIRCULATION TESTS Expected Time Sensor Constant Measurement Sensor Designator Accuracy (Seconds)

Primary Hot leg Temp.

RTD T-2X001 150F 5

Primary Cold Leg Temp.

RTD T-2X005

_6cF 5

Secondary Hot Leg Temp.

RTD T-2X534

+50F 5

Secondary Cold Leg Temp.

RTD T-2X550

+50F 5

Primary Loop Fim FM FM F-2X011

+2%*

1 Secondary Loop Flow PM FM F-2X5485

+2%*

1

  • Accuracy is in % of reading l

l

~~

6071A--602A

1 i

i o All fuel, absorber, and reflector assembly exit temperatures -

10 sec.

o All fuel and reflector exit flows - 11 low flow PSD ECFM (Phase Sensitive Detector Eddy Current Flow Meter) flows are to be sampled at 1 second intervals. These are in positions 1201, 1202, 1703, 2101, 2201, 2202, 2610, 3508, 3606, 3610, 3707. All others are to be sampled at 10 second intervals.

o Temperature channel from PSD output of 1202.

j o The primary hot leg temperatures (RTD's) - 10 sec.

o The secondary hot leg temperatures (RTD's) - 10 sec.

o IHX primary outlet temperature - 10 sec.

o DRX module outlet temperatures - 10 sec.

o Primary and secondary loop flows - 1 sec.

o Primary and secondary pump speeds - 1 sec.

o Reactor Overflow Tank (T-42) temperature - 60 sec.

o Reactor cover gas pressure - 10 sec.

o Secondary loop cover gas pressure - 10 sec.

o Reactor vessel level (from the three PPS channels and the 5 ft. RV operating level probe) - 10 sec.

o Primary pump tank level - 10 sec.

o Secondary pump tank level - 10 sec.

o Secondary expansion tank level - 10 sec.

o Reactor overflow tank level - 63 sec.

l DHX fan speed, inlet vane, and fine and coarse control damper o

positions - the four DHX modules in loop #1 unit are to be on one second intervals. The other two DHX units (eight DHX modules) are to be on 60 second intervals.

Nine exit temperatures from TLLM's(l) (Temperature, Liquid o

Level Monitor) - 10 sec.

(1)These temperatures may be recorded by separate recorder if these channels cannot be recorded by the DDH&DS.

I 6071A--602A __

o Nine intermediate level temperatures from TLLM U-355 - 10 sec.

o Three temperatures from PTP (Proximity Test Plug) - 10 sec.

The instrumentation for the two F0TA assemblies shall be recorded continuously on magnetic tape or similar medium at a maximum scan interval of I second during the test. The channels to be recorded includes the T/C's, eddy current flowmeters and the additional low flow PSD channel for the flowmeters.

The instrumentation for the A0TA and VOTA assemblies shall be recorded continuously on magnetic tape or similar medium at a maximum scan interval of 10 seconds during the test. The channels to be recorded include the exit T/C's, eddy current flowmeters, and the additional low flow PSD channe; ?or the A0TA flowmeter.

Pump coastdown time for each pump shall be recort!ed.

a l

t

~~

6071A--602A

l e

4.0 DEMO PLANT MODEL i

t The DEMO computer code as set up to model CRBRP was used as the basis for the FFTF model.

The steam side of the plant was deleted, and a simplified DHX model was added. Geometrical changes were made to the reactor, IHX, pumps, and piping to represent the physical characteristics. Wherever possible, the same correlations and modelling techniques were used in the FFTF and CRERP versions of DEMO.

l A brief description of each component of the DEMN model of FFTF is given below. A discussion of the reactor post-shutdown power calculation and the i

DHX are included in Section 5.0, since they are not inherently a part of the DEMO model, but are used as boundary conditions in the DEMO calculations.

t 1

4.1 REACTOR MODEL i

The DEM0 Reactor Simulation includes models for 1) reactor neutron kinetics and decay power, 2) therma".-hydraulic models of the core and surrounding regions and 3) thermal-hydraulic models of the inlet and outlet plena.

The standard DEMO medels for reactor kinetics and associated Doppler and sodium density feedbacks were used. These models are described in Ref. 1-1.

The neutronic data was taken from Ref. 4-1.

Decay power as a function of time is input to DEMO in tabular form and is discussed further in Section 5.

The inlet plenum for these pretest predictions was represented as a 1 r. ode fully mixed fluid volume thermally coupled to the su'rrounding structure.

The t

upper plenum thermal behavior was calculated with the VARR II and PLENUM 2-A (MOD) codes as described in Section 4.2.

The core / bypass region which takes flew from the inlet plenum and discharges sodium to the upper plenum is represented by three thermally independent models which are hydraulically coupled at their boundaries - the inlet and outlet plena. The thre7 separate models through the reactor are called the 6071A--602A

~ ~

fuel, non-fuel, and bypass regions or channels. The fuel channel includes all the fuel assemblies and the fuel open test assemblies. The non-fuel channel includes the inner and outer reflectors, the radial shields and control and shim assemblies. The bypass channel includes the vessel liner and the in-vessel storage.

Dynamic flow redistribution between regions is calculated. At each time step the f raction of total reactor flow through each channel is allowed to vary until the total pressure drop across each channel is equal. The total pressure drop in each channel is the sum of the dynamic pressure loss and the static tnermal head. The thermal head is evaluat'd from the sodium temperatures through the channel and the channel e'avation. The dynamic pressure drop is the sum of the friction and form losses through the channel and is a function of the channel flow rate. The static head portion of the pressure drop is calculated directly with the DEM0 thermal model by lumping all the heat capacity of each assembly in the region together into one assembly. The thermal behavior of the DEM0 channel thus corresponds to an average assembly.

The dynamic portion of the pressure drop was more difficult to determine especially in the non-fuel region where channels cf widely differing thermal and hydraulic characteristics are lumped together. DEMO inputs the dynamic pressure drop as a function of flow f rom comparison with the more detailed whole core COBRA model. The COBRA model uses 37 different channels for flow redistribution and also calculates heat transfer between the regions. Further details as to how the correlations were developed will be given in Appendix C.

The difficulty in developing a dynamic pressure drop correlation for a group of channels is that, while the total pressure drop for the region is constant, each individual channel has a dif ferent dynamic and static portion.

The best method found for the evaluation of the dynamic pressure drop was to calculate it using the total pressure drop (constant for all channels) and the static pressure drop (evaluated using mixed mean temperatures for all the regions) both of which can more clearly be calculated for a lumped group of channels. The DEMO dyanmic pressure drop correlations were determined by the l

following method:

, 6071A--602A

i

~#

DYN TOTAL STATIC DEF0 COBRA DEF0 By using the DEF0 static pressure drop and the COBRA total pressure drop this method assures that the total pressure drop between DEMO and COBRA will be consistent with small differences in the static head calculation being compensated for in the dynamic pressure drop correlation. The dynamic pressure drop correlations for each channel are given in Figures 4-1, 4-2 and 4-3.

These correlations are data input to DEF0 for the transient calculations.

A description of the thennal modeling of the three DEF0 reactor dannels is given belew along with the channel elevations used in the static pressure drop.

The DEFD fuel assembly channel (illustrated in Figure C-3) is divided into 3 axial sections representing the active core, upper fuel (fission gas plenum) and the flow tubes.

The active core section models the 36 inches of fuel in the a'tive core as well as the 6.5 inch upper and lower reflectors adjacent to the fuel to the core. The active core is represented with 5 axial and 5 radial nodes with 2 additional axial nodes for the inlet and outlet reflectors. The 5 radial nodes include inner and outer fuel, clad, coolant and duct. The dimensior.s used for this case are based upon an average fuel assembly. All of the power generated in the fuel assemblies (96.7% of the total power) is assumed to be generated in this region with a citopped cosine power distribution.

Above the upper reflector there is a 42 inch fission gas plenum extending to the top of the fuel pins. The upper fuel region of DEMO models this region with 5 axial-3 radial nodes. The radial nodes represent the clad, sodium and duct. The dimensions for the clad, duct and sodium flow area are unchanged in this region fran the active core. Lumped in with the clad however is additional metal mass to represent the spacers and spring which are located inside the cladding tube.

The last axial region in the DEF0 fuel assembly model represents the hardware from the top of the fuel pins to the top of the flow tubes. This includes the 6071A--602A

+

+

l-/

j

//

//

iI t

ll l

5 l/

s i

H mg t

j g

8 I,

a

g' e

1r-l

?

I M

M

,/

h m

~

5 2

/

P$

h e*

l

/

/

ss 8

,/

ll 8 O

g

/-/

5,2,

~

N

>/<

m c

a f

.g

?

O u

g l

no @

=

l h

o I

8._

0S lh0LF.MRON/P ATLED - FFEOC PORD ERUSSERP o

-lE~

-h

-4 d1S-21 h

+

i i-c:

8 r

2_...__

pu.

i i

1 4

h' y ;

=

,W-me-w_-gi.

l I

i l

r

,l i

?

t_

f i

i 4

1 I i f

,I i

i i

i l

i

..15

,i l

l I

}

g l

l h

e L

BE T

l i

)

i.

3 l

i j

i 3

1 i

i i

g 4

I.

g i

}

i

=

I i

i l

g i

e 4

O i

i

}

C_

P^

p-1 1

i

.o p, d r

t 1

6 l

i

/

pf v

ki i

I i

i k

l4g 6

u 1

i i

/

/

f f

,I

.(

dg i

i f

f d

1 j

g e

r E

l k

hl l'l i

i

/

l 5

4

~

3 1

i s

i l

w 2

i l

l Q

g 55 l

ff

?._

4 a

ss

=

1 i

i t

/

i My u

{

}

t r

i a

f4

~'

i

/

(

Z i

y 1

j t

Oh i

{

d 3

g 6

O.

O 8

m

  • l.$ hJld liSON/d Yll33 ' '53303 dOd3 3d3S$3dd

+

+

=lb

+

+

M I

=

EP

=9 8

S a

eA R a

P@$

C i,m 5n O

E 1,

4 4

~x bb E s

44 w

$$ r

+

Y o

9

~

55 5 l

II m 6

l 3*

O e

_9

-8 l

.QS WOLF.RP0N/P ATLED

.FFEOC PORD ERUSSERP o

l

+

-tS-

~

fuel assembly handling socket a small gap and the flow tubes. All the metal and sodium in this region is divided in 5 equal size axial nodes.

The ou..de wall of the chimneys are assumed to be adiabatic, but because they are thin walled (.1 inch), heat transfer with the surrounding sdoium is modeled by increasing the metal heat capacity to include all of the sodium which is inmediately above the core surrounding the flow tubes. The total elevation for the flow tube region in DEMO is 49.4 inches.

The elevation of the DEMO fuel assembly channel elevation is 231.90 in, above the centerline of the inlet nozzle. This includes an elevation of 33.75 inches for the unheated section of the fuel assemblies and 57.75 inches for the inlet plenum. The fuel assembly channel exits 13.9 inches above the centerline of the outlet nozzle.

Therefore the static pressure drop portion of the total nozzle to nozzle reactor pressure drop must subtract the static pressure drop for the 13.9 inches of the upper plenum assumed to be at the upper plenum temperature as calculated by VARR-II or PLENUM-2.

The DEMO non-fuel channel lunps together the safety and control rods, the inner and outer reflectors, and the inner and outer shielding. Unlike the fuel assembly channel this section models together assemblies with varying geometry all of which have low power. The non-fuel channel is divided into two axial sections. The lower section models the non-fuel assemblies below the top of the active core (the first 80.95 inches). All of the heat capacity for this region is lumped together into 7 equal size axial nodes and 2 radial nodes representing sodium and metal. 3.3% of the total power is assumed to be generated in these assemblies with the same axial power shape as in the fuel.

The non-fuel assemblies extend 55.82 inches above the active core. Again all of the non-fuel assemblies are lumped together into 5 equal size axial nodes with 2 radial nodes to represent sodium and metal. This section ends at the inlet to the flow tubes. Although the inner reflectors have flow tubes the majority of the flow through the non-fuel region exits directly into the plenum. Because of this, the flow tubes in this region were neglected. The i

elevation at the top of the non-fuel assembly channel is 194.52 inches above 6071A--602A.

-- g

i J

the centerline of the inlet nozzle or 23.48 inches below the centerlir.c of the outlet nozzle. In calculating the no7.zle to nozzle static pressure drop the additional 23.48 inches is assumed to be at the upper plenum temperature.

The DEMO bypass channel represents the in-vessel storage and the vessel liner. This section is assumed to be unheated and is modeled as a large on node mixing volur.1e.

Because of the long transport time through the bypass it is not expected that its temperature will vary much during a 1000 second transient. The bypass is assumed to end at the horizontal baffle. This gives 4

the elevation of the top of the bypass channel to be 182.5 inches above the inlet nozzle or 35.5 inches below the outlet nozzle. For the static pressure drop calculation this additional height is calculated to be at the upper plenum temperature.

4.2 REACTOR VESSEL UPPER PLENUM During the development of the DEM0 model, it was found that a one-node '.otally mixed plenum model was not adequate for the purpose of calculating the pre-test predictions. Detailed analyses of the upper plenum using two dimensional mixing codes (VARR-II, Reference B-2, and TEMPEST, Reference B-3) have shown that the outlet nozzle temperature transient was much more seve;e than that predicted with a one node perfectly mixed plenum model. Temperature transients as high as 200 /sec. were noted in the detailed analyses, with F

outlet temperatures as much as 700F below that predicted with a one node model.

Appendix B contains a detailed description of the analyses done to describe the upper plenum performance. The end result of the analysis was an upper plenum model based on the PLENUF-2A code (Reference B-1), and modified to match the VARR-II results for tht first three minutes of the transient. The PLENUM-2A (modified) model was u ;ed for design case predictions an.d VARR-II was used for best estimate pred'ctions.

s The PLENUM-2A model is a generalized two-node plenum mixing model developed from tests done at Argonne '4ational Laboratory. Figure 4-4 shows the general

-I ~

6071A--602A l

1

i a

TABLE 4.1 REACTOR VESSEL UPPER PLENUM MODEL DIMENSIONS I

D1 - Reactor vessel inner diameter 19.64 f t Z1 - Reactor vessel sodium level * ~

- #16 ft (varies during transient) 02 - U1S diameter 4.195 ft Z2 - Core flow tube height *

- 3.035 ft-I D3 - Outlet nozzle diameter

- 2.271 ft 1

Z3 - Outlet nozzle centerline height

  • 1.958 ft D4 - Core flow area equivalent diameter - 3.7 f t t

i

  • All heights measured frcm the top of the horizontal baffle.

i l

l 1

i I

l 6071A--602A ' I i

-e,,w

.,c,

-, -, - - - - = - -.,

-r n +-

m wr-

~

e

-w

i

=

_og

=

"m J

V2(VOLUME 2)

DENSITY INTERFACE T2 WARM FLUID

/

Zl (HOT-COLD INTERFACE) k

[ dN Tl[p 1

EXIT N0ZZLE VELOC (10F 3)

ZB F

N km j

V /_

-i D3 1 TOUT i

9 (VI(VO aN y

z2 oa M E 1) v o

/

COOL ENTERING JET

( TINLET )

i i

i FIGURE 4-4 PLENUM-2A MODEL 0F THE REACTOR VESSEL UPPER PLENUM e

i ccnfi guration of tN upper plenum model. The model is initialized with the entire plenum treated as a totally mixed one node region. When the reactor scrams and the pumps start coasting down, the fluid velocity and temperature

]

entering the plenum rapidly f all. When the jet height (detemined by the Richardson nteber, tM ratio cf the gravit ational to t he inertial f orces) drops below a predetermined value (the p'enum height for %e normal PLE?.UM-2A, f our feet for the version modified to match VARR-II), the moct ' switches to a two-node mi xing model. The interf ace layer is initially located at the exit height of the core flow tubes (ZZ in Figure 4.4) and all the flew leaving the i

core is assumed to enter the annular regicn below the flow tube exit. The flow entering this icwer region is integrated over time, and when its volume i

is 1.5 times the voline of the mixing region, tre interf ace layer starts i

f rising at a rate dctemined by the froude number. The size of the Icwer mixing

]

region then expands as more of the opper region fluid h mixed with it due to j

the rise of the interf ace layer.

i j

This model was used for the design case runs only. This was done to avoid a prolonged and unnecess ary detailed analysis with VARR-Il for each design case. The modifications done to PLENUM-2A to make it match VARR-II were based

)

upon the 100% initial power best estimate transient, and were theref ore only correct. for this one case. Even so, comparisons made with this 100% pcwer i

modified PLENUM-2A model against "ARR-II for the 75% and 35% initial pcwer l

model showed good agreement in the short term, with only slight dif ferences in calculated flow noted (see Appendix B). The modified PLENUM-2A model was I

l therefore adequate for the design case runs since it agreed well with VARR-II i

in tre s hort tem and predicted lower temperatures and thermal heads than i

VARR-II in the long tem.

l 4.3 SDDIUM PUMPS AND PRESStRE DROP The OEM0 pump and loop hydraulics model is shown schematically in Figure 4-5.

The figure is strictly applicable to the DEMO model of one secondary loop i

which contains one pump and one flow node.

In the primary loops, the cover gas pressures and free surf ace levels in the reactor vesse' and ptnp tank are l

[ 6071A--602A

l l'tR,1P M0TiO N N

dN 19V(~ " rp + rg + r m (SPEED) is s '.

n r p' rg y l ORRELAT10f4 390 C

h =hp (0/N) p FHICTION T01;dUE d 0 CORREL ATION

,e;.I!

rg = rg (N)

~

PUfJP h

liEAD P POMP!NG TORullE CORRELATION

.00P FLUID MOTION FLOW r' = rp (Q/N)

M U*

If

=h

+hp + APg p

i DATA TiEQUIP.E0 l

l PUfl.P INE RTI A, I, (C0 ASTD0VJfl TESTS) p FRICTION TORQUE CORRELATION, rg (ff), (COAS100WN TESTS)

"!!f/?tNG TOROUE CORRELATION, rp (G/N), (PO't!28 FLOW-SPEED r.1AFS)

HEAD CORCELATION, hp (0/0), (HEAD FLOV!&EED MAPS AfiO STOPTED R0 TOR AP)

TLill3 PRESS'JilE DROP CORRELATION, AP. (CCMiONEf;T FI OW TESTS, STAN0ACD CORitEL. [10NN g

GE07.WTRY & S00illt.1 DENSITY FOR f1ATilRAL IIT.AD CALC 'LATION, b3 FIGURE 4-5 PUMP AND FLOW DYNAMIC SIMULATION.

. --- =. -.. ~. - _. _

[

4 i

)

toupled to the fluid motion equations. Also the fluid equations for each of I

l the two modeled loops are coupled in the reactor inlet and outlet plena. As i

the figure shows, friction torque and pumping torque are calculated as a function of speed and flow. The calculated pump speed and flow are the basis a

for the, mad calculation from the pmp head correlation. Flows are calculated i

as functions of the ccupied pump head, thermal head, and fluid friction and and form losses.

t The DEM0 equation of motion for each pump is:

I

,\\

2 IN Ip+Tf+T

=

6Cg i m

j where N is the fraction of rated rotational speed N, and T, T, and i

R p

7 T are pumping torque, friction torque, and notor torque, respectively, m

normalized to reft-oce torque T.

I is the p;mp rotational inertia. The R

pmp is coupled to the fluid through the pumping torque and through the pump I

head-flow-speed correlations. Data fcr these correlations and the emp inerti a, were derived f rom ext 6nsive pum tests conducted prior to the pump's installation in FFTF (Ref 4-2).

The loss torques modelled in DEM0 account for motor windage, friction between f

4 1

the pump shaf t, and surrounding fluid and 'oearing losses. These are I

j determined experimentally in such a way that the total of the pumping torque and loss toroues is accurately represented for a pmp coastdown. Thus if the pump torque departs f rom Q/N similarity at low speeds because of non-similar hydraulic inefficiencies the torque dif ferences wou 1 be accounted for in the f r ict f o'1 torque terms.

The head-flow spead correlations for the FFTF pumps are as follows:

Primary Ptrnp I

1. 7794 + 0.170395 h - 0.2235(M (h) - 0.12654 (h)3 +0.020458(h)

=.

i 6071A--60?A

~

i i

~,m.

--_,,c,--,

.,. ~ - -.. - -

.. _ _ _. =. _ _ _ _ _

l I

[

i 4

l Head (ft) h = buu (f t )

l

p., Speed (RPM) 1110 sRPM) g, Flow (GPM) 14duu (GPM) l Secondary Pump 1.2265 + 0.3347 h - 0.4717 (h)2 - 0.025033 (h)3 + 0.00739 ( f 4

=

Head (f t )

h = 400 (f t) 4 N = Speed (RPM) 1110 (RPM) q, Flow (GPM) 14buU (GPM)

The pumping torque correlation, which is valid for both the primary and

{

secondary pumps is as follows:

0.43336 + 0.70175 h - 0.21684 (h)2 - 0.5889 (h)3 + 0.30622 (h)4

=

.07663(h)5+.00736(h)6 1

where T=T GPM 0

  1. 6071 A--602A i

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

i

__ _ _ _ _ m- -

l i

i The ef fective pump rotational inertias detemined fran coastdown data were l

2 2

16,728 lb-ft for the primary pump and 8368 lb-f t for the secondary pump. These values were used in the best-estinate transient predictions. The inertias specified for the design of the pump in the Systen Design Description 2

2 (Reference 4-3) were 14.000 lb-f t for the primary pump and 6500 lb-f t for the secondary pump. These values wre used for the design case pretest l

predictions in Section 7.

I The pressure crop data is input to OEK) as correlations of AP as a f unction of mass flow rate.

These correlations are based on test data taken for specific j

j canponents during their design or design verification and on standard AP l

correlations for other components such as piping.

[

Correlations for main heat transport systen pressure-drops are as follows:

I

1) The secondary loop (Figure 4-6).

This is a lumped AP vs. flow for the IHX secondary side, piping and fittings and the DHXs.

[

f

2) The reactor vessel to primary pump flow path (Figure 4-7).

This is a i

lumped correlation for the reactor exit nozzle and the piping and pipe fittings.

^

j

3) The primary pump to reacter vessel flow path (Figure 4-8).

This j

lumped correlation models tne AP of the IHX, piping and pipe fittings.

5 i

4) The primary check valves (Figure 4-9).

f l

1

5) The main coolant pump stepped rotor AP (Figure 4-10).

l Tne development of these correlations is discussed in detail in References 4-2 l

and 4-5.

i-i i

j 6071A--602A

-2 4-

- =

+

+

0 l

t l.)

~

l T

//

eC 9l s

ie/

~

R

/

n E

H i

e e

I u

m 5

a e

/

b 5

./

=

m MDR a

l U

e

/

W i I 2 f' El 8

^

g M

Sg6 "

s 3

/

sww e

/.

/

5 11 o l

//

Ilei

/

i

..e a

/

..o.,,

a

,/

w E

N

/

eC

/:/

d

/

~'

e N

b O

e

+

+

+

+

k i

i l'T i

l

~~

i l

f I

4 s

?

9 cc a

N x

3 c

i

~

l x

a e

d n

5 V

a w

A L

w e

u a

u a w

m i

p:

w k

wE I

t

.O M

?

3 $

\\

il

?

ll 0llh$ll E

22 O

i jl l.

d a

da M iI ll h

ss J.:

x i

EE I'

u r4 l

^

u

/

t4 7r r

l' u"

i 0

i o

R d

d 2

'00 M'H80N / d Vi*130 add 300 dGO ~.MGS3Bd

+ - - -

EM E l

100.

l t

N

?

s 5

(

.s, N

N N.-

n.

N s.

M N

-.s f

s N

~

-e g

7 g

u Q.

I E

E ll 10.

0.01 0,10 I.00 FLWi - NORMAi.IZED (REFERENCED TO 1645.7 LB./SEC.)

useo 7

7 PRI.Le r PtAP TO R.V.PGM.

C

-O-O PRI.LOSP Pur TO R.V. DE N

SE

+

FIGURE 4-8 PUMP TO REACTOR VESSEL PRESSURE DRO.) VEREUS FLOW

+

l 11 A

SC 5

~.

Ln W

C 3

D C

C U

W 5

M 5

- 5 30

//

pu l

/

m u

.l a

e o

//

y=

w 2d U

//

//

. ll$

I

//

g NW$$$

//

d x

U6 a

/)

//

EE 5

9 e

/

i s-E

  • l

/

g

[

/

/

/

1, e

8 5

2

  • 08 M81d'Hd8N / d VIT30 'd3303 d080 380SS%

+

+ i

T D

+

f

. ),l t l

t

.i: :

i its i j 4

I i.....'i, h l

.r n,. i

.4

+

i f{

i

...j.j

!Il E

64'.i,

I

.t t

'l1 d

l,.

! l 'l l

li j

1 i

i i' i t 4- '

+ +

1

++- +-- -

--I y:i g

o

. q.l... l. _ _ _ l' _ _.-.._

li I

y w

..n_.__._

7 1

l a.

il l

q' l' _ _

l l

=

i a

l I

g w

M 8

_Y.S s

Z

.i X

i o

Ill

[

f

!'N I

s ill_j

!l\\

gOa$

~

3 o.

ss s

s n

m n.r

\\

k l'

EE i

x I

~'

. \\

EE k

.\\.

II i

p l

o g

\\

t O

=

1 s

N i

l

\\

O s

,o J.

T u

i m

u.

8 8

8 9

5 d

d 1333 - O m dOd o

+

i 4.4 IHX MODEL The thermal model of the IHX includes buth the tube bundle heat transfer region as well as the inlet and outlet n'ena for both the primary and secondary sides.

The tube bundle of the counter flow IHX is represented by 45 axial sections.

Fear radial nodes, at each axial location, represent the shell and baffle temperature, primary sodium temperature, tube wall temperature and intermediate sodium temperature. Tnis model is based on a one-dirnensional flow model in which flow maldistribution is neglected. The radial heat transfer coefficients are based on the Maresca-Dwyer correlations which were recommended by the vendor.

The tube bundle model is described in detail in Ref. 1-1.

The fluid transport / heat transfer between the fluid and structure in the plena are represented by 8 equal size axial sections each with a fluid and a metal node. An alternating central-backward differencing formulation is used to solve these equations. ibis model is discussed in detail in Ref. 4-4.

4.5 DHX MODEL The DHX model used in the FFTF pre-test predictions is a replacement for the steam generator heat sink model which is a standard part of DEMO.

The DEMO verification objectives do not include the DHX, however, a simple heat sink model was rf. quired to complete the DEMO plant model and make the

(

initializations and transient simulations reasonably straight-forward.

I The OHX is modeled, thermally, as a simple counter flow heat exchanger, represented by 8 axial sections. Four radial nodes at each 3xial section i

represent the sodium fice, tube metal,. fin metal, and the air flov

% DHX inlet and outlet plenum are each represented as a single mixing r oupled to a plenum metal node. A simplified proportional-integral air flow controller was added to maintain a set DHX sodium outlet temperature. Right after a reactor scram, when the DHX dampers are being shut off and the fans j

are tripped by the scram signal the air flow was modeled to match the sodium outlet temperature obtained from an FFTF IANUS calculation.

i 6071A--602A I

. - - ~ - -.

I r

I a

[

4.6 PIPING AND PUMP THERMAL MODEL i

The accurate calculation of transient sodium temperatures throughout the plant is essential to the calcflation of the thermal driving head. Because it is l

the thermal head which determines the flow after the pumps have stopped, accurate model.ag of the thermal heat ct.,acity of the piping and components is necessary. Thermal calculations are performed in DEMO using separate thermal j

models for the piping, IHX, DHX, pump and reactor.

The FFTF plant contains approximately 1253 ft of piping in each of its tHee loops (333 f t in each primary circuit and 920 f t in each secondary circ 't).

The DEMO piping model represents 2 loops ut piping with 52 segments eat..i consisting of 8 axial nodes and 5 radial nodes. The radial nodes represent the coolant, pipe wall, insulation, surface and ambient. With this !ayout the average axial node length in the primary is 4.6 ft and in the secondary 9.2 ft.

The finite difference formulation of this model uses an alternating central-backward differencing technique to solve the equations. This method minimizes the " false mixing" (a numerical artificality that smooths the ca! ; lated axial temperature gradients in the fluid) effects introduced by the approximation of the axial temperature gradient with temperature differences across a finte node length. A more complete description of the piping model can be found in Reference 4-4.

This model har been verified analytically with l

closed for.n solutions (Reference 4-4) and experimentally with data from EBR-II.

I In the FFTF pump, flow is ducted from the inlet to the impeller and from the impaller to the exit nozzle. There is a net elevational change of 6 ft in the pump making the calculation of accurate temperatures in the pump important.

The pump thermal model used in these pretest predictions is a " pipe-like" model similar to the model used to represent the IHX p1ena. The duct is represented by eight axial sections each containing a coupled sodium and a structure node. The structure nodes account for the duct metal, the impel kr, and the sodium in the relatively stagnent sodium region surrounding the duct.

The sodium in the upper pump tank is neglected as are the leakage flows around the impeller. The model used for FfTF differs from the one used for CRBRP in Reference ",-4 because of the difference in pump design..

6071A--602A l

i I.

i 5.0 BOUNDARY C0hDIT10NS f

f Power generation in the Reactor and power rejection in the Dump Heat l

Exchangers, which are boundary conditions for the pretest predictions are j

discussed in this section. Unlike the temperature, flow, and power initial conditions, the post-scram decay power, which is a function of the pre-scram power history, cannot be precisely specified. Similarly, the transient heat rejection in the DHX's, which is a function of air ambient temperature, cannot be accurately projected. Additionally, verification of the simple DHX model is not part of the DEMO verification objectives. Therefore, the approach is to identify the DHX outlet sodium temperature and Reactor power as assumed f

boundary conditions.

Considerable effort has been spent in selecting reasonable boundary conditions; howevcr, it is likely that post test calculations made with the actual measured values f or the DHX outlet j

temperature and decay power based on actual pre-test power history will be necessary, i

I j

The post-shutdown reactor power generation is comprised of the fission power 4

i and decay power. The fission power is calculated in the DEMO code with a i

t point reactor kinetics model. The decay power is input in tabular form as a function of time, with separate tables f sr the fuel and non-fuel assemblies.

The fuel assembly decay power is the som of individual assembly decay powers esed by PNL in the whole core COBRA calculations. The non-fuel assembly decay power table is similarly a sum of the decay powers in the control, shim,.

i l

reflector assemblies and the fixed radial shield. The actual tables were generated by the PNL whole core COBRA developers with the HEAT code supplied by FFTF engineering. The power history for the 100% case assumed 24 full l

power hours of operation prior to the reactor trip. The 75% and 35% cases assumed 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> at 75% or 35% just prior to shutdown. The values of decay power used for the best cstimate predicticns are shown in Table 5.1 for the fuel assemblies and in Table 5.2 for the non-fuel assemblies.

The boundary condition at the heat sink is the sodium temperature exiting the i

j DHX. Figure 5-1 shows the boundary conditions for the 100% power nominal c ase. The boundary conditions for other cases are in Appendix A.

This l

transient was produced by the simple DHX model and air controller discussed in Section 4-5.

i 6015A-602A-(S1478) t

+,e n.,

--r

~ -,,,

.wr,..-n....

,n, -,.,,. _,, - -. _ _. _. - _

O o

o Gl0.

I 600.

7 590.

S00.

1

,g 570.

e a:

~

E.

550.

g, O.

100.

200.

300.

400.

500.

600.

700.

800.

900.

1000.

TIE - SEC.

LEGEND 7

7 S-OHX NA EXIT TEW ITERMCLP)

FIGURE 5-1 HEAT SIrlK B0UNDARY CONDITION FOR THE 100% BEST ESTIMATE CASE -

+

S0DIUM TEMPERATURE AT THE DHX OUTLET N0ZZLE THERM 0 COUPLE

TABLE 5.1 FUEL A55EM5LY DECAY HEATS Decay Heat - (Watts'X10-7)

Time-Sec.

100% Power 75% Power 35% Pcaer 0.0 1.854 1.191 0.5558 1.0 1.730 1.097 0.5121 2.0 1.652 1.040 0.4851 3.0 1.597 0.9983 0.4659 4.0 1.554 0.9653 0.4505 5.0 1.518 0.9383 0.4379 6.0 1.486 0.9150 0.4270 7.0 1.459 0.8948 0.4176 8.0 1.434 0.8760 0.4038 9.0 1.412 0.8595 0.4011 10.0 1.392 0.8445 0.3941 20.0 1.251 0.7389 0.3448 30.0 1.163 0.6734 0.3143 40.0 1.099 0.6260 0.2921 50.0 1.049 0.5636 0,2747 60.0 1.008 0.5580 0.2604 70.0 0.9731 0.5321 0.24E3 80.0 0.9430 0.5098 0.2379 90.0 0.9167 0.4904 0.2288' 100.0 0.8934 0.4733 0.2209 2CC.0 0.7512 0.3694 0.1724 3CO.0 0.6786 0.3176 0.1482 400.0 0.6312 0.2848 0.1329 500.0 0.5955 O.2606 0.1216 600.0 0.5663 0.2411 0.1125 700.0 0.5411 0.2246 0.1048 800.0 0.5189 0.2102 0.0951 900.0 0.4989 0.1975 0.3922 1000.0 0.4809 0.1861 0.0568 2000.0 0.3611 0.1147 0.0535

-3000.0 0.2955 0.0797 0.0372 4000.0 0.2536 0.0595 0.0278

- 3 4-6015A-602A-(51478)

TABLE 5.2 NON-FUEL ASSEMBLY DECAY HEATS Decay Heat - (Watts X10-5)

Time-Sec.

100% Power 75% Power 35% Power 0.0 9.204 4.989 2.328 1.0 8.708 4.618 2.155 2.0 8.401 4.388 2.047 3.0 8.183 4.223 1.971 4.0 8.010 4.094 1.910 5.0 7.866 3.987 1.860 6.0 7.743 3.894 1.817 7.0 7.634 3.812 1.780 8.0 7.536 3.739 1.745 9.0 7.448 3.672 1.715 10.0 7.368 3.614 1.687 20.0 6.807 3.195 1.491 30.0 6.457 2.935 1.369 40.0 6.203 2.746 1.282 50.0 6.003 2.598 1.212 60.0 5.839 2.477 1.156 70.0 5.699 2.364 1.107 80.0 5.579 2.286 1.067 90.0 5.474 2.209 1.031 100.0 5.381 2.140 0.999 200.0 4.808 1.729 0.807 300.0 4.512 1.527 0.712 400.0 4.316 1.397 0.652 500.0 4.165 1.302 0.608 600.0

4. 041 1.226 0.572 700.0 3.932 1.162 0.542 800.0 3.834 1.105 0.516 900.0 3.746 1.055 0.492 1000.0 3.665 1.010 0.471 2000.0 3.092 0.725 0.338 3000.0 2.729 0.578 0.270 4000.0 2.462 0.486 0.227

~ ~

6015A-602A-(S1478) l

It is recognized that the ORr cutiet te ;erature wili cre than likely be different for the actual test than that calculated in the pre-test predictior;s because of diff erent rDient te perature ccnditicrs and crudeness cf the si ple DHX.mdel used. Seasured values cf the DMX sedi r cutlet te";erature transient will therefore be used x.,

an input bcandary ccndition to DEVO, if recessary, during the post-test ?nalysis.

i

)'

},

h I

t c

l i

~ ~u ~

~

6015A-602A-(51478)

6.0 PRE-TEST PREDICTIONS - BEST ESTIMATE In this section the results of the best estimate transient natural circulation pre-test predictions are given for the tests from 100%, 75%, and 35% power.

These predictions are based on the best estimate pressure drop data, pump coastdown model, dimensional data and component thermal models as described in Section 4.

These predictions are therefore the best estimate of the plant performance during the transient. That is, there is an equal likelihood that the measured flow will be more or less favorable than the predictions. The limiting case or design calculation will be presented in Section 7.

All of the predictions developed for this report are based on the boundary conditions given previously in Section 5.

For each of the tests discussed in this section a set of predictions will be included for the significant measured plant parameters. These are listed in Table 3.1 and include primary and intermediate flow and primary and intermediate hot and cold leg temperatures.

These rJeves represent DEM0's best estimate of the actual values of the plant parameters. No correction was made to these plots to account for instrument time constants. These time constants are given on Table 3.1.

In addition to these parameters a more complete set of curves for each test is included in Appendix A.

Test D simulates a transition to natural circulation from 100% power and flow. The DEM0 initial conditions (temperature, flows, and pump speeds) were matched to those listed in Section 2 for this test.

These conditions were obtained by fixing the reactor power, volumetric flow as measured in the cold leg, and secondary cold leg temperature and allowing the plant to relax to ste.dy state. The DEMO simulation begins with a scram of the reactor and is followed by a trip of the pumps.3 seconds later, this simulates a manual scram from the FFTF main control panel. The scram and flow coastdown initiates temperature transients at the reactnr outlet, IHX pr6 ary and 1

intermediate outlets, and the DHX sodium outlet. 6015A-602A-(S1478) i l

. __~

_ ~_

t The primary pumps coastdown and stop in between 90 and 95 seconds after their i

trip. This caastdown is shown in Figure 6-19.

The primary ficw drops to a minimum value of about 2% (260 gpm) of rated conditons 100 seconds after the d

start of the transient.

It then recovers to about 3% (390 gpm) as shown in Figure 6-1.

The drop in flow at 100 seconds can be explained from an examination of the reactor temperatures.

The reactor fuel assembly exit l

temperature shown in Figure A-5 decreases initially because the reactor power shutdown proceeds more quickly then the flow coastdown. Consequently the reactor and the hot leg piping are filled with cold sodium, causing a I

reduction in the primary natural head as shown in Figure A-10.

As shown the primary natural head drops to a minimum of.09 psi at 40 seconds. When the i

pumps stop and the flow is supported only by the' thermal driving head, the l

flow drops to a minimum while the reactor temperatures i'icrease to a maximum value. When the reactor temperature increases the thermal head recovers and the flow recovers.

From this it can be seen that the minimum flow reached by the reactor and the time required for it to recover has a significant effect i

on the maximum reactor temperatures.

Figure 6-3 shows the predicted temperature at the location of the hot leg RTD for this. transient. This l

instrument is located 85 ft. upstream of the pump. The transient experienced l

at this location is the reactor vessel exit temperature (Figure A-1) mitigated by about 65 ft. of piping. Figure 6-4 shows the predicted temperature at the

.ary cold leg RTD. This instrument is located just upstream of the cold leg l

cnc.k valve. The transient introduced early on (first 50 seconds) into the primary cold leg is caused by the collapse of the IHX primary outlet temperature onto the secondary inlet temperature. Figure A-9 shows this t

effect. This collapse occurs because as the flows decrease the dimensionless I

size of the IHX becomes large and the heat is transferred more efficiently

{

from the primary to secondary side.

I The secondary pump coastdown time is between 45-50 seconds as shown in Figure i

6-20.

The resulting secondary flow coastdown is shown in Figure 6-2.

This flow remains higher than the primary side, dropping to a minimur. of about 4%

i (520 gpm) of rated conditions.

The secondary thermal head is soown on Figure A-10.

t '

6015A-602A-(S1478)

l-

+

e o

t>

O 3

8 l

W O

O (D

9 C

n3 8

"E

.i Be la a

zu&

m i

\\

= ~gp h

I QCa ag,e-g s

1 sP m

i

$E*5 'W h d8

\\,

go

$w k 2 E

<c-WG E if u S

95 e

o I

d = tal w

I b~g w e o oze

+-

i

$p BEL i

8E0 5g:- W5t EW "g o

    • k i f o

i l MM w w z a o.

i

~

88

'1 l E

' 8

\\

t,>Q

,u

. m

-p 8

../

6 O

O O

GuuW guuD sus Ndo - M07:1 dDel ABYHI8d

+

o. - - - _ _ _ _ _ _ _ _ _ _ _.

._a

,w4

=

+

+

8

I:,

O l

i

(

Y "

i!

I O

Ii g

l t

O O

O CD n

M Ce n +>

O we m O

=w c N

l-3 O OV Ha i

_ < p t

m

>* c I

3'O e i

ex Co&

4 J-

?

Y I

(Q m

t 82 wmy an H w n=.

i

<W l

  • ?.'

"CW A4.J

^ E~

Ync u

4 m

-- mww Q u.

w --

N g

  • W~

HbL Z<C d-U

~E' br 5

"N 9 h,E i

O Ae; wU wa 0

T m

=

gy U

C W G) w

~k wwk l

MM ECL b8 azo 8

> C -

Ug Z O

n a

wwa O1*

~

I*

l I:

O' E"bI w'

i v

i 6

O

. w i

N C1 b

_a -

I a

t>q a o

w i

O O

}

~

  1. f a,,,,s AA o

O C

8 O

O O

O_

~

Mdo - Mal.-l d207 ABYONO33S

+

+

+

o 8

f o

Q i.

.o 8

i s.

uc

\\.

m am o

e o

o N

y N

a e-i-

5e y 4

anc

-e x

$"y

\\

U, z 3 'B 4

\\

.O.

au G1 m

m t*-

aue g c o.-.

.t. o

,J

,i g

vs s

m Evo t

=

d-w l'

[

E 1

y cm I

h*f w o 8 s

.,/

t__

e-b-u g

s 50 6-. K V owe t;0; eke EW "We 0*E wszfao

/

m omv s'..

8 8 e,,

4m

../

g

/

i W

w 8

css cu u.

i 8__

~.*

o

.4'030 - # 31 031 10H AHVH18d o

+ o

+

ll1l

+

+

000 N~9 1

0 N,

09 008 i

t nnr e

0 n

~

0 nr 7

o N ',

i m

D i

G T

R E A

0 1V L

t 0

H

)

n 6

.N

0. E 0'.

R em 1

I 2

0W u 0C CO r S

CP

~

O IO P t SC Y s

fI R% n t

0 tF A0I 0

C oTT M0 CF 0 E oFI.

~

I 1 r 5 S c(

R( o rO P

f s'

s E TIE E

FT DR n TA l

FM EU o i

x

~-

T FI TT i IT.

cat 0

r I R c 0

HE DE e 4

C EP r IT S

RM r

r. S o

DKPE TC 0

IT MD o CC PP ETN 0

DR(

0 00

~

3 00 11 4

- 6 a

" O ER U

0

- G 0

VO I

F 2

~

00 1

i t\\

~

~

r O

0 5

0 5

0 8

5 3

0 9

6 6

6 6

S u dg,

- ow2 e h{

+

+

,uNe

+

l l

'W N

D

~

e e#

,0 A

t 8

2c 3

m g

8 8

/

8c v

LaJ H

E 4J H^c N

c%

ag w

tw

\\

(O 4

f3

.u.

,n a b, a_

Owc s.*N.,

-=-O-DO

's N

A U-5 U

La.

wv o W y_

m

+

' ~..

t aW 4

S-g

.g w; = -

bt

~...

e-u O w G)

O N

g

.-.- W & E I

MM nwo

, ~.. ' -

8W ru Eoo gv wrz A O%v 8 8 ' =.

't C hh y

aC g

q u.

c N

G 8

eu g

,o d

d d

d'

^

s 8

8 8

d*030 - dH31 037 10H A8VONO33S

+

+ +

w-O

+

e es s

~

q 8

m s.

N,N N

O O

N.. ~.

~

Q n

\\

8 8

~

u

~

C m

.i

- w O d b

kO u

I

~ ~... -

8

, a-c s om y Oft UwE W g:

33

>- o s

.s **.

rn U

xau Q

sn

~*

f's m

ane g

sc 8 to GC 88-O o

v-s m

w w-o e

' = W b

i

/

Wb-gaWe b2wzo p

U <C D

-/

==

~w mau g*

W&

5 us.c u

-> NIL ta tt G. W O 88

~u O

EOo pg w 6- ~

OW f

\\

88 9 o

w E*h w \\

- a i.

lf 8

W D

I e

(M b*

O O

f am t

t t

t

'T s

O O

O O

O 01 CD W

Q CD W

W LD La T 030 - dH31 031 0'103 A8VGNO33S

+

4 +

+

1 The secondary hot leg RTD is located upstream of the tee, before the OHX inlet. The transient induced in this pipe is a -esult of a change in the ratio of the primary to intermediate flows. Figure b-5 shows the predicted temperature at this location. Figure A-3 (Appendix A) shows how this transient is mitigated by the piping.

The secondary old leg RTD is located 60 ft. downstream of the secondary pump outlet. Figure 6-6 shows the DEMO prediction for the temperature at this location. The transient produced at this location is a result of the^ shutting down of the air flow to the DHX.

Figure A-4 shows how this transient is mitigated by the piping and the pump.

Test C simulates a transition to natural circulation from 75% power and flow.

The initial conditions for this test are given in Section 2 with the power, flows and secondary cold leg temperature used to fix the steady state conditions. The shape of the transient curves produced in this run are similar to those for the 100% case. As shown in Figure 6-7 the primary flow drops from 75% to a minimum of 1.7% (220 gpm) of rated conditions at 100

)

seconds before it recovers to between 2 and 3% (260-390 gpm).

In this case the primary natural head drops to.07 psi at 40 seconds before recovering as shown in Figure A-20. The primary hot leg temperature is shown in Figure 6-9.

This transient is caused by the rapid decrease in reactor outlet temperature transported through the piping. Figure A-11 shows the primary hot leg l

temperatures. The primary cold leg transient is shown in Figure 6-10.

The primary pump coasts down in 90-95 seconds for the 75% case as Figure 6-21 shows. The secondary pump stops at between 45-50 seconds. This coastdown is l

shown in Figure 6-22.

The secondary flow coastdown for this case is given in Figure 6-8.

As with the 100% case the secondary flow drops to a minimum of

~

4-5% (520-650 gpm) of rated flow.

iigure 6-11 shows the secondary hot leg temperature. Figure 6-12 shows the secondary cold leg temperatures. Both these curves are similar in shape to the 100% calculations.

~ ~

6015A-602A-(S1478)

+

+

s e

i i,

e e

i O

l

\\\\

?

8 i

a

+

1

}

O O

'a i.

1 m

I i

4 j.

n w

(,

O 3

^I O

o i

J r-wa i

F wa 1.A b

1 I

HN I

E

<v

{

8 I 2x e

Eu cw a-N kN 4

x e n$,

I Li l

O e i

O v

-m

-w US E

O W l

g-

c. H. <

j g

Rt W %.w c W l

wc g

i i

- bi-bf I

O r 5"

l O

op w

v wz 1

c4 w h

a Mm cu. -

gg

<t er i.

+

e- >

I. E <

i I

I UU ww-l AA C a. -

i O

W%

N 3

6c E

u l

Ea g

hL C

P d

N

~

1 i.

l, 1

6

-)

gl i

J #>,

i

.O 8

i i

f Hao - M07.1 deal ABYHiUd

+

+

I - _ _

l l'

(

000 1

y c.

00 9

)

t 0

E n

0 H

a 8

T) tR s TE n AW o OC WP 0

O e

0 L% m 7

F5i7T P(

O t

OR n LE e t

8S 0

9 T m s

YE u 0

fe RMr t

6 lC A

t i

CT DWs LS NOn PE S

GS OLI t5 CF D

s7 E

f N

EF STo 0 O tT E

0 C iF CF DNn 5 E ego a9I S

tai

  1. E CMt cFT E

I c

M uIA DT 4

ENm FI I

nFI 0 T iT REr s

0 NE PNo 4

C AC lT OM Ss MRo Es EEN De DP(

i Y

c 8

l.

0 C P 0

P 5

6 3

5 7

7- -

E O ?

R U

G I

F 0

0 O-2 00 s*. A 1

~'

i

,I h1-O 0

0 0

0 0

0 0

0 0

0 0

1 0

0 l

0 1

1 s

2%.E$ %3>.E@8i

.e*

ll

. _.. ~...

+

SSO.

eW-N N

.~.

ew.

K u

r.

N.-

o

~

(d O

N s

9 N.

to N

850.

\\

o i

8 kJ

~~.

. o.

s m

I e

... o.,,

n I

~

x E

800, s

h 1

i g

a_

a e

i i

e e

e i

O.

100, 200.

300.

400.

500.

600, 700.

800.

900.

1000.

TIME SECONDS s

LECDd

-d" - d" 75 PCT DESIG4 IFFTF75PCTDESIGLCOR Op 7E PCT BEST EST! MATE tFF1F75DESTC9R 98

(

+

FIGUPI 6-9 DEMO PREDICTED PRIMARY HOT LEG RTD TEMPERATURE (75% POWER)

(No Correction for Instrument Time Constant)

i o

o 680.

4

\\

655.

.\\

u.

5

~

)$

~ * -.... - -

___ N

~

N.

to 630.

~

8 a

f y

~

a u

y E

~.

g 605.

_,'~ ~...

' ' G.

580.

O.

100.

200.

300.

400.

500.

600, 700.

800.

300, t000.

TIE SECONDS trcom 7 75 PCT DESIGN IFFTF75PCTDESIGM.CsR 98

- --- O-75 PCT BEST ESTIMATE IFFTF758ESTC8R BI FIGURE 6-10 DEMO PREDICTED PRIMARY COLD LEG

+

RTD TEMPERATURE (75% P0'4ER)

+

(No Correction for Instrument Time Constant)

i i

i 1

4 i

+

+

e

=

1

~

(I l

\\

l i

1

/*

\\

\\

l ma k

,~

e s.

m

,./*.

s v

8 g

e n

C 2

/.

-8

'W W

f e

v W

H C

e on v

-N

.\\.

U"

'wa N"

O

>mL wou m

-m

.t c. m a

m e-a c

8_ b

=M~

C Ln

,w uw vN L O

u S-WS

'g O

g W

te i

s,.

m w

s Oac a

..N

,s,..,

o<

Ww woo j

.J w -

FH-(

$.- g]p 5

u E

w mm x u

a g awo 1

ru

)

w o-Q CO m

w&z i

cxv i

i t>Q (9

I w

N$ e m

8 u.

I CJ l

O j

O I

~

i L-a o

J O

O O

O O

5 8

8 E

8 i

i

.I f 030 - dH3103710H ABYON0')3S s

(

+

+

1 l

i

O o

O l

4 610.

am

.'A 600.

~..

g

'b'

/

g g

590.

N

/.

W

/

n

.\\

/

o w

\\

j J

m o

5

'/

/

T' d

500.

-\\

\\

- o

\\

x

\\

,/

\\

I 0

\\

/

\\

/

\\

/

0w s.

570.

\\.

8 s.

a a

e i

t t

I e

a a

0, 100.

200.

300.

400.

500.

600.

700.

800.

900.

1000.

TIE SECONDS LEcce D" -- D 75 PCT DESIGN IFFTF76PCTDESIGILCER 83 75 PCT SEST ESTIMATE IFFTF768ESTCM 88

+

FIGilRE 6-12 DEM PREDICTED SECONDARY COLD LEG

+

O RTD TEMPERATURE (75% POWER)

(No Correction for Instrument Time Constant)

i 1

Test B simulates transition to natural circulation from 35% power and 75%

flow. The initial conditions for this test are given in Section 2.

As with l

the other predictions deb 0 was initialized by fixing the power, flows, and j

secondary cold leg temperature and allowing the program to calculate the remaining plant conditions.

j For this transient the primary pump coasted to a stop in between 90 i seconds. The pump coastdowns for Test B cre almost identical to those for j

Test C shown on Figure 6-21 and 6-22.

Figure 6-13 shows the primary flow coa stdown.

In this case the minimum flow in the primary loop is 0.9% (120 gpm) at 110 seconds.

P should be noted that the pressure drop data is l

extrapolated to this low flow. Therefore, more uncertainty exists in predicted flow than existed in the previous cases. The flow recovers to about 1.~i% af ter 1000 seconds. Figure A-30 shows that the minimum natural head reached in this case is.03 psi at 50 seconds.

This is lower than the 75% and 100% power cases because of the reactor AT's are lower resulting in lower initial natural heads. The predicted primary hot leg temperatures are given

)

in Figure 6-15.

At this low flow the transport time through the loop is much j

greater than either the 100% or 75% power cases as-can be seen from a j

conparison of the primary hot leg temperatures on Figures A-1, A-ll and A-21.

For the 35% case the sharp temperature drop at the reactor vessel pit does not reach the primary pump inlet ($ 150 downstream) in 100 seconds.

This compares with 500-600 seconds for the 100% and 75% powercase. Figure 6-16 shows the primary cold leg temperature. This transient is similar in shape to the~

previous tests but expanded in time because of the lowr flow rates during this test.

i j

The secondary pumps coast to a stop in between 45-50 seconds.

The predicted secondary flow coastdown produced in this case is shoun in Figure 6-14.

This-flow drops to a minimum of about 3% (390 GPM) of rated flow. The temperature for the secondary hot leg is plotted in Figure 6-17 and for the cold leg ic f

Figure 6-18.

For the initial portion of the modelled transient, a large change in temperoore is not predicted.

l l

6015A-602A-(51478)'

. _, -., - -.. _ _ _ - _. - _ _ _ _ _ -..... = - -.

O 100000.

-r e.4eD66*=

P*

m m.

m

-.d.m 5.h Mh+

mm 4 _ W6 9*

10000. P ga.

O g

d n.

_i 2

5 i

i w

=

s I

1000.

J.

u IE

\\\\\\\\

\\;

\\\\

\\1\\

o.-..

.3 1

.I

,00, 0.

100.

200.

300.

400.

500.

600.

700.

800.

900.

1000.

TIFE SECONDS LIGEIO 7

7 35 PCT DESIGN IFFTF3SPCTOESIGN AI

-O---O 36 PCT BEST ESTIMATE IFFTF35PCTNSINAR AI

+

FIGURE 6-13 DEMO PREDICTED PRIMARY LOOP FLOW AT THE PERMANENT MAGENT FLOW METER (35% POWER)

(No Correction for Instrument Time Constant)

(~

t j

i fflOl' l'i '

l i c-l i

I

't i i l

i

,; i I.

l

<;t i.

f l

j, li 8

,i I!

n I

e i

4 c

I j i

i

.i J

[

}

l, i

r,,

i 1

,i, t

11 i

i l

1 O

! i

)

O i

l' i

m

' ; j I'

I

' - 1 i;

i t

I l

l i!

It i

c t'

I l i.

1 2

w,_. m I I I I

{l 8

=xc HW C

,,l ',. - l 6

IU

,' )r jj

< c c,

Ho

e, i l

ff en-E a

Ili i

i3 :

i l1

{

l 3

O i.A H 6

4 g

n I

8

'~t

!i -

l i

y i

i-

.t to wx

.- w 8' jl1 I

li

~

<H a

59

,4:

i i

m

-wL 1

-9 cE-i i 6:

C) ob w

m o d WW E* C i.'

N.

mo-t

'.i}

l

-a-

{

o u wa 4

m w m

-uu i

r o

l 1

ahW 2

o l<

l lt W bb*1-

~H%

K 1

J W

t

<t i

)

wm cm e A

mua 1

O H

.l 1

I l[]

H <E w

,I l

?

if lj, l

l O

M v

t v

5" i

4 o

i t l

cH a

'j l

i m

w= L f

il I

g Cr; L J L i :..I i I, ano i

! I i.4 t

i, j

c h' u w

vu 8

AA Ex0

!1 I ;I9 i )

l i

4 a

wwz

,l,

- 1  ! ! i M

88 oc-t i

.l,l ll l!

l I }l 1l! !'

=

1 i >

"h h

~

i i C

i i i f

i t

i!

e o

,jlt l

l llj f

8 N

l J

e i i i.

al i

il l

i lll t

i1 i

i, t

n 4

iii 8

r i

I: Ij l

!. i.!M

'l,6 i

1.,

p<-

i t j i '1 !,

i.. l I

5

>>4 w

.o o

o o

d 8

8_

o o

Hd0 - Mald d207 A8VCN033S

+

+

O o

O l

700.

n a,.......

y s.,,.

ns.

J 740.

w O

l w

\\

730.

\\

\\

a a

N 7

J n

e l

I 720.

\\

i i

.x

\\.

1

~

EA.

x 710.

., ~

700.

Y

"'W.

0.

100.

200.

300.

400.

S00.

600.

700.

800.

800.

1000.

TIME SECONDS LEGD0

-D"--- O 35 PCT BEST ESTIMATE gFFTF3SPCTNetVAR Al

" 3S PCT DESIGN (FFTF35PCTLESIGN Al

+

+

FIGURE 6-15 DEMO PREDICTED PRIMARY HOT LEG RTD TEMPERATURE (35% POWER)

(No Correction for Insturment Time Constant) f," '

7

-um p_%_'

J.....__.

E e

8 O

i e

a

,i

./

u.

^

l 3

Cs m

/

u m

8 C

/

C Q

U i

g E

./

s w

l Y

O c

{

N Ceo 3

/

JM&

m CW O i

(

o 3'

/

L g

Oo O.,

O X

C

}

8

<t a ? w

/

2C t.1 w

-mL g

I

.h$

XwC i

c.

)

/

D an W

)

Q Q*

46 O%C W O,_ O r

~

-w d a vS o <; u gh i

v

-av O

to

_nw

=uo g

M gWH

}

W $i._(-

u c. L

% "E L awe su

~

.S m w

s

.w n

b

-m s.a W Z NN OEU j

bM o

7 i

~. _

av AA O

I bb 66 8

o L

o l

8

~

\\

1 l

l O

)

/

O O

r s'

e

,s***

,f ~ ' ~,,

O

.d'030 - eH31031 GT) ANYHIUd

+

+

I I

1 O

+

3 l

i

~~

,/

p W

\\.

\\

(.

\\

'N,,..%

c' m

a a

'~

N s

8 g

a s

N..,s 6

g o,--s ze wu

_3 c eoe gag 8

4 z at L G

omu 3

umm N.

==

wwc m

vs

~

M gh w

.N

  • g rm aze vu wmo O G
a. -

es-8 " TN MMU s,

W 9.1 owe w c. L 1u exu W

c. w o i

5a su o

s

.o EOo MW o"~

mm w r-z

~

~~

s I

MM Te NN.

=

w n

t@

Q e

u E@

i 8

i N

l 8.-

.m d

o.

o o

o.

o.

N 8

m

\\

8 m

d'030 - dH31 037 10H /HYONO33S o

, I

+

+

a l

~

l

/

d~

/

~

8 i

/

',,/ *.

m

+.

a ag C

e c

w o

8 a

u N

a c

-.w

--=

u

.~

a cr a C

C 8

<z aa u

/

w f

.m w a u ee u

,f

.o. b was wn w: w

-i m

m

=

Pfw

.e se p

by a m,

/

y_

wc u

u o

u

--o o w 9onw v w

m m U ~w.-

-a S1 d i_ _g j

ax e w=

0

/

cc

-J

~

E *%

O

.s gw 1 u O

9 "n-o c

r i o

ww v

ww

'o wE a ct.

Owe"u

--gg ao

- =

e cc -

gg 8

66M o

oo u

E>

i 8

CJ 8_

l A

t 1

.o o

g' 0

m l

  1. 030 - dA31 037 0703 A8vcNc333

+

+.-

o P

+

S 4>

4)

N 8

-=-

sa w

e 2

L 5s.

82 i

G w o.

m _.3 l~

Nc 8 '-

-i s

va i

W D

Z dbW l

N

-s e i

W If w

/ -

p

=

50 st t

=M

/

=o a-8 EE i

~

88 s

i E*Q E

(

./

C3

)

s-e i

,/

caQ

!/

/.l' N

..a

^

o Hd8 - t TV3013 033dS dH0d* iud-S

+

~

e O

+

g 4

O l'.

o

.M f-1 i

m 4

N g a<

~~

_ f =.

a, 1

2 m 5*

O. Y W

. a y, g

~

M tic co 9

dC 3w d

1

. e,- - --

o Zu6 L$ $ d b

I

~ w 3_w aa

~

k a-L. t X

gZ >- w '

N

=

z 0 - b..f e-rJ

~ 6-CH L% tn

J o gg wo

..w-Eg.

oo 88.m

- o 1

5 Wr g

ND-w m

ea

/

~

/

.o-o o

o o

m o

o s

m N

Hda - (TrNT C3?dS ErGd*LNI-S o

+


m

A

-s b

W

'l I

l 0

o e

o 4

g

-==~a**==-

=

W G

S 1) 4 e

e W

C O

  • e 8

3 g ".--

v.o_ a OO Y

H3 j

mu mo I"

m C o.3 l"

h7

~

4

>d bh 4 N

.U to W b t_,_g

b. x o a w

vs_ >- 2 Ma

/

5"'

s a-

- R at I

W sn Q. N M D=

l

~

l h

b j

t>b w rd l

>Q

.C.D.

D

/

t u.

,o

/

9 y

l e

W w

O e

e a

a a

a 8

a a

8 NdB - (*1Y30m 033dS dH0d*18d-S O

+ t

+

+

- I;I,

l l

I:

li-Il4 05 1

7 O

A 52 1

i I

iWW 1B 0

B OO 0

DL 1

MCMTF S

CT AD LS NE O :;

i GS CA IS ST P,

EF DT NE S

W D

TF l

CF FO N

Pt P

O D5 J' 5

. C N7E EFT I

5 E GTA A5 7

S EFM D7 LFI M

E IS CI T

M NE G

E '!

I T

!T S F' SS EE DB 2

TT 2 CC PP 6 S5 0

77 E 5

- R

?O U G

d,

~-.

I F

?O 5

E 2

-,s' i'. i, '

it i

1.i

..o-y O

0 0

0 0

0 0

0 O

0 0

0 0

0 0

0 7

6 5

4 3

2

[

rc.a:,. dooJ-ow " n.ga..;,u o

t

+

n i7 c

1 7.0 DESIGN CASE PREDICTIONS The predictions shown in Section 6.0 were based upon best estimate pressure drop, heat transfer, and dimensional data input to DEMO.

These date were based upon detailed plant analyses of both test data and extensive computer and hand calculations, and are felt to best represent the expected behavior of the plant. There is an equal likelihood that the measurements taken during the test would indicate a more or a less f avorable transient than that predicted. A set of design case predictions were therefore made based upon typical design data--biased to include data uncertainties. As long as the measurements, accounting for instrument accuracy, are more favorable than these design case predictions, the DEM0 modelling and data used for FFTF will be jcdged to be adequate.

The changes to DEMO for the design predictions are discussed in detail in Section 4.0.

Briefly, they included:

a) increasing the primary and secondary loop pressure drops, b) increasing the reactor vessel pressure drop, c) reducing the sodium pump's rotational inertia to the.'. sign value, and d) reducing the reactor post-shutdown power.

Since the loop flow will be the governing criterion for assessing the adequacy of DEMO, the uncertainties were stacked up in a manner to arrive at a minimum flow. The decay power input table was therefore reduced by 25% - the typic-1 calculational uncertainty on decay power given the pre-trip power history.

It should be noted that this does not result in maximum reactor temperatures.

Therefore the design case reactor temperature curves given in this repcrt are not limiting values and it is likely that they will be lower than the test data.

If temperature were used as the limiting criterion it would be necessary to increase decay power to account for the uncertainties. This case, which would produce limiting temperatures but not minimum flows, was used for the design case COBRA analysis. 6015A-602A-(S1478)

Design caf e calculations aere cade for each of the tests B, C, and D.

The results cf these calculations are shcan cccpared with the best estinate calculaticis en Figures 6-1 thrcugh 6-22 in Section 6.

The changes to the pucp model for the design case resulted in quicker stcp times icr the pumps.

The primary puras coastdown to a stop 15 seccnds earlier and the intermediate purps coast to a stop 10 seconds earlier. For the 100% pener c ? sign case the reacter flew crops to a lower flew (1.9% vrs 2%) and reaches

.. einimun ficw atout 15 seconds earlier than the test estimate case (Figure 6-1).

Figure 6-7 shows the results of the 75% power design case. Again the flew crcps to a lcwer value (1.6% vrs 1.7%) and reaches its minimum 15 seconds earlier in the transient. The sa.me effect is pecduced by the 35% pcaer design case as Figure 6-13 shows.

In this case a minirun ficw of.E% is reacned in the design case at SO seconds ccTpared to.9% ficw at 100 seconds for the best esticate case.

d

~ ~

6015A-602A-(51478)

8.0 ACCEPTANCE CRITERIA This section delineates the criteria and logic with sich the adequacy of the transient natural circulaticn calculations, and by implication the models and data in DEMD, will be determined. The criteria are intended to be used in post-test comparisons of the pre-test predictions with data measured during the testing. The post-test analysis will assess the effects of actual test conditions as opposed to those used for t5e a-priori predictions, accuracy of data used in the analysis, effects of measurement uncertainties, and consistency of the models in DEMO with test results. However, in the end the questions to be answered during the post-test analysis are:

1) How well do the predictions and measurements agree? and 2) Are the differences between predictions and measurenents acceptably small?

8.1 POST TEST ANALYSIS Considerable post-test analysis will be necessary to support and qualify a

" bottom line" acceptance criteria. These analyses will include the following:

1.

Detenrination of the effect of decay power and heat sink boundary conditions on the pre-test predictions.

If differences between the assumed and actual boundary conditions are significant, the predictions will be re-run with a decay heat based on the actual power history and the measured heat sink sodium temperature but with the same DEK) model and input data.

2.

Assessment of the accuracy of the data used in the pre-test predictions. The objective here, to the extent possible with the instrumentation in FFTF, is to determine how much of the difference, between the best estimate calculations and the measurements, is due to uncertainties in the data used in the predictions. Particularly significant are pump data that would affect the flow coastdown and pressure drop data that would affect loop flows.

The steady state natural circulation tests (pretest predictions were made in Referer.c.e 4-5) will provide total loop Ap vs flow information.

The actual measured pump speed coastdown and stop time will be used to evaluate the accuracy of the pump friction torque data.

6015A-602A-(51478) l i

1 i

l 3.

Determination of the accuracy of cocponent thermal mocels.

The outlet thermal response et tne piping, punp, IhX, reactor fuel 4

assembly group, and reactor upper plenum will be examined for given ceasured ficw and temperatare input to the extent that that's

[

y possible with the installed plant instrumentaticn.

4.

An overall assessment of the accuracy, DEMO and data.

Based on 2 and 3 above and the differences between neasured and best estimate ficws, j

an overall assessment of the accuracy of DEMO and the data for i

calculating natural circulation flows will be made.

j S.2 EOTTOM LINE CRITERION i

The bottom line acceptance criterion has been developed in ter: s of primary f

i flow rate. This is because the primary flow available for decay hest removal is the key variable calculated with DEM0 in a natural circulation analysis.

1 Figures 8-1, 8-2, and 8-3, for transients from 100%, 75%, and 35% power i

respectively, each show ficw as a function of time for the test estimate case, l

the design case, and the acceptance boundry. As oiscussed in Sections 6 and 7

}

the best estimate case is based on best 6timate system data and the design

[

case is based on data for pressure drep, pucp inertia, and decay power with l

typical design uncertainties included in a directicn to produce minimun flow.

The acceptance limit curves were developed from the ficws calculated in the l

1

]

design cases by 1) increasing flew by 2% of the calculated value to account

)

for the magnetic flow meter accuracy and 2) adjusting the calculated ficw with a one second first order lag to account for the magnetic flew eeter tire constant. The acceptance limits correspond to the lowest reasured ficws for which it could be argued that the actual ficw wculd not be less than the I

calculated design flow. Thus, the acceptance criterion is that if the minir:um j

measured ficw is greater than the minimum in the acceptance limit, it would l

then folicw that the DEVO calculations made with design data conservatively envelope the actual natural circulation flow. The data and codel would therefore be acceptable for design case transient natural circulation predictions.

4 4 6015A-602A-(S1478) 4

- _ ~.

+ - - - _ -. -,

,-_m

1 l

l l

rb d

a b

y-LO H<

2" J

YLU y

D K.x ji

=

l C

I

~E

=v e

HW 1'

mm f

5 u

w L

=

W Ei e

oc h

--k

.-e e

-w m

tn wN i

l I

f a

x is m

c e

k-3c' i

i A

J 8

\\

B-G t

gsu

\\

2zwW s

u W W i,

-. v

~

8 W.s a e

- n W

t s

U=

1 I,

l i

g o

$1 I

r Il

],i W:

i

~/:

/

a'

/,

j'/

w-

,+

-8 s,

/,

t

/

l/

i

/

o g, -

~

x-go Y

h o

s'y /

w w

/"

8 e

e o

e o

o e

o k

b h

b (Wd9) t013 dOO7 ABVWIBd-l m

b l

W LW W^C*

F-- ct C

Z1 i

LD y

t i

m-D-F=.

l m

o 2O~

,a zo W1 e

>- W W m g

4./J 1 W

  • E
  • L 4 U,
  • ~*
  • C W

W X

sn sn e W.

cC

=1" v W.

~

> = = -

U UD N

i (A

J C

D-M Ch. 1 4

g(,

gW 8

4 e--

r W

L, a

r I.

de }

H)

>- M 6

me m; to MN t

i w

y 8 y r

N r

['

~

CCi C

w il BJ j

+

4 3

i i 1

Ai 1i i

i 1

e i

i 1

N tr ;

I, e

g r

i 3

g l5 i

i I[

5A. I i

I 3

l t

i e

h' a l)

I I

f t

t m1 I

i C

(

6 1

{-

1-l g

i 4 -

i

-^

f i

I

[

l

'I i

1 h: >.

y l-.

-6

j. 4

...t.

4 W

i i

i i

.e.

Jf f.

W 4

.i

+

W ut

{

[

ji q !,

t 1

F**

A i

-e i

sb j l

k, I-

)-

- c.

i l

l g

i 1.4.)

i

'g b

l

+

l-i I,

1, i

4 N

e-i m

m 3

i 8

e 1

i

=

O 1

t.

i z

i-1 I

i i

8 O

,f c

,6

+

i g

1 C

n.4.

4 y

CL' 4s

+

i

/. -

i vf ri

/r

/:

+

6 o

,/f, N

e,

/ /

/*

/*y -

y/e O

{

w r

/'

t t

i 1

i s

6 m

~

LO O

O O

.O.~

O O

O 8

O O

Q O'

O O

5 to Ln v

W N

r-*

(HdD) 1011 dOO 1 ABVW18'd

I

(

(

(

(

t

~

700

.\\_._

.__ _ a_

l i

BEST ESTIMATE I

t i

DESIQ1 CASE

_ 600 \\

l DESI44 CASE WITH INSTRUKNT EFFECTS l

.(22% FLOW,1 SEC0fl0 TIME LAG) l g,

1 500

\\

\\

.g.

_400 h

9

\\

\\

300

\\

\\

\\

200 N

x x

~ ~ ' '

s (s

- - ~

~

~ - -

.100 0

50 60 70 80 90 100 11 0 120 130 140 150

. TIME (SECONDS)

FIGURE 8-3 ACCEPThiCE LIMIT ON FLOW FOR THE 35% POWER CASE

9.0 REFERENCES

1-1 W. H. Alliston, A. Batenbur9, B. Dvorznak, et.al., " Clinch River Breeder Reactor Plant - LMFBR Demo Plant Simulation Model (DEMO),"

CRBRP-ARD-0005, February,1978 ( Availability:

U.S. DOE Technical Information Center).

2-1 FFTF TS-51-5A008, Rev. 2, " Specification f or FFTF Transient Natural Circul ation Test," dated September 1979.

3-1 FFTF TS-51-4/009, Rev.1, " Specification for Steady State Natural Circulation Test," dated January,1978.

I l

3-2 Fast Flux Test Facility System Design Description for Process Monitoring and Control System, S00 93, PARA II, Reactor Heat Transport Instrumentation System.

4-1 S. L. Additon, T. B. McCall and C., F. Wolf e, Simulation of the Overall FFTF Plant Perfonnance," HEDL-TC-556, March 1976 (Availability:

U. S. DOE Technical Inf ormation Center.

4-2 H. P. Planchon, R. Calvo, R. Schurko, S. S. Lin, " Comparison of DEMO Calculations to FFTF Punp and Flow Coastdown Measuronents,"

WAR D-NC-94000-1, August 1979.

4-3 FFTF SDD51, FFTF System Design Description Reactor Heat Transport System, 4-4 H. P. Planchon, W. R. Laster, " Loop H3at Capacity Models and Their Effects for DEMO Natural Circulation Transient Analysis," WARD-NC-3045-2, September 1978 (Availability:

U.S. DOE Technical Inf ormation Center).

4-5 H. P. Planchon, W. R. Laster, R. Calvo, "0EMO Pretest Predictions for the FFTF Steady State Natural Circulation Tests," WARD-94000-91456, October.

1979. -

6071 A--602A

l APPENDIX A TRANSIENT CALCULATIONS FOR THE TRANSITION TO NATURAL CIRCULATION FROM 100%,

75% AND 35% POWER (BEST ESTIMATE CASES).

100% Case Figures A-1 through A-10 75% Case Figures A-11 through A-20 35% Case Figures A-21 through A-30 l

6071A--602A

-7I-

t.

g r

o i.

a l

l

/

/

~

r e

/

i p

l

/

b

/

?

2

/

/

/

l

)

/

,o

/

~

/

/

l

/

h

,/

/

/

/

./

.j

/

/

[

~

g

/

Z

/

o u

3

/

r.

o m

o I

i cz n

a ;

,./

a-ws n

l

&,e

,./

1 t

on

/

=R j

O D

ji Q p. e,.3 d./

I e

u Eu i

r=

4 "bOE-

=~

_--w ax P

xm-~2 o

- W w

g C'-

o u) w =w -

9 O

g 'a - w.,

~ v:

u ts. -

uw u>

s

- Cr_

w we of "s #7 v

?

w 5 c2 u-

.C ED2. EL. M E5 1

5 w

nd.

oa

/

m v

5 rr M.E.E.E.E.

w ~"

T e

e.,

u m.m.. n.

\\

P

9C53 4

l N

o w

1 o

'D.^.<i:s s

a o

iT C

3 r.

8 e4 A

N 8

4 h

e i

l-w

.O h

d*033 - Stin1YE3di31

+

+..

I i

b O

+

6 I

i i

o:

i n

O I

?!

1 i

i/

I.

I s

o l l

~

ll

/

g l

i 1

w

\\

i em 8'

og wz i

2 i

~

e-

//

dg i,'

ll i

i tu

-~

wa i

~

ee

/

./

/

m a

-m U

w

~

m ow l

l I

c 8%

w-y u- %

en m

HjH n

/

??$ f

~

'*???

d

/

DQQS w

/

/

ll 8

a s994 c

,/

r i

e l

/

-/

/

jlj 8

\\

i

)\\

/

8 V...-'

. - /

^

.d 0

d'030 - 3BA1VB3dH31 O

o

- 7 3-t

o 4

a

/o:

l

/"

Y

~

/

9 x

l7 I

/

<a li i

~

4'

/

t I

/

I i

.1

,i I

w

a. s i

n w

o e-a g

Q wA k

N JZ

,\\

r se

,'.g 9-

  • \\

n yb

.s i

s ee

.N

/

8, 6

e e

a

.N ir

?

25 =

s\\ /

rt ur w2 w

s.

U o

me

\\7'a /

8 1-r o'

o o

n; y,2 "m

o H

m powo aw u

z-

-e W $55 d

/

om l

s

.. w NM y ~6 l

.n

'A_. B _ -

a m

o f

o F200 E1 r

/

v

--as uw

/

A, A A. A.

m C"QQ 9

/

s

./

8 cDdy "e

o l

s 8

n 8

r

... ~ ~.

w ups o

o o

g o

E 8

8 8

o

.i'030 - 3801V83eH31 0

+ _

}

i

+

o i

I 4

s s

g O

9

\\

1.I

\\

\\

\\

L

'. \\

\\

i

\\

\\

\\

\\.

\\

j y

8 em

~

m-

'/%

'N

/ ' s, s Sg

/

s or

/

a "a

/

/

't s

$8 iN

~

$~

5$

/

wy O

vs

/

m

/

d&

C~Te~_9-Mx o

g"3Go" S'

a -

o If 3 - d-$, d in

,e o

g t;g e'

./

askh5M 5$

~

/

l f

w. g

/

~

~~

D$$55 "m

f

/

a s----

o~

Ehhhhb 55

/

c.Qg$no o*

'.l

'lt!

T g

toctyo

/

a m

a

/,'

8 C

/,,*

. ~. _ _

b

\\

p

/

~%

8_

r

.m

.o O

th h

S E

d*030 - 3&11Y83dH31 O I

l o

8 o

f

)

l>

e t

o h

{

i 2

g I

4' o

l..

o.

1 p

I t

y 8

t t

m m

=-

~. =,

=,

c.,

J 8

- 5 61 r-

?-

1:

5..

4 I

-o cc i

8 rm 2-

.J.

I

==

w

~

x

~

=5

!t:

t,. u-u w e

=-

j

.u r_<wn~

e

==

c

',, a -.."w=

'z Lr-t a

5 J,5 2 %y '

=x i

~

3 p;

Y,

, { *'". '3-w _

z_

2W

[

p.

w u >

-a 4

t'C a f.2 ;__.

E?

tv

~

f.\\

-- 2 <

5 o

> a a s,.2 :,

~_

u o

u.

y

= > w L-z>

c, ex-a t,

t e

e,

n I.,

>o :c=$

r:

4-t

.=

f*

Il '

. 4a c

I s

8

e..: <,: r o.

c

./

o e

/,

~

w=

/

~~'

O

! e e

/

4 ee ca W

h m 4.

(

I

~~N, I

8

.... g l

P.,,

s m

W N

w '

'"h 8

d s.

5 r~

gm

+

4 6 _ _ _ _ _ _ _ _

1 1

1 o

+

i f

.i 1

  • Q Q'

i n

h 5.'

i j

i 1

II 4

h G

i I.

me

.s

m r

  • l l

l j

c:$

8 o

or o

'o

e i

i e-mo t

I wa i

g-

s rw i

<z cr r

!l E=

mm e

6hy$

m

(;

U

$8 W

9

{

8 9~ %

U$

? Whh*

o

$ >-:s

.m-mm

.)

l p

W em w

b

[;M.E..

8 m._

we um am u.

l

( *CO.Su 2d i

l wo W N:'Y i

8 T

C) w

\\

s 8

2 w

g ll f

3-8._

/

V i

r m

.o o

d D30 3BA1VH3dH31

+

f O

+

1.0 A

0.8 "N

/

__~

"v x

G

_J La.

o 0.g 44 m

J t-o e-.

ta.

O.4 e

o w

co

=o i

g_.

o<

cc LA.

0.3

g.0-:

fr/..

-. z......

0.0 0,

100.

200.

300.

400, 500.

600.

700.

800.

900.

1000.

Tile - SEC.

LEODO O----d FUEL ASSY FRACTI7N NLH FUEL ASSY FRACTION

-t}---G BYPASS FLCW FRACTIPN

+

+

. FIGURE A-7 DEMO PREDICTED. REACTOR FLOW FRAC-TIONS FOR THE 100% TRANSIENT

i i

6 1 O

+

I

O I

~

Q u

a f

I bl l

r i

i e

s l

e 1

M r-r i '.

l j

O e

'l l

a Z

a

<D l

x0 l

. i w-j %

2M e

OZ s q d

h i

l N

O i

sa I

8 OO mO l

RE o-i w

%W g

==

d A

p CL I l

O l

v 3

ze wo d

o' l

7 1-um e

CE.U.Q e

8 l

W n

.0 a

8 D

e 8

C e'

E M..

[

e i

/

8 i

ca g

p*

t 8

__ y -. -

d '.'

. /

~

l j

up o

9 9

9 o

d 8

e IN3383d o +

i

" - " ~

+

+

1 1

s q

O 5

g a

o g

1 1

a l

~

l t

a5 8

5 i

m=

4 6N g

j xu t

a 1

=o

-O I

e Q

Oww Ch s

s-ww S

O O

w l

W gM -

5E mw QJM to

, th wf ow c-

_ _vo

.??VO E

9

.o h

>h,h,h C

wer s &

D e

C

@c. a E

r 8o 8

N g

E n.

W -s n

e d

.i 03G 3Bn1YB3dH31 l

l'

+

a.

/

/

/

I i

h*

I t

i, E

s 1

w

\\

\\

fN 8

ad

-i i

w nm

\\

94

<z*1

's a

an

\\

W WE

  1. w af S8 s

a-W

}.

E. < E

  • EE S

O 5

58

/

h E

~

/

./

==

7 i 6 E

l a

s>O S

I u.

I i

I i

'i 8

g S

U 9

o o

o o

o o

o r

a ISd

+

m f

+

i

+

1i' l

/

/

s'

/

/

/

6

/

/

n

/

/

s

/

a

/

/

/

/

s*

/,

/

/

I

.oo I/

/

m l

//

/

/

1I

/

l I

4 8

w" r

/

r-

?

/

nw

? j C1 w

i e

8 4

i s

  • f Ing.d g 3

"Wre =d

/

~

sp lf i

u

~Cnnd x4 dW Us2-

-d cu$

f

_w2 1

o e

i m

MS "*

WW

_- c" g>33 WE C

03___M I

S" Y$55@ cmd2

' "???? "m t@99u EW i8I d5 s

b

'99YY 4

)

N 8

8 l

\\

w tr o

O l

l

~

3 d'030 - 3801YB3dH31

+

)

l l

i

+

+

a

/

l l

l' l 9

' 9.i 1

i !/

/

E lI

/

i ;

/

8

/

=t s

m

/

w l' i k

O N

l',l O

/

n e

w i

o,

I li

/

  • l

/

l ow r

?.

8 Ok.

i en 55

/

l

/

E gm

~

I

.L. &

/

,9 I

/

j u

La mg gg wg@

c-g

/

/

8 9 wt Sw 8

rz

/

W %.E..dW SH

/

/

ou Eh_lgl WE

/

/

m.,,

g e

ena aw l'

/

5%%%

55 t@99 l

l m

/

u 8

W9Y w

/

o a

/

D

[

O

/

C 8

N W

l 8_

/

../

/

id k,C... * *. j/

I e

a g

nV

.O O

O d

d o

8 3

8 8

8 8

O

.f 030 - 3Bn1YB3dH31 o

+ _

a

=;

ft

~

Q

/ s

/$

s.'

h

'e

/#

5 ff I

ll

=

e 1

f,e 8

/

e r

=

.e1 2

~

O O

~

s.\\

a s

\\.

}

w s.\\

a W

Cb g

8 t-

. g

==

a s.s be M. 'f"*.

3. '

e= H

.j e~

L\\

w

_. g O

U

$ h$

o LJ

~5 U

6 ~ *e C O wp s,

f O

~ n

m yd UR gg g

e N

z-C

/

ws<

u, w

        • N wi CWo yH t

=

Y

~- ;a.Y. E

~= =c

  • y n

O w

ac h

O "E. E. 5 5s. g 2w v

a i.A memm

,R._s C

q e a

>$OM IM C GW

  1. )

e t M

n

@0 M

e K

C f

...y

,a-8

/

O f

a 445$

  • m,*%-.~

I I

1 C

O O

O O

O O

O y

CD C

v N

CD CD CD 3

m w

d*030

  • 3 d Td'M1

+

+ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

l l

+

o a

p.I

\\

f o I

\\

~

\\ l

@i NJ 3.

,t i

i i

\\

l 1

\\

\\

/

/

's

~

g

/

/

w

/.

g g.

x 8

w 4N r-H

/ N s N

/

N

~

e

,\\

/

8 cw c.

[

\\.

J I.

or s.

e w

/

s.

G b

4 b cy cx e

/

bg >> <EN

=

/

/

I U

"w w

dW C-E 3 6 t-

/

o 9 d %;5.-

L

.I W

wgz mw mm gl 3 - o.y o

~

c d

ww i

i-E m m t; E 7

8 5

E.E.mmum

g'

/

ummamm m

hhbO EE

"' q'.

t-

'*il Is M hO l l'

' ' ~..

L' l

l w

m, o

e O

~

f N

w

,4 2

4 I

a f

a f

6 1

f 1

1 1

t t

I O

O LD O

t-T N

UJ W

W d*030 - 3E11Y83dH31 o

+ l l

1

+

a_

I t

O i

5 2

'l 8

m Ia l e'

~

w

'l r-ww o-r-

u-c il m&

'l' N&

o H be 8

2R d

"5' w l.el wr gCC g

  • m b

88 v

w me it w

,_ ;5 N p, c u.

9 ' k' C &p~, b a.

ll 3=megg c,e a

mm

~d w

a 8

p 0: mm y m m m% w >a-w 8

      • o l

1 v

up l

AR E' "> s w l

I W$

,o6460 ' ",

c o

,e t ta 8

//

c99490 o

i

/

e m

l 7

~

b

(

a.

~~\\

J

,, ~

i4

,m.

___.-,--r-i v

o 3 030 3801VH3dH31 i -

it 6

l v

1 a

i I

d h

i l

I I

}

f I

i 8

w l

l l

O ww l

O

=~

n ww I

r e<

t j

ox l

'H r

m >.

8 WR I

-tw #w I

WW= 5p

. h!

_% g mb $$

E mm am g

N N (L.

W ED g

$ d 'hh[ b$

tm m <

d a w >-

H %J a

O Ok mmz.b.e..?Mb Ww I

o

_u Q O

i i: il ed 8

t o'c'ty 'x m=

r et n

l

~

bz l

8

=

~

4

~

W D

C3 8_

e

.l g

.u.

.o d 030 38r11Y83dt31 C'

+ ~

J

i 4

+

e a-

,e g,

4e se O.

\\?

A R

e e

e e

e e

8

=

e e

O e

~U e

s==

e p

8 O

8 et h

O' e

W e

1:e E

!e a

6 8

u i;

e eH e

s 2-vy e

g J

.E. w m e

m..

he 9;EW WH W O gypg owe e

e E

p; E"um

- a 1

n. =

q g

p e

4 O

.4

  • 1 M e

WIS w; o e

e k$.m.

Ok e

F*-

e e

le-g

.e 4

il

'99 e l

o

\\e Oo e

to

\\,;e II N

N A<

g l

O 8

w N

O d

d 1

j M073 3V3B TV101 30 N0113VB3

+

+

t

< 't.

o

+

e 8

O i

c.

h

~

8 O

i.

j i

I f

l O

s O

CD

f l

l 1:

t o

p Oe-

.I l

It kt in i

3 t t 8

i J

sg l

1 o

A I

i eR l

d

!i!E

+

o w Ud "5

o th 3:.

(D o <n e

a%

x.E oa edd w t--

j W

.o. u, i.-)

l z cN dd* W

>>d zw

o. z

-v ac t--

[

N4 o Im 664 82 8

c c)

. 4 C e

. e r-I C *hd. d w

/

3 N

c w

L u,,

v-b

.i I

~

+-

p

, s

.m

/

,.a.p*

l s

s O

O.

O.

O.

8, O.

O

==

.u emu e

1N3383d

+

+ - _-

+

i.

=

0, m

.b i

i a

O l

,.4 O

~

I t

8 t

8-m m

n m

f n

y I

Q m-

~=

-mm m

ysu-g = m u

=

.w e

a8 7 8dd d

$.ee e N.-. m o

._ o m os c. ? ? ?. ?. w m o

m=

.s e

W O

a. s si m c

wo

+ 8 I C6

D.Q.

p i

8 4

O La x

8 t

8.

~

G

/

8 f

i l

l 1

J

~.

.d i

.l 030 3&l1YB3cM31 Y. _ -. _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ - _ _

4

.+.

0.6

/

0.5

,,,~..,*.

0.4

,O

0.3 E

/

v 0.2

-~

0.1 0.0 i

0.

100.

200.-

300.

400.

500, 600.

700 800.

900.

1000.

TIE - SEC.

I wave 0---b s-MllMARY NATURAL EAD i

s.INTEREDIATE NATURAL EAD 1

l FIGURE A-20 DEMO PREDICTED NATURAL HEADS

+

FOR THE 75% TRANSIENT

O

+

a ll0 a

[

iu g

8 f

E O

w O

a N

Ew aw a

8 n

$% g,-

<o aw*a

=a Bj-~5 M

e"

  • 8d 83 o
gme s

8 ea6S m m, W $s22g Sm p

g***y tii=

d t3_33_

25 m O

mo e

w EEE

='

A$

i 00 t Op$a 5g i:il o-8 i-

\\

(799tf i

o w

&g-8~

O

\\

O C

[

t i

e e

i e

j 8

8 8

8 8

9 8

~

~

r.

~

d*030 - Sun 1Y83dH31 0

+.

\\

8 O.

/

e

/ O 9 l I'

o 1.

/

[

f i

l l

1 l

l I

I o

j i

g l

I

.I g

e O

I l

l 1

O h

r-l t

w

/

i j

c-wm o

J=

0 e'

/

O cc JW i,.

/*

8 Qun m

g

>M

~

Q H W

g f

>=e.-

w

$W l o.

~

1- ~

U$ d r!

'i u

0 EH LU O M

(:.

8 8dBcd ee cu cc e u m

./

Ed W

=AU HM 8

.J

.x

>*uw 8

gg e'

-[

h k

OD xs wH O

-u cc <

    • - M @

O cu ct e

ggs j

v w

c a.

.I MMM.M EE e:/

5009 e

yyo-f

~

?

8 vo.o.k i:

O Y

W e

Cf D

Q e

8 N

)

8 o'

?

(

  • e-3

,O O

O u

a 8

g O

f 030 - 3801YU3dH31 o

+

-9 3-

O

+

's.\\

/

~

,\\

d

'R \\

/

'.\\

/

m s,

s\\

/

's \\

'/

's,\\

./

.s O

s\\

=;

s s\\

-e e

'd

  • i

)

'f N

=

)

Ow

\\

l wwan 1

1 s

i O Cd j

1 I F-i g

> D '.

N in i.

N

/ lg

<M O

W Z L.J

-/

OC

.O[2J UW O u) 5e U**

i o

Q W*

n o

, e' Q u.

=

L2

/

dhUp HW vw

/

e' H

g

-C

)

QO j

-{->

wg O

ec O

V2 U C. M v

-- a w

bbbb fh

\\

ww E&&&

QW l

m

/

g

.' e.

~

c494 a

a

,*l l

wW D

/

e

,/

~

8 CJ i

/

/

O O

s

~

, ~ -.....

~

g e

o o

o o

o m

e o

N o

r-N N

b=

N N

  1. D30 - 3801Yd3dH31 0

+. _ _ _ _ _ _ _ _ _ _ _ _

o

+

L i

i f

/

gi c)

/

/

L'o i

l s

I s

I o

\\\\

}

/

/

( '\\

/

~

/

O 2

8 chi

\\-s

/

,i

~

cz r-ww a~m

\\

/

,/

a<

s

's.

/

Om or 8

-x.

o mm o

o gm f

zw o=

/

A.

v vH

~

w

.w o E2 mm w llEud=8 o

s59-5 e e e'

8

/

s gg

,/

h, h.i:9 SR

~

-s..

m.

/

O 5E.EEMM 9 7777 9g=

,/

[

mmmmmm ww

.hbo 6

c -

..sl n

~

o e

l/

's 8

. y!j:0 d

t-w f.a

's m

p e

a

~

1

.s 8

l a

i

../

/

8 7

.O o

o m

o 8

E d*030 - 3801YU3dH31 0

+.

-o o

800.

i

~

i!

's

\\

s l

\\

I

'(

A 7so. <

,/s

\\

l s

l l

ls

\\

\\

u.

l

/

s

/

s

~Q I

'\\ '

~

O 1 W

700' (ygV

' ' :. :. -]-%-...

~

y Dg g

/

4 i

//

T w

./

650. h l

i

,1 i

O w

~.

..,,,,g,,,,,

~

600.

O.

100.

200.

300.

400.

500.

600.

700.

800.

900.

1000.

TIME - SEC.

LEGEND 7

7 S-RV IM ET N3Z TE:*

-O -~-~O RS' LONCR FLEhtM TEMP

-D--- O-FUEL ASSY TEMP t TAC) - d-FUEL ASSY CUTLET TEW

~ $' b 8 (( N M E JRE O

+

+

FIGURE A-25 DEM0 PREDICTED REACTOR TEMPERATURES FOR THE FUEL ASSEMBLY GROUP FOR THE 35% TRANSIEllT

..,s o

i l

800.

/

l 750 r.

l

-I u.

_1 o

g

-l w

700 s'

\\

e-

<e w

e-Q.....

. s. g. g. g...g.......

650.

(W

. o......

GOO.

O.

100.

200.

300.

400.

500.

600.

700.

800.

900.

1000.

TIME - SEC.

LEGEND S-RV INLET N0Z TEMP O- --- d" S-RV LCkER PLENUM TEW

-D... G-NON FUEL ASSY TEMP I tac t

-o- - d-NON FUEL ASSY GUTLET TEW

-si -n-RV OUTLET TEMPERATURE

+

FIGURE A-26 DEMO PREDICTED REACTOR TEMPERATURES FOR THE NON-fun ASSEMBLY GROUP FOR THE 35% TRANSIENT

..T

~~

c

O

+

1 8

O

e t>

Q P

8 o

is le le e

b

~

e

=~

, e.

'-=

w G

oz

_J <

e w Cc 8

H

a N

e o ae t'

Hw om

$. z i,

u

.eo cc =

ww W

V' kNU o"

O

dM w cc i.

v cr Ho ta

<>G vL

'. *e g

xu.

'Qg om wn c Jd C

w' p '^

"a.--

O

, D

~Q oo O

T

C <

G w ct 3.

w c.o c

L._

j.

I o PQCl N

N t

t s

O t .l.

oO C*h w

=

cc

n

~

D

e C

w e

8 N

~

\\

l

.o O

't.

i J

i O

O O.

O.

v.

g o,

O O

O o

o l

~

401.4 3Y38 'lV101 30 NOI13t13 1

O

+ 1

100.0 g

,c1 l

14

\\

i I

i 4.-

3 g'

10.0,_

A

__.D L

ut i

v. g

\\

l e

3.k1 g

z p

-D---

y l

5 0:

I I I

_4 14 g

1

.t t

....O---

a.

i g

8 1.0 m

--, N - ~',

ix 3

a f

A V

i N

i g,

y 5

l i

(

W N

\\

__s 0.1 O.

100.

200.

300.

400.

500.

600.

700.

800.

900, 1000.

TIME - SEC.

LEGEND

-7 7 - TFERMAL PS ER O-----O PRi VOLUME FLOW PERCENT

- D --- O - SEC VOLUME Fu m FERCENT

-d- - d-AIR FLSW

+

+

FIGURE A-28 DEMO PREDICTED POWER AND FLOWS FOR THE 35% TRANSIENT

+

+

+

+

8 I

O.-

6 I>

9 8

0 L

s a

f a

e I

a

~

i f

6 f

O l

8 8

I l

I g

iH d

55 u.-

l hN f

CEi i

wu U,e l

s Cm m

l O

Q wubh I

ww H

oe a

O U

,f,, g_

WO e

14 gg O

0) p cr O

O sd D

m Qu

..m mz W W >- > >

WD

==UU

.k.N..

h O

j

~

O co m m. u.s a

o a

~

"h@lY T

E g

y s

x D

e sm u.

8 8

N O

O

~

?

I

=

A I

1 f

I eg f

d O

O O

O O

O c) p N

O (3

b P

b Q

Q W

d 030 3P111VB3dH31

,g,

o a_

\\

\\

8 1

i.

8 m

w m

J 2' 2-8

< BR 2" LO

  • i a"

/

a we a v'..

/

U g;

d 5

Ee Wg wo e t..

W

'u, W gg<

Oa E<

/

w ww OI 9

O V

EE

/

4, 4 w

t&

a0 a

/

m i.,.

t@

l I

/

\\

\\

8 i

y

/

1 I

i 9

8 g

.i 5

.'N O

d d

o ISd o

+

-101-

I APPENDIX 8 FFTF UPPER PLENUM MODEL A cetailed analysis of the FFTF Rerctor upper plenum was conducted to prcperly calculate the sodium temperature at the plenum outlet nozzle. This appendix describes die plenum mixing analysis that was done, the results of that analysis and the impact of fluid stratification in the reactor plenum on natural circulation flows.

REVIEW OF SCALE MCCEL TEST DATA The results from a number of FFTF scale model, plenum mixing, tests conducted at ASL were reviewed in order to obtain the best data for input to the OEM0 olenum modal. Unfortunately the tests were not designed to specifically examine natural circulation. The signif" ant limitations were:

1)

Scram transients were simulated with flow decreasing to pony motor levels

(#10%) and with fluid temperatures, entering the plenum, continuously decreasi ng. Flow during natural circulation is typically as low as 2%

and the fluid temperature from the core is tenporarily higher than the plenum temperature (see Figure A-5).

2)

Flow in the tests was simulated to enter the plenum from a single mixed central flow channel. However, in FFTF hotter flow from the core area enters the plenum via flow tubes in the instrument tree above the core, but cooler flow (about 13% of total reactor flow) from outer reflectors, shields, and the bypass enters the plenum at the icwer horizontal baffle level or at the exit of the thermal liner. Not modeling this distributed in-flow was a severe limitation of the tests since it precluded stratification during normal operation with forced circulation.

Notwithstanding the test limitations, the data infecated that the plenum did stratify early in the transient in the ANL tcsts, the active mixing region appeared to be reduced to as little as 35% of the initial volume of the plenun. During the scram, the plenum was divided into two distinct mixing

-102-6015A-602A-(51478)

regions; a hot upper region, with a relatively cold lower.egion where the fluid entering from the core was mixed with the plenum fluid.

A very slow upward movement of the interf ace laye; was noted, with some of the upper region fluid mixing with the colder lower region.

DE1 AILED ANALYSIS The upper plenum was therefore analyzed in detail to determine the sodium temperature transient at the outlet nozzle for a transitio.

' :atural circulation. Four separate codes or models were used as fol' s:

1)

A FULLY MIXED SINGLE MIXING N0DE: A simple standard representation of a well mixed plenum.

2)

PLENUM-2A (Reference B-1): An empirical code developed to correlate the various upper plenum mixing tests conducted at ANL.

It has not been validated for a transition to natural circulation event, where the tore outlet temperature is greater than the upper plenum temperature for part of the flow coastdown period.

3)

VARR-II (Reference B-2): A generalized two-dimensional mixing code. The region to be modelled is split up into discrete nodes and the equations of motion and diffusion are solved explicitly for each node. This code has been found to correlate well against a number of mixing experiments, and has been used for CRBRP design analysis.

4)

TEMPEST (Reference B-3): A three-dimensional mixing and diffusion code general enough for any geometry. This code also has been found to correlate well against experimental results.

A comparison between the reactor vessel outlet temperature obtained using the fully mixed plenum model, the PLENUM-2A and tbc "cRR-II results is shown in Figure B-1.

Figure B-2 shows the VARR-II and FLs.aa-2A results for the first 160 seconds of the transient. Figure B-1 shows there is an increasing severity in the outlet temperature transient predicted when going from the

-103-6015A-602A-(S1478)

Pt it NATLRAL CIRCtA.ATIGN I

i I

I i

9.

i

<-<,--t 950.

4 I:

l e

o

.'.g 910.

t k

i w

v

^

H Q

Y 870-a '.

/

+t..

w p

~

,. " - -Dk

'N g

g 8

n

\\y e @>

' ~., ' ~.

/

g*

.s.

830.

7..

g

\\

f v

.. I m.

.4

~

l

,?-

790.

750.

O.

250.

500.

750.

1000.

TIE SEC9POS LEGDO 6... 6. NEN PLEMM

",tsam.a

- D - - - G VMut-i l - 4-M PLDUI-2A FIGURE B-1 REACTOR VESSEL OUTLET TEMPERATURES CALCULATED WITH VARIOUS PLENUM MODELS

-104-

(

(

(

l i

FIGURE B-2 FFTF OUTLET N0ZZLE TEPGATURE DURING A NATURAL CIRCl;LATION EVENT LEGEND Uniform Core 980 Flow Only

~~-"

VARR-II Fuel, Non-960 Fuel and Bypass Flow

/*

PLENUM-2 (ANL) 4

_N a

920l

^

o' s

TEMPEST - Fuel, Non-a L

A Fuel and w

Bypass Flow 4

h 900

's, w

5 H

880

's 4

5 l

IB 860 a

/

..____-~

/

_/"

's 840 a/

kl

^

^

A A

a A

A A

820 I

I I

.- -. L i

i I

800 0

20 40 60 89 100 120 140 160 Time - Seconds FIGURE B-2

f ully mixed mode, so PLENUM-2A, then to the VARR-II codel. The fully mixed plenum model assumes that the total upper plenum is well mixed for the entire transient. This mitigates the effect of the rapid change in the core outlet teng erature (shown in Figure B-1).

The PLENUM-2A model uses a totally mixed plenum for the initial portion of the transient, when the jet velocity exiting the core is sufficient to carry the sodium coolant to the top of the plenum. When this velocity collapses 3t about 20 seconds into the transient, the plenum is then split into two regions--an upper stagnant hot region and a colder mixing zone surrounding the chimneys (instrument tree flow tubes), W1ere the core outlet flow mixes with the plenum sodium in this region only. This reduced mixing voltr.e results in more rapid transients than the 1-h0DE upper olenum model predicts. The volume of this region then exoands due to the rise of the interf ace surf ace between the two regions. The velocity at which this interf ace rises is based upon the Richardson number ( AogD/p Vg 4 ), the ratio of the gravitational to inertial forces. As this strf ace rises, more of the hot upper region sodium is mixed with the lower region fluid. The peak rate of change of sodium temperature as calculated with PLENUM-2A was 5.5 F/second.

The VARR-II calculations wre done using the two-dimensional,14 x 21 node cylindrical codel of the plenum shown on Figure B-3.

The Z axis is the plenum cen te rli ne. The transient inout for these calculations, including reactor flow rate and core assembly exit temperatures are from a DEM0/FFTF computer calculation. Results of the VARR-II calculations in Figures B-1 and B-2 show U

that the temperature f alls fran 950 F to 830 F at a rate of about 20 F/second during the initial portion of the event. The "saw tooth" shape of the VARR-II curve between 200 and 500 seconds shown in Figure B-1 is caused by a straight line interpolation used to represent the VARR-II results in DEMO. This small approximation in temperature is not significant in the calculation of natural heads or natural circulation flows.

}

A major contributor to the differences between the VARR-II and the PLENtN-2 calculations is caused by the steady state temperature distribution in the pienen before transient initiation. The PLENtN-2 code was developed by ANL from 1/10 scale outlet plenum codel tests.

In these tests, all reactor flow 6015A-602A-(S1478)

-106-

22 l

l l

kC

!.INER FLOW I

I I

l 20 i

l l

l 18 l

l l

l 16 l

14

f

,/

l l

PLENUM VERTICAV 12 CENTERLINE 10 I

J_'

I I

i

-l l

I LJi 8

l l

l C)RE FLDW l

l l

6

/,l[l,\\{d l

l l

%'i,N6)tEl

% s X _ J K lor,r,,n c>

uTLET N0zztE d

4 z (rT. )

>G @X issHi6 etoW I

i A2 XIX!XiX. A-1 IA

/

BYP SS Fl OW 2

4 6

8 10 12 14 16

> R ( FT. ) HORIz0NTAL BAFFLE FIGURE 8-3 FFTF OUTLET PLENUM MODEL FOR VARR-II AND TEMPEST l

l

-107-

was through the core, resulting in a fairly.. 'orm plenum steady state temperature distribution. The VARR-II calculations included the reflector, shield, and bypass flow, which is colder, and enters the plenum at the horizontal baffle elevation. The calculations show that this colder fluid stratifies in the bottom of the plenum during steady state operation. Figure B-4 illustrates the steady state plenum temperature contours. During a transient this c'.d fluid is forced out the outlet nozzle, causing the maximum rate of temperature change to be app, ~.imately 200F/sec at the outlet nozzle.

A check of this theory was made by making a VARR-II calculation with all of the flow entering the plenum through the instrument trees. The resulting transient, also ploted on Figure B-2, is even less severe than that calculated with PLENUM-2.

The VARR-II results were checked with the TEMPEST code (Reference B-3).

The plenum modeling, inflow and outflow transients, and the inlet temperature transients used were identical to the ones used in the VARR-II analysis.

The TEMPEST results matched the VARR-II results within 50F for the first 3 minutes of the transient as shown in Figure B-2.

The TEMPEST plenum outlet temperatures also showed good agreement with the VARR-II results all the way cut to 1000 seconds.

The cause of the rapid decrease in the plenum outlet temperature is further illustrated in Figure B-5.

The figure shows velocity patterns and isotherms at 5 times in the transient out to 40 seconds. The scale of the velocity vectors changes (increases) for the frames at larger time,.

The cross hatched isotherms are the highest and lowest temperatures at the time of the frame--not necessarily the same temperatures frame to frame.

Initially the velocity pattern is a distorted torus. The vectors and isotherms show poor mixing with sodium in the lower part of the plenum. At 16 seconds flow and temperatures exiting the flow tubes have decreased and a column of relatively cold sodium exists above the flow tubes.

The rotation of the toroidal vortex reverses washing this cooler fluid down over the horizontal baffle and pushing the previously stratified cooler fluid in the lower part of the plenum out the outlet nozzle. The temperature decrease at the outlet nozzle lasts from 16 to 25 seconds.

6015A-602A-(S1478)

-108-

_ _r --

1 f

943 V

927_

N d'

r__--

all 895 P79 3E2

~

nna

\\_

FIGURE B-4 FFTF OUTLET PLEllui1 STEADY STATE TEl1PERATURE DISTRIBUTIO!!

-109-

9 8

4p-*.=

y 1 ww 'J E

.l t u's % -

s y,\\e. )

(, <a.

h *$.

}

-)

s N'/,[' 's, 3.,J 3

j i ~ s

{ d

\\ \\'

\\ y, D 9 \\.\\g ifV s v

, j' s

N V.

  • ii. 't t

, ; w i

i 9

f I.g \\bU.1.{$.Aj@r,%i

,f*-

j,,4 V3

[

6

O

i. i. -- /

e s

% j/ p L.

g p

1 A kl T Ns A Ku//,'fj V9-m!

i e -

g. -

D,JW:':

/i

>,k b'[/

' ~YO' ', -j

-5 4

, 4 4%7 D%

  • ^*

l I

'~._ e. ;i

,p t +

\\m,t 2

u_

n~

j!/l[ l C' + ~

i \\.

t,,

A%<

6

.ss_

i pl/3 N i f

?::

4 N.

,e

  • g l

.l/ ' f, N

' ) ; t.3 s

.)

4. (. ('

N, s'

r um-

~. y / + ~,

'E

=

i/g

','J.,

. 3 /, s'..'

1, N,

~ '

m, y }-.=

2 s.

s s, % w ' '*-

-"-*.-).

s

/,s j l sgj /

~

,,.~..

~' ]j '/,,r,f

,/,/ / i

. v.m -

w

, :, /'A

~ *///,1

..,/

~

^-

1

' ^

W

{.

}.

'a=

t-

,*,,/

t i

~

,. 7 "

,.T E

,, M_, N i f j,. 6

}

s.

,,i,-,,

'Ws

_W s' W

  • /

,,?f f

l'

.d f, %<.h d

g,

,*s q\\,

Mb/lpd a.

M.

. t v,

1,

%.)

\\

w f.

t\\

j, >

t' a

,,2.,7/,/, i i+

1

#/,-
-

/

1 ~,

i,>

.N 1

3,

s I

f L./

e

- s. *p ! m'

-9

}

.4 3

ffj Z,,

,.~ <---" j y /,j/ ;

w--,--e

.==

/ q',g/

s,%,.-..~

g

} ;,

..s.y, s w ----

jfj/, 3'/;

b-

,,/-

/ _ /,e,, jr; l s,%

__p v.

n/A b

f'

~ ~

C.

I#

y,# r g

=

-~.+.-t

, u e- - ~~ w,,,// ~ } } ';,

',7' (p[~d 5

{3

.-x.:,wn,] '; ;,g l

.-p. e y

\\N d

.C

<& A g y; j\\

/

j l 3

l/

'kg/ p

-7 I f'

\\

l so e

t i

/

i1\\

,'(.W, (' /.#

I

/

f

.,1 8

t 7

M M,y y/.

%, J,) [ y g /</, ~

}, ~ ~..., n,',.

, N

, [

y

'2

M.-

. =:

~Q q

e Jf,

/

5 3

~ y.. ~.

~

-r y-w y

y g,

~

-~.

=

)ga,p.~ p rq 2

i

i. __ _

mMMmeese g % g g

  • x_'.'a a

i -

i k IN d_

1 -

p_

_-./ l

}......

w.

o a <

  • g

. [. ;p.

.\\

in y,

i, o

~,.

s' s

u e"*

W ggy

.f

~

s n,;sj,

.u f

I v'///

  • ~

_/ fib SB0133A All3013A 01073 S3Nn30SilVWH3H1

-110-3859

The PLENUM-2A code was modified to more closely match the results predicted by VARR-II for the first 3 minutes of the 100% transient. The modified version was called PLENUM-2A (M00). The changes included reducing the mixing volume below the elevation of the outlet chimneys (from 840 ft.3 to 215 ft.3).

This had the effect of increasing the rate at which the temperature drops for the initial portion of the transient. Also, the rate at which the interface surface between the two mixing regions rises was increased to match the temperature rise between 25 and 40 seconds. This had the effect of adding high temperature fluid from the upper region to the colder lower active mixing region. A comparison between the modified PLENUM-2A, PLENUM-2A, and "ARR-II models is shown in Figure B-1.

As shown, there is good agreement between the two models for times less than 180 seconds.

Past this time the VARR-II model predicted higher temperatures than PLENUM-2A. This is probably due to VARR-Il predicting a greater amount of mixing.

The modified PLENUM-2A model, with volumes and mixing adjusted as explained above the for 100% case, was used to gerierate results for the 75% and 35%

power transients.

For these cases, the initial rapid decrease in plenum outlet temperature shown by VARR-II was predicted by the modified PLENUM-2A model. The comparison is shown in Figures B-6 and B-7.

The 35% case (Figure B-7) showed the greatest differences, due to the power-to-flow relation during the transient not being similar to the 100% and 75% cases. At longer times PLENUM-2A (MOD) predicts lower outlet nozzle temperatures because of less plenum mixing than is calculated with VARR-II. As a result PLENUM-2A (MOD) produces conservatively low predictions of primary loop natural heads and flow y

\\

rates.

IMPACT OF PLENUM MODELS ON NATURAL CIRCULATION FLOWS The sensitivity of natural circulation flows to plenum mixing was determined with DEMO Calculations. Figures B-8, B-9, B-10 show the resulting reactor exit nozzle temperatures, primary thermal heads, and total primary flow for a one-node model, the VARR-II model and PLENUM-2A (MOD). As shown, there is a significant difference in temperature between the 1-node plenum and VARR-II.

The differences between VARR-II and PLENUM-2A (MOD) are typically 30 F.

-111-6015A-602A-(51478)

FFTF PLOTS 75 PCT.R PLEN2A 1,

i.

1 t

}

i l

i 4

i 940.

.l n

i 2

i u

7..., -

! ' t ~.. : N.

's 2

. J 4

.f l

'?-

's s

i l

i s

s j

i i

t i

's 900.

s

{

i

~.

0, t

---f - f- -- +; - { -

m, l

t I

-L.

-+

{..

---4 A - -

- - - +

, 4. _.

I i

i I

~

i

.I 3-N.,

4 i

l

{

I i

I 860*

3

- O' t

i a

I l

i d

g

_ [.

L _. l

.__ p m__.s__ g _

. h

--_4

. -- t.

l l!

f7hb I

sk lE a

o Lo E20.

e j

=

j,!

o i

s Y

I f%

a _+-.

N. L

.L s

i Ni t

I i

j bI j

~

i i

IN i

s l;;j q,

t I

i i

740.

l f

l

?

t l

l t

t I

i 700.

O.

250.

500.

750, 1000.

TIME - SEC.

LEGD0 7

7 G-REACTOR YESSEL EXIT TEW. (PLENUM-2A (MOD))

O -----G-S-PRIMARY PN IM.ET TEW

- E} --- G - S-PR IMARY PUW EXI T TEW - S-PRIMARY IHX IPNBZ TEW. D-S-PRI IHX TUENX. IM.ET TEF s

a VARR-II S REACTOR VESSEL E%IT TEMP.

FIG'JRE B-6 COMPARISON OF UPPER PLENUM EXIT TEMPERATURES CALCULATED WITH PLEiM!-2A (liOD) A!!D YARR-II (75% POWER)

-112-

FFTF PLOTS 35 PCT.R PLEN2A 760.

. _. _ i. a l

i j

P

..........:;4g t.

_ _.1 i

l l

i i

_. 9

,l

)

3 77 I

l

l i

l iiii i

t 735.

l i

k k

I f

I i

?

{

I I

I i

i t

t

- +.

i l

1 g

.. _ _. __I.. _ -

. __p 6

t.

I t

L d - -+ -

t i

w t +i a

-_ 4

+

1 I

i I

710.

g

.in a

e

_l _. a_2_.* _ "

g x

/

u s

__. LL_

/'

N t

r

~

,q

\\

._d

__p x{

\\ _u t

_}

W

_,l\\

/

i 685.

3

\\

.h

\\

/

v l

i

__ -.7 _.

660.

0.

250.

500.

750.

1000.

TIME - SEC.

LEGEPO 7

? S-REACTOR VESSEL EXIT TEW. (PLENUM-2A (MOD))

O-

-O S-PRIMARY PUW INLET TEMP

- D - - - G - S-PR IMARY PUW EX I T TEW - d-S-PRIMARY IHX ItMBZ TEW.

-H

-U-S-PRI IHX TUBt0L IPLET TEW e

a VARR-II S REACTOR VESSEL EXIT TEMP.

FIGURE B-7 COMPARIS0N OF UPPER PLENUM EXIT TEMPERA.uRES CALCULATED WITH PLENUtt-2A (110D) and VARR-II (35S POWER)

-113-

FIGURE B-8 COMPARISON OF PLENUM TEMPERATURE AT R.V.

OU M T NO R FOR DI m RENT M NUM MOD M 950 940

\\

s 4

s 920

{

~

LENUM-2A (MOD)

~

l 1 N0DE MIXING MODEL I

s w

[

N d

900 i

8

\\

I t3

.L g

880 i

f 8

i

[

g u

VARR-II 860 s'

~

b

\\

/

h

['

\\

/

84 0

/

/

/

/

~

~

s'

~

820 800 i

1_ _ _

u _ __._ i i

__t

_1 1

1 0

20 40 60 80 100 120 140 160 180 200 220 1

CrrnNnt AFTfD TDIP

I

(

(

I FIGURE B-9 COMPARIS0N OF PRIMARY THERMAL HEAD FOR 1-N0DE PLENUM, VARR-II PLENUM AND PLENUM-2A (MOD) MODELS

.4

.3 l

~~

~

s s'

l n

w G

i S

i a

ob

/

VARR-II PLENUM a

t f

,2 i N0DE PLENUM t'

5 u,

x

/

A PLENUM-2A (MOD) l 8

S

/

g

\\

s

/

g:

,' 4 -

/

.1 A

0 i

i i

i i

i n.

,t_

i 0

20 40 60 80 100 120 140 160 180 200 TIMF, SECONDS

i E

I

=

- ~

E O

m w

a e

?

J G

O C

E v

U W

l Q C

1 4

E

', CO O

N D

I Z

Cr a

z I

w r

I 3

i W

1 a

i 1

4 O

}

4 l

-, C i

{.

mm 1

c Y

l' aa Ls. O

\\'

E E I oE D

wD z

m OZ w

C

<W J

$ N U WJ 4

C 4-u

%A t

m LH i

es l'

S

=

'4

=%

l Ow ML l Q

~L-t O %

%~

~. - y

<O j

f b

C55 dols dxrld l!

um m

O O.

n 1

../'

'fX O

CO CC

  1. 7 '

y D

4

.o b

O 7o O

O O

N N

1 I

f f

- O C

M N

O O.

O.

O.

O.

(!;0113Yd.1 SSYd) M014. E013V33

-116-

I t

I figure B-9 shows, however, that the differences in primary thennal head (which are essentially Caused by a different average temperature or density of fluid in the 12 ft. reactor outlet pipe) are not large--the PLENUM 2A (MOD) to VARR-II differences are negligably small. The bottom line of the comparison is the differences in calculated primary Icop flow caused by differences in mixing model. The flows are shown in Figere B-10.

As shown plenum stratification clearly decreases the flow. The flow minimum, just after the pump staps tnat was calculated with the f ully mixed, one-node plenum model is 15% higher than the flows calculated with VARR-II.

It is also significant that despite e30 F difference in nozzle outlet temperatures (Figure B-8) the minimum flow calculated with PLENUM-2A (MOD) and VARR-II are practically identical. Further into the transient the PLENUM-2A (hul) flows are conservatively less than those calculated with VARR-II.

PLEhuM MODELS FOR PRETEST PREDICTIONS, The pretest predictions were made with a combination of PLENUM-2A (MOD) and VARR-II models of the FFTF reactor plenum.

VARR-II was used for the best estimate calculations. The less accurate but nore conservative PLENUM-2A (MOD) was used for the DESIGN case calculations.

The considerations which led to this decision were as follows.

The PLENUM-2A (MOD) model is relatively simple and was easily incorporated and coupled to DEMO. This arrangement resulted in a fast running straight-forward method of including plenum stratification in the overall plant calculation.

Flow feed back effects on the temperature and flow entering the plenum are automatically accounted for.

However, as previously discussed, the nozzle outlet tenperatures are conservatively low.

DEMO, with PLENUM-2A (MOD) incorporated, therefore meets the requirements for a design code--it is self-contained, straight-forward, and conservative--and this combination was used for the DESIGN case caiculations.

VARR-II and DEF0, in contrast, could not easily be coupled because of their size, canplexity and running time. Therefore each plant prediction required the following:

6015A-602A-(51478)

~

I 1)

A DEM0 run (with PLENUM-2A M00) was made to provide flow and temperature to V ARR-II, 2)

A VARR-II run was made to provide plenum temperature input to DEMO, and 3)

A final DEMO run was made with the VARR-II plenum temperature input to DEMO. Differences between the flows calculated with the first and last DEM0 runs were small (Figure B-10), however, the sodium temperatures at the reactor outlet were measurably different. This arrangement was therefore used for the best estimate calculations.

o s

f

~

~

6015A-602A-(S1478) l

i APPENDIX B REFERENCES B-1) Paul A. Howard and Juan L. Carbajo, "Plenua-2A: A Program for Transient Analysis of LMFBR Outlet Plenums," AhL-CT-79-17, Argonne National Laboratory, February 1979 B-2 James L. Cook, Paul I. Natiayama, "V ARR-II: A Computer Program for Calcuating Time Dependent Turbulent Fluid Flows with Slight Density Variation," WARD-D-0106, Advanced Reactors Division, May 1975.

B-3 D. S. Trent, " Tempest - A Three Dimensional Time Dependent Computer Progran for Hydrothermal Analysis," FATE-79-105, Battelle Pacific Northwest Laboratory, July 1979.

6015A-602A-(51478)

~

~

d APPEh311 C TFE REACTOR WCDEL ine purpcse of this appendix is to descrite the reacter eccal and cata used in DEMO for the FFTF pre-test predictions. Along with a cescription of the CEMG redel this includes a description of the interface betneen the DEPO rodel and the rcre detailed ccre nodels used in this analysis, CGRINTH and COERA C.

The nethadolc;y used to derive integral data froc these detailed codes for use in OEMO is presented.

The basic philosophy for taking natural circulaticn calculations with the above renticned corputer codes is as follows. Three codes arc recuired for the natural circulation predicticns: DEFO, COLR A-WC and FORE-EM. DEMO, the systers code, is used for whole plant analysis and provides the bcundary conditions for tne COERA-WC analysis, namely, the total reacter fic* (or tctal reactor pressure crep) and the reactor inlet terperature.

Eecause of the

. hole plant sccee of D040, tne purpose of the code is not to necessarily provide detailed redeling, but to include sufficient detail and cceplexity to accurately predict the o<erall response of tne coeporent.

In keeping with this philosophy DEMO represents the plant components, such as the reactor, with equivalent models with lurping based on nore cetailed calculations. The COSRA-WC code uses the boundary conditions, total reactor pressure drcp and reactor inlet temperature, Orovided by DEMO and provides cetailed whole ccre analysis considering inter.

ntra asserbly heat transfer and fica redistribution. Asserbly ficas ano t2er:al tcundary conditions calculated fren COERA are then input to FORE-2M wnich is used to calculate het fuel pin terperatures.

Figure C-1 shcws the general layout of the FFTF reacter. The significant thermal hydraulic regions of the reactor are the iniet plenue, core region and the outlet plenum. Flow enters the inlet plenan froc the prirary cold leg 0

piping through a nozzle inclined downs rd 45 to proncte mixing. Flox from the 10w pressure inlet plenu enters the inner and outer asse:Olies and bypass thrcJgh the co'e support structure consisting of the basket. entrance pler.un,

~

~

63094-502A(S1475)

..~..

- -~.

ri

/

,, I' r,i i,:',. o__2 f

500iUM ib i.'

n 4,

,.;h 0-5 I

i i

l,j t

e i

j..

[1 I

,a n

[;j CU~LET N0ZZLE l

r,j r,,

.. / '

J r.

-. i c '

M t,i

!; a

~

g l-l..al

-_ 2 gLo.p t

G -,j 7 ( 'l _;_ _.', ;

I l

['[

-.n 24

-[ l C Iq c

I s Blifi h

l l

l l

I,N t

l I

1 L

L I

I i

('

h::

l I

f.

I

{\\

m w

w c

I 11 2 i

[

g it

~

c G

Y N

,l R q d r j-

.um m-

~

s u m.,

t h

d8

,j I=

1,

_. m>.

~ >..\\

r-3 f.

) E He

=mj

~

4

=e

!g a.

um

,1

,_y a

z.-

ti
  • i o,t ;y L

I. !,, a- { + :=_. = =.=:.c.T=:

<7,----3

])h'*'d,T 1-==-= w=:=:35 t1

- L.

f

-t_.

4 '.. a, p(c... -...

,3 t

a'

- [*/

t

-- 1 s.

q ikJ d

r,

../

\\.-

D L

h.

/l4 m.( v s f'

L s

'- VENI N '..j s.N4,A SASi'.ET I'

.d-

/

Ne u.

v L... 3 PERIPHEPAL q w.N.-i'91

. i t.

P.

N.,

g, ' DUM /,

PLDiUM

\\

j-s I...u BASKET BASKET r-==',

g -,,1-ENTP 'f'C E LC't! P PE5 5 UT.E Oi?x.C E -- y' _.._"_ =

PLENUM PLil;UM

/Siu.Y.

s

's INLET Pt.ENUM FIGURE C-1 FFTF REACTOR VESSEL ILLUSTRATING THE SIGNIFICANT FLOW PATHS

-121-

~

basket center plenum and peripheral plenum. The total socium volene in the inlet plenum regicn and ccre support structure is 2000 f t.3 At a typical natural circulatten fica-of 2% the mixing time constant cf this regien is 1290 seccnds.

The ccre regien is hydrodynamically a parallel net.ori af th a total of 199 assecolies including 73 fuel assemblies, 9 control and safety assemblies,105 racial reflectcrs, and 11 shie asseeblies. Flc also goes through the shields and bypasses the ccre through the in-vessel storage, the reacter vessel ccre barrel annulus, and the vessel liner.

The up;er plenue is a large free surf ace sodian volume witn a mixing time censtant of 35.5 seconds at full ficw (1730 seconds at a typical natural circulation fic* cf 21). Flow enters tne upper plenum thecugh the instrurent trees and fica tubes above the ccre, from tne cutlet of the reflectors and shields at the hcrizontal baffle and from the bypass at tne baffle liner interf ace seal (BLIS) approximately at the location of the radial gap be meen the top of the assenblies and entrance to the ficw tubes and free the lirer near tre top of the sodium level in the vessel. Fica exits the upper pienem thrcugh three outlet nozzles located approximately two feet above the uccizontal baffle.

In the develeprent of the OEMO reactor mocel two detailed reactor codes, CCERA 'WC and CORINTH were utilized.

These codes provided a check en the lu ping assurptions of the CEPO rodel and were used in the development of integral cata used in the DEMO code.

The CDERA "a*C code w35 used to predict the core aide ccolant and red tercerature distribution. A;;roxicately cne-sixth cf the core was modelled including bypass ficws and pressure losses abcwe and telcw the ccre region.

Detailed te peratures and flew districuticns were obtaired fcr the two fuel open test assemblies (FOTA's). The COSRA model used a hydraulic pressure drop network for the pressure drco above and belo the core

  • pin bundle region".

In the pin bundle region the conpleta set of CCERA equations were selved. Tne fica network for the regicns above and below the pin bundle includes only the

-122-6309A-502A(51478)

stctic head and orifice or friction f actor pressure drop.

The loss coefficients were calculated from the known steady state pressure drop between two points and the corresponding steady state flow rate.

From the basket entrance plenum, four groups of flow were identified-each group sharing some common inlet and exit loss coefficients. Group one consists of the reflectors and the inner and outer radial shields. Group two includes all the inner core assemblies except the open test assemblies. Group three is the two fuel open test assemblies. Group four includes the bypass flow patn. COBRA does not model the lower plenum thermally--it uses the DEM0 calculated lower plenum temperature as a boundary condition.

The COBRA-WC code modelled the pin bundle region using a generalized subchannel method where several standard subchannels, os

m. : assemblies were lumped together to form a single subchannel.

Figure C-2 shows the noding scheme used to model the bundle region. A total of 39 assemblies were modelled in various degrees of detail, including 15 fuel assemblies, 3 absorbers, 1 in-core shim, 1 peripherial shim and 19 radial reflectors plus a bypass region.

The 9 outer reflectors were modelled using a single flow channel; all of the other assemblies were modelled using one or more subchannels. For the section of the core modelled, the two F0TA's in rows 2 and 6 are modelled in the most detail.

Thirty-seven subchannels are used in each FOTA model. The other fuel assemblies were modelled with seven channels except for a few which were far removed from the test assemblies. These assemblies were modelled with only one channel.

Each of the radial reflectors and the in-core shim were modelled using a single channel. The absorbers were modelled with seven channels.

In total, 212 channels were used for the entire model. The bypass was split into two channels, one for the thermal liner and one for the annulus including the invessel storage region around the core barrel. The BLIS near the top of the annulus was not modelled because earlier COBRA sensit'vity studies indicated that while the particular model used for the BLIS had a large effect on the bypass flow, its effects on the flow through the core channels was negligible.

6309A-602A(S1478)

-123-

R Av,N E

m S

R p

R f

3_

O O

T T

C C

A E

E

/ Z vs nv L

R F

E R

F T

/

F A

mn,*';;,Vs.

vs 7F 3

F E

O a/

fa L

2 E

D

(~

%j,O M

S b

C E

R J

50v mV W

E IV A

R R

D h Cl(I B

D N

O

)

C I

F O

b NVs.

Z NO I

I T

L AR T

e-S

,l U

L r.

L I

2

-C

/

%k9(/

E

~

R UG I

F

'-dv r

?t 1,

lll 1

l

The significant features of the COBRA model are its detailed analysis of selected assemblies (the FOTA's) and the inclusion of inter and intr,. assembly heat transfer.

The CORINTH code is a themal hydraulic code which models flow redistribution between parallel flow paths. Each of the flow paths are lumped themally and hydraulically and there is no heat transfer between flow paths.

This model is a generalization of the FLODISC code for themal hydraulic transient calculations.

The CORINTH model of FFTF includes a model for the inlet plenum using the DEMO calculated reactor vessel inlet temperature as a boundary condition. Also included is a model for the outlet plenum. Through the core region, CORINTH models 22 flow paths representing 4 open test assemblies, 2 channels for the lower bypass, and 16 core channels. The 4 open test assemblies represent 2 fuel assenblies (the row 2 F0TA and the row 6 FOTA) and 2 non-fuel assemblies (the V0TA and A0TA). The 2 lower bypass channels represent the lower liner and one channei lumping the invessel storage and the reactor vessel core barrel annu'us. The 16 core assemblies represent 6 fuel assemblies and 10 non-f uel cssemblies including the in-core shim, peripherial shim, 4 reflector channels, safety rods, control rods, the inner st'elding and outer shielding.

Above the core all the flow tubes are lumped together into one lumped chimney. The BLIS flow is modelled and the upper liner is neglected.

The DEMO reactor model lumps the flow through the reactor into 3 flow channels representing the f uel assemblies, non-f uel assemblies and bypass. Models are also included for the upper and lower plenum. The power generation model includes neutron kir.stics. and decay power. The purpose of this appendix is to describe the ther%1 hydratlic representation of the 3 reactor flow paths modeled in DEFO. Details of the upper plenum model can be found in Appendix B, and the details of the power generation models can be found in Section 4 and 5 of the report.

The thermal modelling of the three flow paths in the OEMO model was relatively straightforward. Figure C-3 shows the 3 DEMO flow paths and how they were 6309A-602A(S1478)

-125

l i,yV Wr.f urs ' f',

ru i.f~ ' urV Jf.

t A

i i

UPPER PLENUti i

I i

l I

i FUEL ASSEWLY l

i

\\

l NON-FUEL BYPASS

/JSSEWLY i

I I

I l

i 1

i i Fi m -Tube

[

fl 44.4 in.

j.

(

l L!

~l r_

23.45" ~

v h,

i l

j Uccer Assy. i Waper jj i

l I

42 in.

I Non-Fuel !{!

55L82 in. ;

j I

l t

v r-i i

,d A

i Jnuaud_

Refl. 6.5 in. g t

l f lI g

i Bypass r- ~ - -

--l Paflectorg ;Q 124.75 in.

Co r'e i

g i

Shield 1

36Jn.

80.95 in!

1 I

l' l _ Re fh 6. 5 in.

i 4

i r

(,

F-l 7&er Assy.-

i l

I g r; 4 r--

g B

,, y, '

33.Qin.

ig m

i o

s l/

\\

"I Ple'un Plenun s t n

plenun 57.75 in.,'

557.57 in.

57.75 in.

s

~

I s

y

- -m s

/

p

/

LU.iER PLEMU?4 s'

4

'l/

/

J FIGUPI C-3 ILLUSTRATION OF DEMG mDELINr, OF FFTF REACT 00

-126-

split into tarious thernal regions.

The fuel assembiy channel was split into 4 thermal regions. The inlet region having no heat generation was assumed to be at the lower plenum temperature. Thr fuel pin region was split into two sections the active core and the upper fuel assembly which includes that part of the pin from the top of the upper reflector to the top of the fuel pin.

For both of these section calculations are made on the basis of en average fuel pin. Dimension for these regions were obtained from FFTF fuel assembly and fuel pin crawings and are consistent with IANUS data given in HEDL-TC-556. The final thermal region in the fuel assembly rhannel models the handling socket on the fuel assemblies and the flow t oes.

Data for this region was obtained from FFTF drawings. The sodium in the upper plenum above the core and between the flow tubes was lumped in witn the flow tube metal to simulate heat transfer with the sodium in the upper plenum.

Because of the similar geometry of the assemblies modelled in the fuel assembly channel the DEMO lumped model correctly represents heat capacities and heat transfer coefficients in the fuel assemblies. Comparisons with the more detailed CORINTH model during a natural circulation transient indicato the DEMO lumped model is good for the fuel assemblies.

The non-fuel channel in DEMO includes reflectors, inner and outer shields, the safety and control rods, the in-core shims and the po spherial shims. The channels in this region were lumped thermally by adding the sodium and metal heat capacities and thc heat transfer coefficients of the individual channels together into one lumped C'annel. This method of lumping correctly represents the total sodium and metal heat capacity of the region but does not take into account time constant variation. between channels due to differing heat transfer coefficients. The variations in time constants was not a problem for the non-fuel region because the main portion of the flow is through the reflectors and shields both of which consist of large sodium volumes with long transport time constants at natural circulation flows. Because of this geometry, during a natural circulation transient, there are only slight differences in temperatures at the lower non-fuel region exit (approximately the top of the active core) and none at the exit of the non-fuel region.

Comparisons of temperature response with those calculated with the detailed modeling in CORINTH confirm the validity of the lumping used in the DEMO 6309A-602A(S1478)

nodel. Had the differences in time constants between the assemblies produced significant differences in the comparison, an optimal estimation technique would have been used alcng with the CORINTH results to develop equivalent themal prcperities for the OEMO non-fuel channel model.

The themal behavior of the bypass can be handled s aply because this region is unheated and has a large time constant (500 seconds at typical natural circulation flows) which makes this region essentially isothemal during the natural circulation transient. DEMO models this region as a large one node mixing volume.

The lumping of the hydraulics of the parallel network of flow paths into the 3 channels modelled in DEMO was not so straightforward. The 3 channels are coupled by the total pressure drop boundary condition measured from the inlet plenum (elevation at the centerline of the inlet nozzle) to the outlet plenum (elevation at the centerline of the outlet nozzle). With this model the flows are allowed to redistribute so that the total pressure drop between these boundary conditions are equal for all three flow paths. The total pressure drop in each channel consists of two parts; 1.) the static head portion wtiich is a f unction of the sodium temperatures (density) through the flow path and 2.) the dynamic portion which consists of the friction and form losses through the channel which are only a function of the channel flow rate.

(There is a slight dependence of the dynamic pressure drop on temperature as it effects the physical properties of the fluid, but this was found to be negligible during the transient).

The static head portion of the pressure drop is de' ermined by the sodium temperatures calculated with the thermal model of the channel. The lumping of the flow dependent friction and form loss tems for the parallel network used in the OEMO model was best accomplished by input of a correlation for the dynanic pressure drop versus flow for each of the three lumped flow paths.

The correlation was based on the results of a more detailed core calculation.

The dynamic pressure drop correlation was obtained fran the equation 6309A-602A(S1478)

-128-

Total - OP 0

0 Static Dynamic

=

DEMO COBRA DEMO where APTotal and APStatic are determineu from a series of steady state calculations at various flows with the power to flow ratio set equal to one.

This method matches the total pressure drop betaeen DEMO and COSRA.

Any slight differences in the thermal model are compensited for in the dynamic pressure crop correlation. Although the calculations were performed at a power to flow ratio of one, the dynamic pressure drop correlations obtained should be generally applicable to any case. For a different power to flow ratio the effects on the dynamic pressure drop from flow redistribution within a lumped group of channels and interassemlby heat transfer between assemblies within a lumped DEMO channel (which is a function of the poner to flow ratio) would not be included. While the power to flow ratio deviates significantly from 1.0 during a natural circulation transient, comparison with CORINTH and COBRA have indicated that the correlations are not significantly affected and are adequate for natural circulation calculations.

Tne 'uel assembly dynamic pressure drop correlation is shown in Figure C-4.

Also plotted on this figure are the COBRA data at steady state power to flow I

equal one upon which the correlation is based. Data from CORINTH steady state power to flow equal one calculations, and COBRA transient calculations are also plotted. There is generally good agreement between the data from all sources. This indicates that the functional dependence of the dynamic pressure crop correlation on power to flow ratio is weak (inter-region heat transfer and intra-region flow redistribution are not significant) and provide confidence in the DEMO corr!1ations.

The non-fuel dynam'; pressure drop correlation is shcan in Figure C-5.

Plotted against th's correlation are the COBRA steady state points which include inter-region heat transfer from the fuel assembly channel, the COBRA transient calculations which do not include inter-region heat transfer (because there is insufficient time for it to be significant), and CORINTH steady-state-power-to-flow-equal-one calculations with no inter assembly heat transfer.

Since DEMO matches the total pressure drop with COBRA in developing 6309A-602A(51478)

-129-

o I

i 6

I i

i t

i 1

-. ~.

't i

l n-i.i.

i i

r i

i i

e

'l l

1

-l i

{l p

j t

t

_ML l._ j

,3 i-y l

l ll---}i---

,y l

l 1

+-

t 3

w 3.9 o

- " +

.#_ N m

e i

CL i

t e r ix i;,

N N1 l

I.

s t

N.

a.

+

~-v

'i-j i.F i

.q

. >$4

'.TT j;iil!:.;.ghir, M

r

~

~

_~l M

I;

..e wl4.

.,: p.

I i-

! ~

1 w

,1

..l..i i

. t.;...

-d m.

l i

i' u.

e. ?

-1 :

).

i

,e g

g il ?-

i :
j.:

i

.,n....

.- j

-j

,:=-

j g

p. 4

.u.-

a u

CL 1

i i

1 l

g I

i

- 1 i

]

7 g

g_

_p

._ q_. a g

ij

-pp t

y-i

$?

n ~

GjiniCQE RA TEAN$1ENT c E

w 2

88 l

REEA 10M'

- {

{

i.

a:

.{0F 3 t : :-.

-+-

~

h.

.*.s

-i i.

.8 h. = 1 LL W.-

p

.... :u-4 a.

i39'il LB./SEC.

. __+

iM..

u: c.

. 1..

u

, :nne:.,

. p u nL

' ;;,:.. r x+C:.

ar.~

1+ p;;;.

n:

. 72. -

..,..i.

,,2 0.01 '

O.1 1.0 FLOW FRACTION,II g

F GURE C-4 PRESSURE DROP CORRELATION FOR 1

THE FUEL ASSEMBLIES GROUP J

j

[

-130-4 J

the dynamic pressure drop correlation the effect of inter region heat transfer from the fuel assemblies to the non-fuel would be accounted for in the correlation.

Inter region heat transfer from the fuel assemblies woL'd increase the non-fuel temperatures and decrease the density and therefore decreast. the static head portion of the pressure drop in the channel.

Since the total *eactor pressure drops in DEMO and COBRA are made equal, and with inter-region heat transfer, the COBRA static pressure drop would be less than DEMO, the calculated dynamic pressure drop correlation for DEMO would therefore be less than the actual pressure drop by the difference in static heads. This effect would be greater at low flows because there would be more time for inter assembly heat transfer to take pi3ce and the static pressure drop wou'd be a greater portion of the total pressure drop as the flow decreased. This effect can be seen from the COPRA steady state power-to-flow-APdyn equal-one points shown in Figure C-5.

As the flow decreases, the oints W

decrease indicating that the pressure drop through the channel requires an exponent on normalized flow greater than 2 at low flows. This is not physical and considerably less than the CORINTH steady-state-n ser-to-flow-equal-one Joints which do not include inter-region heat transfer.

It is also less than the COBRA transient points which reflect non-fuel temperatares before the inter region heat transfer had time to take place. When the effect of inter region heat transfer is removed from the COBRA steady state calculations, the calculated dynamic pressure drop agrees with CORINTH ar.d the COBRA transient calculations as shown on Figure C-5.

The inter-region heat transfer effect required that different correlations be used to represent the non-fuel assembly during the steady state tests and the transient tests. The correlation on Figure C-5 for steady state pre-test predictions is almost a const:..t times flow squared. This effectively includes the effect of inter assembly heat transfer in the dynamic pressure drop correlations. The static head calculated by DEMO would be higher because of lower temperatures without inter region heat transfer but a lower dynamic pressure drop would be calculated so that the total pressure drop through this region would be correct.

6309A-602A(S1478)

-131-

o it I

~

~

l

..Ii I

l i i I I i

i l

I jI l ti l i

I i

t n

6_

f 4

_ ____ j-..

___1_1 a_ g {_._4 L.J-

- y --_l..q -,-.!_._l-.. y -

9 I

i t

j 7 ; l T M:

p}m; F

i i i i

l e

i i

.-i

t " - --

s 4

+

i..

3 f! !'!l j_b l j

l k ',_f ' !

I i

1 l l }'

I l _

l l

1' I

l l l 4 'f f l j l!! I 3 A l I I " f, '.l_ l_' ' i !'~ __t_ -._ f,-l -.._ 4.4j s.s. i_- + m 't I i I l -l l i l l I l l C-i- j 7 c4 i 3 t S 13 "%i l l.l l ,t l f s 'O.,g [_ - { l g 's t _ _..- 7._) _i C-i.s A - !, l ' i i Ii

r..

j j l %.!, li ~ ~ ~ I w i i } r e a ,I 4f,' f- - -. !~.e l .1 H .f f b '.L: h --l' t i i M _T !_ i .i j i i i. i-i i y j. 1 - -1 I t 1 I 4 j l 4 I u I'- .]. j l i' t w f I l CODR/tr-STEADY-STATT. } i l I 4 ? i.. I > > l ! I CD I' v i d-1 M --td G MITH INITDEC[C!OIfdCE,hd 1 s 9-cpa-qRgs.Itat). - y l }. = = a C.0RItG H I! l __ij._ l i t + g. 1.... .::.t ..i I. ... i i a j-. g _... -

.4

-COS _(4MTE M.ESTON-4EATH y .x.~ W ~l ! ~I R[ MOVED) ! l l !! i ! ~} i I I ~ I' li


C9RifEEATICT4 USE6~i"O" [5TE DY

^ lY l ~d ! i 1 -SINE; EMEMCU09S 8 r ._:. - bCQRRELATIOil USED ;FORi l .,d' i . -. 1.i .j- .l. - g.

q. p ;-

TRANSIFNT PREDifl.lDN$_ ! ' {.]! ; f - ii! =. Fl!0W.. .ij.l.) 2I^

1.... j i
r: g
g=; :gru:.739-j g cg;p 7-

, p: ~ i. ! .t t-J't-i. e q j - ; ....,., i. 2 u, i o 0.01 0.1 1.1 N0!i-FUEL ASSEMBLY FLOW FRACTION, II FIGURE C-5 PRESSURE DROP CORRELt.YION FOR THE NON-FUEL ASSEMBLIES G0RUP l I -132-r 1

The transient case presents a problem because the inter region heat transfer effect is not important during the initial portion of the transient. However, at 150-200 seconds this effect becomes significant. It is not possible therefore, with the adiabatic region DEMO modelling assumptions to accurately model the non-fuel region pressure drop in both the short term and long term. For the transient pre-test predictions the interassembly heat transfer effects were not included in the dynamic pressure drop correlations because this would be most accurate for the first 50-100 seconds of the transient. The 50-100 second time is the most critical portion of the transient because the minimum flow, used for the design criteria, occurs during this time. The effect of neglecting the interassembly heat transfer effect during the transient would be to decrease the non-fuel channel flow by 20% and the overall flow by 2% et latter times during the transient. The dynamic pressure drop in the bypass channel was modelled as a constant times flow squared. This is physical for the geometry of the bypass region and agree well with both COBRA and CORINTH down to 6%. Below 6% small differences in the upper plenum models and the differc,ces in the elevations of the liner and IVS and RVCBA, modelled separately in COBRA and CORINTH prevented good comparisons of bypass AP between DEMO, COBRA, and CORINTH. As a check on the thermal hydraulic lumping asst:mptions in the DEN 0 model calculations for a natural circulation transient were compared between DEM0 and the more detaield COBRA-WC code. The transient used for this comparison was a transition to natural circulation from 100% power with 24 hours of full power operation decay power assumed before the trip (test D). Comparisons were made of both the thermal and hydraulic responses of the models. Figure C-6 compares the sodium temperature response of the DEMO fuel assembly channel with the COBRA mixed mean temperatures for the fuel assemblies at the top of the active core. Good agreement between the two codes is obtained for the first 100 seconds while the flow drops to its minimum value of 2% of rated conditions. The sodium transport time between the two codes is equal as the peak temperatures occur at 130 seconds for both calculations. The peak temperature in COBRA is 200F lower than DEMO. This difference was traced to inter-region heat transfer from the fuel to the non-fuel region. -133-6309A-602A(S1478)

f 8 q ~ / / / 4 / / / I o h l . 2 5 / $w / N / Q d" C 8 f mw a u o '- E / e Y: O <3

u. -

U sw I o= -n \\ E 'o- \\ G a a o "b "W N $P \\ m 8s N v< s x - Se @M .A g <= 25 wwok Y

u. w or 5E

~ ~ L't Eo Sh SE e O _ S M8 C <\\\\.. \\ on S I s 8 8 8 8 cn en m m ~ (3.) S3B01VB3dW31 NV3W 03XIW S3IlS43SSV 1303 -134-

Figure C-7 shows a comparison of the DEMO non-fuel channel sodium temperature with the mixed mean temperature of the COBRA non-fuel assemblies at an elevation of the top of the active core. Again the temperatures agree well until the inter region heat transfer *

ignificant at low flow af ter 100 seconds.

It is obvious that tne differences between DEMO and COBRA are due to inter assembly heat transfer because of long transport time through the region and the power generation and flow rate during this time. The AT in the non-fuel region, with inter-region heat trar,sfer from the fuel assemblies included, is 150% of what it is with adiabatic regions. This indicates half as much power enters the non-fuel region by inter region heat transfer as would be generated in that region by fission and decay power. Other cases show that the effect of inter-region he&L transfer on non-fuel tempeatures could be even more significant at steady state. Figure C-8 shows the mixec mean temperature of all the sodium including the fuel, non-fuel and bypass at the top of the fuel pins compared between DEMO and COBRA. This resembles the fuel assembly comparison. The peak temperature occurs at the same time for both codes-at 180 seconds. The DEMO peak 0 temperature is 40 F greater than COBRA. This comparison shows that all of the heat leaving the sodium in the fuel region is not immediately reflected in the non-fuel sodium temperature rise. This is because some of the heat leaving the fuel sodium must go into increasing the metal temperatures in the non-fuel region. DEM0 will therefore calculate more heat input to the upper plenum than COBRA early in the transient. Figure C-9 shows a comparison between total fuel assembly flows calculated with DEMO and COBRA for matched plenum to plenum total pressure drop during the natural circulation transient. Since this comparison was made, changes in the DEMS thermal model to more accurately represent the metal mass in the upper fuel assembly region would bring the calculated flows closer. As shown, the calculation agrdes well above 6% flow and to within 3% at lower flows. A similar flow comparison for the bypass was made as shown in Figure C-10. The two DEM0 calculations represent the steady state dynamic pressure drop correlations shown in Figure C-5 which include the effects of inter region heat transfer on total pressure drop in the dynamic pressure drop correlation 6309A-602A(S1478) -135-

I l' l i ' i i ~ l I l 8 i a 9 _ ~ i l t 1 } 'f 2 f o () - m y I i i i }. i i l m j l 4 W i >% gt 4 4 I h i O . _. 'f 4 .g.. j _ JC I M[. j. ba m> 1 i t > I WW I 5 I i 'It i i t l 44 t l .t 11.. i 4' yi, g l .,l. W r y g. l WW m l 3% O d. [ 4 i ]) -i wo W c. t i ev O,. _. .l 1 i ~ i gw. <g Q. _.5.._.. 9, z.u t i i e i W W-j i l \\' j I ou i t .,A. '} g 3

9..

i i \\ l l ^ 2* W o =.- f. \\{' 1 i i { l I i M -p H. I I \\ O O ,g9 .I {- 4 .. I()..j W = t <w l a i t .i. , - Q, i !u 3 6 k { Ob W t t m g<F-j i. -w i -, f.. _,l g U l_ ._{_ a.. w l i = i 6 t i i M k f . H 1 d. j l 1 t I I i Ob l I s t O E' UC ..,i 4' _ i j o..-.. w -..._ ( + -.9 ....g_4..- i H, CW i l i i L; Z% i I t . a <D i. + I l 1 ( . w...i.. !. *- . W j ) O< . xn .s I . y \\ l i f ww i H CA t I'. j y + 1 j I L 4.J l i i i l I .1. f .. I _,> l i...;..g f o i or o __..._.]

3.,

..q. .--.3.. g) u. .4 I 3 i i .O< r f-J i- - t i W LaJ + E j 1 I l l

  • i t

Z e i f; 'dD 1": CL1J j -i - I l --- j - i - f i l i i EM l l l I i f' e f j } l O "r t t i v j t I - _.i... 4 4'. . g'.... _.. g _ _ _... C. t e i .%i i i. I a i ~ g .) r y. t (- i 4 ( i k[ WI l f II i l 6 s i 5 cr: : ...i -q _.7 p.

w. i j

1 I I C.0 4 I i l i i i l' W I r l I 1 l i 1.., ... 4._.. - l ,o, i I i i l i 1 l 1 i. i I i i

'T t

l ~; i. .1 ] j i _ _...} i j l ..j . 7. _. i i i i I i i I I l d - m o .i 3 i O i O O .m. 1 i.....q. . m 6 i a r ,_i....._.. _ j.. g. a .Q 1 5 g r-- O 3, I i 'I l I {~' ~, .I I } -136-

t. l _...7_ i i j' r .j. L 1. i I 6 O j t .w o .l. i . ~..... i, t. I i i i i l, ' l 3 i v I P ...'.-._..g __ f...... I ,. b.. _..... .f i j ..g t l p i 1 Y. !. . f, I ,i ik { l l. i I l i . : a 3... d... r.. o

p. ~,

q-j co v p l l ~ i e i i i I l l i z _....o. 1 I I l r.c.:. -.. \\ _...._J _. i - I -....,. f 1 i <l . >j. }. i. .{. .}. { 1 3. i _ d2_ L._. L _;g_ _..;... u.. 4 o ._c.._...__.._.

.... -. _ a

.e l .LN i I ~ 1 .a.yj .l { I - of l l g ; [ _. . a.;... .. }-. . l. .j- .q g ~ .g s:::.u.' j . ; K_ .:cg. l l cc vt n.;. w t. 1. l 9 j:, m.__ 4... l _..a.., l l e J i . v, i s 1 t t. om

j :,i :

I .I 2* W j '.i. M .. 'i.... } ' 1..a_ .......;.. g.. --6 5.. i.g L.. .f.\\ l 1' m i .g"Ck Q H. ..t..... 7,- p i \\ o M% .w. $H ... F. _..!

w. - ol.

I O Z. .l. j e-.. ol ... ! W 1.u ! .i $ p: l l + =.m - ~ ' .. w. . =. ....t_.. .e 8Q 1..... ..L. v ~~ ig ........gg 9- .p .I 4 .p.'_.. .I . -.J- !.. . c. w - _4 _-._.. O cr .m __~H i 1 i m' .k . :I,. .1. ,g u4..u.___4._ . L. L

u. w j-

__. A. 4 i x. p m

2. r 1

..l' .I, I OH i i 6 4 H ). .g .) i .. _** %..: ^ }. I' I ~' 't. .w.._. ..L. .A -_. . + -. .e; .I cc, cr t. 1 i 2.cO CL..' j'_.. 9:4g 'i - t l l n W +t ____i__4.__.. L- _. __4 2.. ,4 4 i i t ~.. V. E 1'

{'.

1 i i I 4

co..

4 8 i ..... ~,. A . 2 .1.. o r .W. I cg i i 'j .") .i. i. 0 . i. + i ^ .._.7 l i -..LI.J.J.. - iL _._. .._. L. .~1...,._ .j_..q.. J. .;... a j _. j.-..{.. L i t . I. ; ..i.... 1,. 'I.. .i. i j. .g 4 4-I. '{: j ' j' .__i_...._...g . i ) i l .._.1____ y

1.....

O w 3.. } 4._. i 4 c!. i. i. t j i l l l I ' i .l g . en f.e ', cm. ... co .. on n, n, ) ___.!. _ i

t' 3db'.T1 l

m i m_ u.[

F

. M1.._1Vd NV3... 33X 1W..... 61 tt013V33 :i i W1 l. 4 'A W J. I' F' .Si j.. 1 t

2. _.a.

= -137-t - j i

~ b - i 7.. i I_..T. 1 ' - - -1

]d.

e ~ -l .l t -f .. _ i _.,. _1_. f _.. E 1.,.... _.y-_,. 6 t g ..p. j;.j. .j.. ,p.. L.-.f._.... '. j l O... ..j. 4._ l j s,, 1 e j p } .l-l ..;...i .p.,..i ~.. _. _. _..! .' - _..i i ..p_.+ . __. j p I I _}. ... 1 .1. .g. { e I 4 y __.

g. __ - _. _

i }.. . f,. I .j.._.,. g C' .I. .4.. l i 6 i i i - A _. .I. a -_.._1 _. C' '- _y_ I ... h. _ _ j. .. l. _g .e I -l-l ) J.. ...I. ..p.. l ._L_ f. O _' C g _,y. __. a _. !J _. 9._ _ .p.. ...'l... = g I i .t .'9_.,..._. p .._..c._.. .....w o. %lv. l I t l' F r-{Ll . r t r a g

[

] -'s i- - -i--H 1'.. i' ] l to j t: o! .m j Q' u .:5 : w '.. h. w[ 't' E. ._- Q._ _ . _j. 67 ,,_ [,,,,;j ,j,, __5 i I. ~ ., '. r3 _ _. t-' . g: - .i l ' ! ' i: f }l. ~ 'l ~'i-l~ ~ l? y s a -u l 4{. .m I. Q :...,. ' _ _ _. l.._.

7. _

_. [. .I .I... _. ___. g g._ q. j M ._.7; .v i .t t- 't O CL' 3 - QT .. Q L. **. - e pI.._ _. '. '- li ' $.i_':.... :!,.. !. .. !..W ' I ;t ' '..9_._.. :l Hj. I '}' M .__'_q_.. _.... _..... _. g ..[,

[

,3 a; - N 7 .__... w ; u%_. -._ y! ,e., i.c -Yl HC p -1 + q.- - -- L l-- s> t:It l E -Y. ---Y{ ' 'N k}- @ +- - i. N... -M{ - -k-!. - --- E -- h -C-'- * -- -+ [ y e .1 gg 7. '{ 'l -{ "'f:

3..

i y' g* i w a Os .l . 1. :. .1. t s; e ...u. .l L {.'_ _.._ L, y j.p.. _il: 1.i__.. g.4.-.. t. 6 - n.} __. e .1;c __.. L ; g) _ o.J..y_ f:

  • *t g

. f.'. g l llt .i w

  • t:

4, .m-i i. n. __ E. 3 g g e l i 4 J.: (.. '; 9,.. .... _ _...-4.._ l l t. { ] ~ i n. j ] 8 d i.. c. I c. 1: .__1.. e._. - j._. _. _.v... y gK_ -.a _3 .__d. - _.i 3 0 i t l l +

s j

.. m...-.: ,y. I-i.

l'

] D 3 __S x-U c' e l l .i-6

1 i

t.

  • t, j

.L i L. .4. - : u i. ......g. g

l t.

_.9. l

i..

l. l t- .t g .l j. 4 e .:t. { -1.' '1{!' .. x...-. "'.- ( i i I., '. g. e q.-..

q...q_.

_g .c i u_ ._.t.. ..... ;.:_.. ;.Lu. ... = 1 . lu , _x 1, _. i -} i q j.. ,x I. ..;. x..g.g...,..

q
.1 -

iq: f ' l. . i. l-j- 6 I I h. .i ~ i i o O O O O O O O O O 8 W f M N e-(%) @no u anou9 l'ana Cn CO N -138-

O A 1. I. .g 6J

j

.. L... _...!.,.. ^ j ...... '.. f. .}. ....l....N l, ...' L.l.. i .i. 7l. I l .I p __a pr.' l- '1'^ m l~~~N -- + i 1: -- t._...s..i. _. i t

1. _

. a... ..g_4.3... ...e.._ i' - Lee - 2 i i g. l 1 g 8.j_ gg-'g g !_ + ...[._... bn r l l ..l. ..i 1 4. w cc;.c 4..._.-.._ ;. . l....,. [ '- uw j 1 l I O e+ W< 1 { l g - n 4 %s4 l l I CD >* C3 % 3 I e .....a. ...._.9......4_ 4._ _.. _ _..... y._ t -l_ 'l. i i 5 M' 2:;3 F i

i.

- e i le 4c_e m _;._L_.1._.l...a. j -

1. 2...

-= .u....= i j O J-um ' l.i l g' l } i. we w i < ft --t------ ~ -G 4 i a u w ce cc 22 w : t i g- . l u. - g.l ; . _.. _... q.. S u y . o.4 _o g...j.. I y C, F-k r-L - .t i - o W' I, 2 l d _ 3 __v._ 3 %n. "* . - :<3 2

  • 3

.'s .]:, l ,l. - p .ft'

. i.

i.i j ^..j' Q J< '.l: - Q l _- 1 . l- ' T. : !n! ' 9, l _.i__. _j k "d : U

..t..

~1: - b-N: un Q 'C .E g 4

. l e i,.,

. i. i. i.~. x ~ en N 3 } }. e=; m O . a. - ~' tC

.g;:.

i:. __N U I E ..o. e+r..- }, n a2 ..}. .e y za ...~ ? i l W 4-J }' W '. :l ' - o' t' !' 4' f 6 T.' .h:. - 1. : '. j: i t. e:.

3..:

z >. 't' i __[ l.. - l. ' c;; A yy

m.. w O cc

... r>

t
.

. j.7 ; ./j-9.. {' - - -j. >g

}.. --

Lw E g: ..; j g; p .a .l .r ..j.. g 4 e . }. - om 3 w z< =_L; ..L.- g f i ,.W 4 O 1 ma ..l 3 ! -. p: i in -w i t t 0: ~1 ma 1 /e i / w cr. <L ' f ' '!.. I ' f, ! /- l l' f ti I t. . '. _ _. @d i l ] I {.'

  • )===

CO g t_ L_ l J 5 1 - ;; Uz 1 I, r-- 7: e. i 4. .. c.. a y ) a I 4 I ~'O y d i i 1 .1 y [ .._ d _L .....i ... 4._, ..h 1_ i c j - [.;..q.. a., ] ;- l l l

t. :

.... g a._..=. 21_i. - ..l... I s 6 u - i 4 j +::- L .i-I w t [ f [ ~ ..,._m... 4:L... ]y l

l l

1 4-I' .~ n.l i e I a g e J d O l l ! l N . 3_....... y. l e 3,. i J + 1 i I ,m, o o o O O in y M N O ( as/ 87) Mo u -139-

l CONTROLLED DISTRIBUTION FOR WARD-94000-00321 DOE /RRT Acting Branch Chief. Engineered Components Branch DOE /RRT Acting Section Chief. CRBR Program Section DOE /RRT Acting Section Chief. Components Section DOE /RRT Acting Section Chief, Reactor Analysis Branch DOE /RRT Fuel Systems Section (R. J. Neuhold) DOE /ARD Sr. Site Representative, Westinghouse AD DOE /CORO AAOD (S. Zirin) (2) 00E/COR0 Dir., Reactor Prograrns Division CRBRP/PO CRBRP Project Office, Asst. Director, Engineering ML Associate Dir., Engineering R&D ML H. Golden HEDL Director HEDL S. Additon LRM/0R Project Manager, CRBR Plant, OR, - W. J. Purcell '~ CRBRP/0R Project Director, CRSR Plant, OR, - L. W. Caffey CRBRP/0R Chief. Engineering Support Branch, OR, - R. E. McCall e 4

s ..s 4 j ATTACHMENT III DEMS PRE-TEST PREDICTIONS OF REACTOR INLET f FLOWS AND TEMPERATURES DURING THE l FFTF TRANSIENT tlATURAL CIRCULATION TEST 1 e c 7 f I 1 e* l t _. 4 o l i 1 i i l l i ..,_.. ~ - ~. _....,.,, -

ATTACHMEtiT III PRE-TEST PREDICTIONS OF REACTOR INLET FLOWS AND TEMPERATURES DURING THE FFTF TRANSIEkT NATURAL CIRCULATION TEST Pre-test predictions of reactor inlet flows and temperatures for the FFTF transient natural circulation test from initial conditions of 35% power and 75% flow were made during the first quarter of CY.0 and docu-mented as a part of the Attachment 11 report WARD-94000-00321 published in March,1980. Subsequent to the issuance of this report, there have been two changes in data input for these predictions which have an im- ~ pact on the results. These are:

1) The initial predictions assumed a power history for the test which is equivalent to 1 hour of operation at'35% of 400 Wt p wer.

The actual power history will be in excess of 5 hours of operation at 35% power and may be as long as 56 hours.

2) Pressure drop tests of the fuel assembly inlet nozzle / shield block assembly conducted at 1cw flows typical of natural circulation have shown that the a P's at low flows are less than those assumed in the analysis reported in the March predictions.

The significance of these two aspects is the following:

1) Higher decay heats associated with t.he longer times at power will cause measured temperatures within the core (in the F0TA's) to be higher than those which have been predicted for the 1 hour power history. It was thus necessary to evaluate and quantify this effect.

It also means, because of higher core temperature, that the flows will be higher because of the increase in the thermal driving heads.

2) The reactor flows will be somewhat higher for a given thennal head due to the reduced reactor a P's associated with the inlet nozzle / shield block assembly.

III-1

_ - = _- For these reasons, it was decided to supplement the analyses described in the WARD-94000-00321 with additional analyses of several cases to provide data for FORE-2M analyses so that the upper bound on fuel assembly temp-eratures could be developed. Four additional cases for the 35% power /75% flow test were analyzed. Each of these cases was calculated with the same program used to compute tne results given in the March report with the exception of decay powers and reactor a P's. The new decay powers used for the 56 hour decay heats are shown in Table 1. For those cases using a " maximum" decay power, these values were multiplied by 1.25. The difference in the correlations fce the reactor a P versus flow for the fuel assembly, non-fuel assembly and ty- ~ pass regions are shown in Figures 1 through 3. When the case is identi-fied as a maximum a P case, the uncertainties given in the March report were used. The four cases (in addition to the "best estimate" 35% power /75% flow case given in the March report) are as follows: CASE 1: Same as the "best estimate" case given in WARD-94000-00321 except a nominal decay heat based on an assumed power history equivalent to 56 hours of operation at 35% power. CASE 2: Same as CASE 1 except maximum decay heats (+ 25% uncertainty) for a 56 hour powel history. CASE 3: Same as CASE 1 except the reactor pressure drops based on the revised correlations (Figures 1 through 3). CAS E' 4: Same as CASE 1 except that high side (" design") pressure droo uncertainties (as described in WARD-94000-00321) were applied to the pressure drop ccrrelations. 'e 4 ) III-2

The new reactor vessel pressure drop correlations used for the average fuel, non-fuel, and bypass assembly are shown in Figure i through 3 respectively. These correlations were calculated from data taken from COBRA-WC analyses which, in turn, were based on the new experimental data on pressure drop through the fuel assembly inlet nozzle / shield blocks from HEDL. The rethodology used in the development of these correlations was the same as that described in the March report. ~ F.igure 4 shows a comparison of the predicted primary icop flow in GPM be-tween the base case (results reported in WARD-94000-00321) and the four additional runs. As expected, comparison between the base case and case number 1 shows that higher decay heat (56 hours of operation) resulted in higher primary loop natural circulation flow. The maximum difference in predicted flow due to the new decay heat effect alone is less than 15% for the first 300 second, transient. Comparison between the base case and case number 3 shows further increase in the predicted primary loop natural circulation flow due to the lower reactor vessel pressure loss coefficient correlations used in case 3. The maximum difference in predicted flow between these two cases is approximately 20%. Finally, the comparison also show, that case 4 utilizing the " design" pressure drops and the nom-inal decay heat for the 56 hrs. power history resulted in the most severe treesient in terms of natural circulation flow. Figure 5 shows the same comper'. ion on an expanded scale for clarity. 4 f e e P e f O III-3 ~~

TABLE 1 DECAY HEAT

  • FOR 56 HR. INITIAL OPERATION AT 35% FULL POWER TIME FUEL DECAY N-F/A DECAY' HEAT HEAT (SEC)

(If4) (MW) 0.0 6.769 .0990 1.0 6.769 .0990 3.0 6.111 .0903 6.1 -5.459 .0816 12.0 4.998 .0754 24.0 4.488 .0685 42.0 4.053 .0627 60.6 3.769 .0589 78.8 3.561 .0560 97.0 3.398 .0539 115.2 3.266 .0521

13. 3 3.157

.0506 151.5 3.064 .0493 169.7 2.984 .0483 187.9 I 2.915 .0473 i 206.0 2.843 .0463 l l l o i i

  • Neutronic heating needs to be added -

to the decay heat for total heat to i l fuel assembly acd nonfuel assembly. l 1 w,, 4 l III-4 . ~.

DYN PRES.DRGP R.V. FUEL ASSY I icoo. r i u e. x r .i p. i o.- p.. y

,p 8

x l l -c w i o

l i...
o..s s... '.... __. _....... _...

e . s.N i.. t i m u b s. e o ; I v l : j a i .,~ . i i .. s.:% & 6 g 8 ,i'<.s i .... ~.. ~.......... ~,.......... m y it soo.. 0.0I O.10 i.00 ETI4ALIZED FLeil trc. '7--F1.'O. LMY l'EES.[rr#' II/ec CORE 1. O ---O FLE1. ASSDC.Y PRES.DhP DESIGN CASE REFERENCE FLOW: 3918.82 LB/SEC Figure 1 + 8 - - - - A

^ ^ DYN PRES.DRSP NGN-FUEL ASSY I 1000. 1 8 95 i C G (I' l 3 '( e t - j.. i A i

  1. o i
5 j

,3 4 I ..g....... i o i e i I d A ... ~ ~ ~.. - j i j! I i t a i ~9 - e l.: l </ gg q7 l t o ~..,,s,. w .s..- 4 + 1 ~.. ' g i ,,~ i ~.~........, w L'

300, O.DI 0.10 1.00 1

NORMALIZED Fl.0M t.rctm 4"-- o.nw na ASSBD Y PEES.T# 1I/80 COREl. 7 *2H FVil m::;oet.Y Prf.S.DRW ESIGN CK:E REFERENCE FLOW: 507.39 LB/SEC Figure 2 +

+ DYN FRES.DRGP BYPASS I / ~ gg 8

  • 3 i

t i d. i r x .l '. 6 1' i c-d.- pg

c. -

. ~. t l (V/ T,! i .4...,..,.. q i o [g. _..;....;. _..... e i i Cd .7,.__.,i._}I_ _. i o 1 i i A M u h l I t g v I e e .~: _.o ...- -....-.:_.. h, q_[._ _ J ___ _. __p.__ _.. _.. s.;. _.. p i t w s. y. r e e, k i 1 .t I00. 0.01 0.10 t.00 NORMALIZED FLOW t LECD C -7 7 BYPASS RE1. ASSDH.Y PRES.DRFP II/00 CEREL .O-- -O BYPASS FW1. AS3CCL.Y PfM S.CRCP DESIG4 CASE REFERENCE FLOW: 432.79 LB/SEC Figure 3 + / s /.

7_. e O o O 1 ) ,1,t 6 h. l ~ g 4,- t J d.j 1 s. 9 t i I A + p O N}. in I N mm.m I ^ i ,00c00 = mmmmm i l , 2 ; memem t' covuuv 8 9 - l . t, s r_ a

  • 3
i e

a l J J I W J.J w 6 I a 1 JZwwm ~ WD . t .1 I LC U 'C C' O ? c * 'I GUU u vu o u o. e !'E;

  • 'I

+i i l i 1 O e@hh@ I 1 - 1

1.
  • 3 I

N o o m cc o i 1 S u. c. o. 1 ~.c g". l8.i. l [r l .i I; m'wmmW l i m w w o: l .I I w e. cc c: a ( I 8 m LL t 1 ; i A> e 9 s e.i i 1 3 >gpyg mm wo o j '.' ! I. g + ff) QOCo o O s cD (D N co n. o. n.. ss ra Z VJ O c) s i e,- i ) In v

  • < 6-Z,-.ct H

Ohg<v4.k N y I; i l l- ~ u v.b. @ l V) I. g .-

  • g I

2 w i,t

  • 3 wuou 3 til 0

.I 5 4 EP cuow ou CD 1 t J, ww o.- Jo t; g y< o.,, u g; G l ii EoG G O i .m.m. u m H 1.'.d, w t,g s w. 11 i gw --u 8_ ggMg U .-~u-s- ~~-m.m-O 1 mv

  • i w w.. g w 4

/ -, 2.- - g f mmumm www a / awodi E / g ,4 EMEME E-- 88888 g o to t.opo g .sl / Y9997 i O o O O d l 8 8 8 e l O O [ O No - PD'id dOOl TdVH1W 4 l + .Ak*Ec' ~,

i FFTF 35 FCT CC?ARfSZM i

900o,

.._. ~ x 0 c'. i000. h ~ ~ ~ ~ ~ n o _\\ .i [ m s 3- \\ + k- \\ \\ \\ 7 ((d e.*..U *-T. * :-T.*[ * *.~. L* * * -h *..~ *.*.* %.~. I1 #.*. *< p N d \\ ,, M kg - - _ - :h.T..*.*..* *M.-* ~ jy. \\ \\ No.,- - [d 100. ' - J-8 [ 50. 100. 150. 200. 250. 300. TIME - SECONDS / 7 LECDC 7-3S PCT DEST ESTlt! ATE.163 DEC.l: EAT.3/00 RV PRES DRCP COREL , Case)l) , Base O -- --O 35 PCT DEST ESTIlfATE.ECIR DEC.lEAT.3/00 RV PRES DROP CCREL 1 -EF-- G-35 PCT DESIGM.12SPCT - SCIN DEC.lEAT 3/03 RV PRES DRCP CCREL( I -o 35 PCT DESIGil.Scin DEC.l: EAT.DESIG?i - 3/00 RV PRES DROP COREL , aSe aSe t -H -u-3S PCT BEST ESTINATE.SCIR DEC. HEAT ll/6n RV PRES DROP COREL aSe l- + Figure 5 .,........g.g.-See.e.8m .}}