ML18143B360

Vapor Circulation in Upper Vessel of Surry PWR for Tmlb Accident Sequences, Draft Ltr Rept

ML18143B360
Person / Time
Site:	Surry
Issue date:	06/28/1984
From:	Henninger R LOS ALAMOS NATIONAL LABORATORY
To:	Han J NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
References
RTR-NUREG-0956, RTR-NUREG-956 Q-7-84-318, NUDOCS 8508090749
Download: ML18143B360 (13)

Text

( c-( Los Alamos National Laboratory Los Alamos.New Mexico 87545 Energy Division Dr. James T. Han Reactor Safety Research Branch Mail Stop 1130SS US Nuclear RegulatoYy Commission Washington, DC 20555

Dear Jim,

DATE June 28, 1984 IN REPLY REFER TO: Q-7-84-318 MAIL STOP: K556 TELEPHONE.

(505) 667-2023 or FTS 843-2023 Enclosed is a draft letter report that presents the results of the TRAC-PFl upper-vessel-circulation calculation for the Surry TMLB' accident sequence.

The AB sequence calculation is completed and a discussion of the results will be included in the final version of the report. Please call me if you have any questions or require futher information.

Sincerely yours, Ru~ Henninger Safety .Analysis RJH:bn Enc. as cited Distribution:

CRM-4 (2), MS-A150 c. Kelber, NRC R. Denning, BCL File (RJH) 8508090749 840628

• PDR ADOCK 05000280 P PDR An Equal Opportunity Employer/Operated by University of California

( :. -.. lbRAFT VAPOR CIRCULATION IN THE UPPER VESSEL OF THE SURRY PWR FOR THE TMLB' ACCIDENT SEQUENCE R. J. Henninger I. INTRODUCTION The above-core structures can provide a significant heat sink during a degraded core accident.

In order to determine the extent to which the structures affect an accident, TRAC-PF1 1 calculations for the Surry Pressurized Water Reactor were performed.

The sequence chosen was a total loss of feedwater with failure of the emergency core cooling (ECC) system (TMLB'). Core outflow conditions, that consisted of time-dependent steam and hydrogen mass flows and vapor temperatures, were used as boundary conditions for the TRAC-PFl ( *. calculations.

These core outflow. conditions were calculated by means of the MARCH code and were provided to us by Battelle Columbus Laboratories.

2 II. MODEL DESCRIPTION A. TRAC Model The TRAC model for the upper part of the Surry vessel is shown in Fig. 1. The model consists of 7 axial levels, 3 radial regions and 2 azimuthal sectors. The inner two radial nodes model the region inside of the core barrel. The first five axial levels correspond to the region betveen the core support plate and the upper support plate. The upper tvo axial levels model the upper head. In each of the four nodes inside of the core barrel there is a pipe that provides a connection between the bottom of the upper plenum and the upper head. These four pipes represent the 53 control rod guide tubes (CRGT) in the Surry vessel. Flow through the CRGTs is restricted by a small total flow area of 0.12,-4 m 2 near their tops. Small-area flow paths between the downcomer and upper head and the downcomer and the inner radial regions were aJ.so modeled* t The hot leg with the pressurizer is connected to one of the t~o azimuthal sectors.

(_ In ~~-PFl I only one heat slab is allowed per node* The vessel noding was therefore chosen so that *thin" structures within the core barrel could be sep~rated from "thick" structures such as the upper support struceu[e and vessel walrs. Thin* structures were typically 0.006 to O .008 m thick. The heat slabs '::. in (he fifth axial region, which includes the upper support are 0.022 m thick and those in the the outer radial node which models the vessel range from 0.15 to 0.30 m thick, depending upon the presence or absence of nozzles and flanges. The aass of the CRGTs was divided equally between the vessel component and the pipes used to represent the guide tubes. All of the heat-slab masses and surface areas were obtained from Westinghouse via Battelle Columbus Laboratories.3 B. Boundary Conditions One of the boundary conditions for these calculations is the pressure in the hot leg. The other boundary condition, as indicated in Fig. 1, is the flow at the core outlet. The conditions for the TMLB' sequence are given in Figs. 2-4. As shown in Fig. 2, the mass flows decrease from the time of core uncovery at 5730 s until approximately 8760 s. At 8760 s, the core slumps into water below the core region producing an increase in core outflow. The vapor temperature shown in Fig. 3 increases with the center (higher-power) node leading the outer node. The calculation was stopped when the temperature returned to the saturation temperature and the flow from the core region was steam. Figure 4 gives the total pressure and the hydrogen partial pressure at the core outlet central and outer radial regions. The total pressure remains near the relief valve set point (16.3 MPa) throughout the transient, and the fraction of flow that is hydrogen increases as the accident proceeds.

C. Initial Conditions The TRAC-PFl calculation of the TMLB' sequence was begun when the core was uncovered and the vapor temperature at the core outlet began to increase above the saturation temperature.

The initial conditions for the TMLB' sequence were that the vessel and hot leg were at the saturation temperature corresponding to 16.3?a (622 K). This seems a reasonable assumption especially for the thin structures.

The vessel temperature is not very important in these~alculations because, as we shall see, there is not much flow to either the uppei head or the downcomer.

( . c* e III. RESULTS FOR THE TMLB SEQUENCE The TMLB' calculation was run at the initial conditions fo 1 400 s {from 536~ to 5760 s). As the temperature of the vapor flowing fr:.om the core inc~ased alter 5760 s, a flow pattern similar to that depicted in Fig. 5 is ~-devtloped.

• This pattern consists of two major convection cells. Most of the upflow is in the central radial cell that is on the side of the vessel next to the hot leg. The returning downflow is in the outer radial cells. The axial mass flow at the top of cell 2 in the vessel is given in Fig. 6. This pattern persists until the rapid flow increase that occurs when the core slumps at 8760 s. The flows within the upper plenum can be compared to the inflow from the core and the outflow through the hot leg which are given in Fig. 7. Comparison of Figs. 6 and 7 shows that the flow within the vessel remains large compared to the inflow and outflow. The vapor from the core is therefore well mixed with vapor in the upper plenum before it exits through the hot leg. This is seen in Fig. 8 1 which gives the core outlet vapor temperature and the vapor temperature in the hot leg. The decrease in temperature is a result of the mixing of the vapor from the core with the vapor already in the vessel. Until the flow increase at 8760 s I the temperature in the hot leg remains relatively low. The energy flows in the vessel are given in Fig. 9. This figure indicates that the major energy removal mechanism up to 8760 sis flow out _through the hot leg. The heat slabs participate very little. The flow i's therefore driven by density differences between the vapor in the vessel and the vapor leaving the core region. The vapor entering the upper plenum from the core is less dense for two reasons. The first, of course, is that its temperature is higher. The second reason is that the fraction of the flow that is hydrogen is increasing.

The importance of these mechanisms in driving flows will be examined by J. Dearing with his two-dimensional MELPROG flow module. Flow through the CRGTs was 5 kg/s or less and not important for energy transport.

The flow directions with the CRGTs, as is indicated on Fig. 5, were similar to the flow pattern in the upper plen~m, namely up the center radial pipe on the hot-leg side and down the other pipes. Flows in the CRGTs on the hot-leg side are given in Figs. 1~ and 11. The increase in mass flow associated with core slumping r~sults in an altered flow pattern. The flow pattern at 8800 s is shown in Fig. 12. The convection cells persist but some of the vapor flows more directly to the hot leg. As the flow increases, so does the importance of the heat slabs. Figure 6

-!'. (' . \::. shows that the energy flow to the heat slabs becomes significant after 8760 s. The temperature of the heat slabs in Figs. 13 and 14 increases 13ignificantly t fol~owing i~creased core outflow * * ~-IV.; CONCLUSIONS AND RECOMMENDATIONS

.. The important conclusions that result from these calculations are: 1. For the flows provided by Battelle Columbus, the vessel structures were not an important heat sink from the time of core uncovery until the time of core slumping;

2. Flow driven by differences in density between the vapor exiting the core and vapor present in the vessel resulted in mixing and lower temperature vapor exiting the vessel; 3. Flow areas of the connections between the upper plenum, upper head, _ and dolofflcomer are too small to be of any importance to the energy flows; and 4. The vessel structures become more important as heat sinks when the c~-core outlet flow increases following slumping.

I believe that a coupled multi-dimensional analysis of this accident that included the core region would produce higher flow from the core region thereby increasing the importance of above-core structures.

I therefore recommend that this calculation be re-run when such a capability exists. This accident sequence will provide an excellent test for the multi-dimensional version of TRAC/MELPROG.

REFERENCES

1. Safety Code Development Group, "TRAC-PFl, An Advanced Best-Estimate Computer Program for Pressurized Water Reactor Analysis," Los Alamos National Laboratory report LA-9944-MS (NUREG/CR-3567), February 1984. 2. Roger o. Wooton, Battelle Columbus Laboratories, private communication March 27, 1984. 3. Peter Cybulskis, Battelle Columbus Laboratories, private iommunication October 1983. t

( . C ( LEVEL 7 6 s J -*. SURRY UPPER PLENUM AND HE.AD ! I i j i .. --CONT -ROL ROD GUIDE TUBE COOLING NOZZLE ----! 4 -U) ' Cl .:,t; -ii: 0 i:::: Ul Ul C 15 10 5 0 3 -* 2 -l t : ... ! I i---f--t t CORE OVTFLOW PROVIDED AS llOUNDART CONDITION Fig. l. t HOT LEG *-TRAC noding diagram for Surry upper vessel. ----------

CENTER CELL -~-OUTER RADIAL CELL .' ' I ' ' I -5..a.,~.....-

........ ~~.....-~~

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

...... ~~""' 5000 5500 6000 6500 7000 7500 BOOO 6500 9000 Tl ME (s) Fig. 2. Outlet mass flow for the two radial regions in the core. The flows are azimuthally symmetric.

( -. . 't: . " (*: e e 2soo ..... --------------------.------.------....------,-----"""'T------, -~-....... Cl,. . .. ::, -0 .. ca, Q. E ca, .. 0 Q. 0 > 2400 2200 2000 1!00 1600 1400 1200 1000 BOO 600 ----------

CENTER CELL ~~-OUTER RADIAL CELL .......... . . . : . . . . , 40C.~.--~~...-~~~~~~

..... .... ~--~ ......

......... .........

5000 55DO 600.J 6500 7000 7500 BC:JD B500 9:lOD Tl ME (s) Fig. 3. Outlet vapor temperatures for the two radial regions in the core. 1eooooa~-r-------,.------.-------,-------,.------..------,-------,.--------, 16000000 14000000 12000000 -~10000000

....... ca, 5 aoooo;;o UI UI 6000000 c.. 4000000 2000000 . I I I : : . . . I ' I . ----TOTAi. P~ESSURE ; I I ----------

HYl)l10Gtf.l .PARl1"1.

Pf-SS1[ OF CENTER R(Gl()f.j HYrlROGEtJ PAF.11"1.

PF,'r:.SyRE Of C.UTER Ri.D1AL REGION . ' ' . . ' I I ' ' . ' . ' , ' . ; . : , : , : ' _. . _. ,' .... 1' ... I ** * ,~.--** -------------------*r

"* !".** \ L n '* " '* *: :: ,: " I: .. 0: :: :: '* *: '* *: I: ,: '* '* '* '* *: :~ '* ,, : I -2000000~.----~

...... ~--.... ~~--...... ~--...... ------...... ....... ....................

5000 5500 6000 6500 7000 7:,0a BOvO 8500 9000 Tl ME (s) Fig. 4. Total pressure and partial pressures of hydrogen at the c-0re outlet. Pressure drops through the system are small, so the total pressure throughout the primary remain near this pressure.

.. ( LEVEL 7 6 4 3 -*-* ... SUllf UPPER PLENUM AMDmAD ! i ; ! T ..... .... * --...--++--i---,H---+-'"1*

• ---I ' H --* -*t-t--t----i---: -t--1i-----

.... W i 1-----... HOT LEG I _ _._._.,~4--~-.-~-M--+-*-:..*.l.J..-4

_____ __. 2 r--,--**--,-**:...--

.... ;---**-1--.*. ;-----1 * " I -... 1' 1' t CORE OUTFLOW PROVIDED AS BOUNDARY CONDITION Fig. 5. t Flow pattern in the upper plenum for TMLB' sequence from core uncovery to core slumping.

This flow is driven mainly by density differences between vapor exiting the core and vapor already present in the vessel. The vessel heat slabs are of little importance because of the limited core outflow in this time regime. Flow in the CRGTs was limited and of little importance.

-;. -100 80 60 -IQ ' 17'* 40 .:,t. ....., 3: 0 20 i:: U) Ul 0 0 :::E ... 0 -20 0. 0 > 60 ' ..... -8'.) ' 5.'.100 e r**. I ".,,* . : .... .... _ ... ...... -----"---f ~-***** .. I --. _/'-, f 55CJO 65.JO ............

Ulltl 2 1/ERllCAl r,ow u**fll R,\()l&L R[C./c,14, ('IPHiC .. 11{ Hvl L[C. IM<R 9'AOIAL ftfGIC,", H~..,1 l(G S10C CU,[J,r RA.DIAL A(C,1011.i.

c,.+{'5.1T[

+.-1T L.lC DlllEA RADIAL A[~1,)I;, HC,1 1u; SID( ***** ... *-. ...................

' 7JC,j E5C.J Tl ME (s) Fig. 6. i. SJ)O Vapor mass flows for the four segments within the core barrel at the top of axial level 2. This figure shows that the vapor flows up the center cells and down the outside cells as depicted in Fig. 5. -U) ' 0, .:,t. ...... !Ir: 0 i:: Ul U) C :::E 4'0 35 30 25 20 15 10 5 0 ----------

IN FROM CORE ~~-OUT HOT LEG . . ' ' ' ' ' *********-

..... __ .. -....... --------------s-'--~--.~~--.-~~-,--~~~~~~--,~~"""'":""'~---1 5000 5500 6000 6500 7000 7500 8000 8500 9000 TIME (s) Fig. 7. Mass flows entering the upper plenmn from the core and exiting through the hot leg for the TMLB' transient.

c* . . i'" '=: . " ( 2600 2400 2200 -2000 :lie. -ca,*. 1BOO ... :, .., 0 1600 ... QI c.. E 1400 V I-.. 1200 0 c.. 0 1000 > BOO 600 400 5000 5500 ---*-*--*-

CORE OUTLET ---HOT LEG ........ -** .. ... -** .. ... -***** _,,.-*** ,, .. -*" 6000 6500 7000 .7500 TIME (s) Fig. 8. , . .. .. .. I . . . . . . : I , : . : _ ....................... . _ ... .1 80:)0 6500 9:)00 Core exit and hot-leg vapor temperatures for the TMLB' transient.

The vapor temperature in the hot leg does not . increase significantly until the core slumps at 8760 s. 'b -* -;= ...... it 0 r;: >, 0, .. QI C w 35 30 25 20 15 10 5 0 ---TO HEAT SLABS *---------

FROM CORE ******************

• OUT HOT LEG --------... ',.;:,._ I : * * * .. .. " .. " .. 1: .. ** t: I" :* I!: 1:1 1:1 1:1 *:: " :, r .. , .. _____ *_ .... ~_ .... _ *.* _._ .... _**_-.. _ .. _ *** _. *-'"...:"~:.::.~~:.:*~.:.:-

-:.::**=",, """--------~**::--,)\ -s~-----------------------....-----

...... ----...... ----....... ------1 5000 5500 6000 6500 7000 7500 BOOO 8500 9000 TIME (s) Fig. 9. Energy flows from the core, out the hot leg and to the vessel heat slabs for the TMLB' transient.

The heat slabs are not important until the core slumps and the mass flow from the core region increases.

• 0.5 CRG1 flow Outer Rodia! Node 0 -0.5 -1 II) -l.5 :::::E ,-.... UI ' 0-, ...._, 0 UI U) 0 :::::E 2.5 * -3~.---,-,----,----.,....--~---,-,--.....,---,-----!

5000 55~0 6000 6~GJ 70JJ 7~00 BuJO 85~J '9~~0 Tl ME (s) Fig. 10. Flow in the CRGT in the center on the hot-leg side. 6---~-----

....... ---.----------.----

5 CRGT Fl ow Inner Rodi ol Node 4 3 2 . .. .. 0 ,,--,--r----.--------,.,....--

..... ----.---,-----!

5GOO 5500 600::l f-:JO 70DO 7$JO 6C*:l0 E,:,JO TIME (s) Fig. 11. i f Flow in the CRGT in the outer radial node on the hot-leg side.

. . t. . Ir. ( l .. e LIVEL 7 6 5 4 3 SURJl1' UPPER PUNUM AND HEAD ! ! --4-----4r---,...--~-Tr--t---rT-*-P--

i .-i ->--* *-~_:_ -i--+-i....-..+--+--4-++..

I ,__ _..._ ______ 1-1--*---!-+

.. -~--+-I ,, ! , 8HOT LEG -,-1*--:-...... **i****-*--!-:---.:*.cJ.

..... -....-----

2 .__ ,_" .. --.----_-.:_--_'---1--***===::-+--~

.. 1-* --:l--* -' I , I .,,,,., t t CORE OUTFLOW PROVIDED AS BOUNDARY CONDITION Fig. 12. t Flow pattern in the vessel at 8800 s. Some of the vapor from the core exits directly to the hot leg. --I) ... :, -0 ... I) Q. E " r " ... :, -u :, ... -II) 1300-r---....---....---

.........

---.------

...... ----..---1200 noo 1000 800 100 700 BOO lCVEL 1 ---INl*R RADIAL tl[CION. OPPOSITE HOT l[G ****------

INN£11 RADIAL REGION, HOT UC SID£ ---OUTER RADIAL RCCION, OPPOSITE HOT LEG ....................

OUTER llADIAL REGION. HOT UG SID£ --~= ...... --------;*" ,, 90,c,y----,----,----,---------.,.---..,...---4 5000 5500 &ooo asoo 1000 1soo aooo 1soo sooo TIME (s) Fig. 13. Surface temperatures of the heat slabs in level l for the two radial regions within the core barrel.

(. ( \ . -. t. . .. ¥ .. -Cl :::, -0 ., C. E II I-Cl :::, -u :, -en 1150 800 750 700 650 e L[V[L S ---INl*R llADIAI. lttGION, DPPOSIT[ HOT U:G *---------

IMCR OJ)IAI. RCGHJH, HOl U:G SIil( ---OUlCII IA!>IAI. R[C\ON, OPPOS11t HOT UG *** *-*************

OUTEli IADIAI. 11£GION, HOT U:G SIDC *' &IDit----r----r----r---,----,----,-----,--.i 5000 5500 GODO 6500 7000 7500 8000 8500 9000 TIME (s) Fig. 14. l Surf ace temperatures of the heat slabs in level 5 for the t1o10 radial regions within the core barrel.