ML19269D109
| ML19269D109 | |
| Person / Time | |
|---|---|
| Site: | Big Rock Point File:Consumers Energy icon.png |
| Issue date: | 02/22/1979 |
| From: | Bixel D CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.) |
| To: | Ziemann D Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 7902270333 | |
| Download: ML19269D109 (12) | |
Text
-
CODSum8IS 1 Power J CompBDy r,: / o ; ', Iff!
General Offices: 212 West Michigan Avenue. Jackson, Michegan 49201 e Area Code 517788-0550 February 22, 1979 Director, Nuclear Reactor Regulation Att Mr Dennis L Ziemann, Chief Operating Reactors Branch Ho 2 US Nuclear Regulatory Commission Washington, DC 20555 DOCr2T 50-155 - LICENSE DPR-LIG ROCK POINT PLANT - ADDITIONAL INFORMATION CONCEERING ECCS ANALYSIS Your letter dated December ik, 1978 requested Consumers Power Company to provide additional inforntion concerning Topical Report " Big Rock Point LOCA Analysis Using the Exxon Nuclear Company Hon-Jet Pump BWR Evaluation Model - Large Break Example Problem," :CI-NF-78-25, Revision 1.
The re-quested information is provided in the attachtent.
David A Bixel (Signed)
David A Bixel Nuclear Licensing Administrator CC JGKeppler, USURC 7902270333
Responses to NRC Request for Additional Informatien on Topical Report XN-NF-78-25 Reauest:
1.
Page 5 Provide additional description on the opcration and modeling of the emergency condenser.
Discuss condenser cooling water, riser, and downcomer flow response as well as the heat transfer modelin'g used for LOCA transients.
Response
bbdeling of the emergency condenser is essentially identical to that presented and used for the development of the EN Hon-Jet Pump BWR model(l) and the ap-plication to the Oyster Creek reactor. 2) Differences relate to the system design input and the Big Rock Point system configuration. The emergency con-denser is a natural circulation device with a design capacity to remove about h% of the rated reactor thermal power.
Steam flows from the steam drum through risers to the emergency condenser tube banks, condensation occurs within the condenser tubes, and the condensate returns to the steam drum via downconers.
The condenser and downconer are initially liquid full and operation is initi-ated by opening a valve in the condensate return line.
The emergency condenser is modeled analytically with three volumes representing the riser, the tube banks, and the downcomer.
Phase separation occurs in the condenser tubes and downcomer regicns, and is modeled. Two valves are modeled, one representing the actual valve in the downcomer and the other, operating simultaneously, is a model device to avoid unnecessary time step reduction due to calculating a lov magnitude oscillatory flow during kncvn zero flow condi-tions.
The valves open linearly over eight seconds. Heat transfer is modeled with the flow dependent heat exchanger option available in RELAP4-EM.
The heat exchanger coefficient is determined from design data and the secondary side is assumed to be a constant temperature sink input at 100 F.
Calculated flows to and from the emergency condenser are shown by Figures 1.1 and 1.2.
Reauest:
2.
Page 9, Figure 2.3 Provide justification for the following modeling features:
a.
Two volumes in the lover plenum.
b.
Two parallel volumes for the steam drum risers.
c.
The use of a bubble rise model in the recirculation loop downcomer.
1
Response
a.
The lover plenum regien is actually modeled as four volumes, which were selected because of the particular geometrical configuration of the Big Rock Point reactor.
Referring to Figures 2.2 and 2.3 of XN-NF-78-25 (Rev 1), Volume 1 represents the stagnant volume between the vessel bottom head and the support plate, Volume 2 represents the volume between the sup-port plate and the bottom of the core external to the flow tubes, Volur.e 3 represents the volume internal to the flow tubes connected to all the core except for the single maximum power fuel assembly, and Volume h is the flow tube volume connected to the hot channel or maximum powered assembly.
b.
The two parallel volumes for the steam drum risers are modeled as five com-bined risers and one single riser. A single riser was input to facilitate revising the input model to con;ider a break in a steam riser.
For the example problem calculation (a recirculation line break) the risers coald have all been combined.
c.
The bubble rise model was applied according to the approved UJP-BWR ECCS evaluation model, and consistent with the Oyster Creek application. That is, the same system regions in the Oyster Creek analysis used the bubble rise model.
Furthermore, consideration of the bubble rise is required where a mixture level must be calculated for either trip signals or com-ponent uncovery.
Reauest:
3.
Pace 15, Table 2.5 Describe the method by which a reactor water level is determined during blowdown for use in actuating core spray valve opening when water level has been lowered to the 8.77-foot level g4.ven on this table.
Response
Table 2.5 in Reference 1 gives the plant set points for operation of the core spray valve which is a coincident low-vater level and low-system pressure trip.
In modeling a coincident trip, the controlling factor is the last trip signal to be reached.
For all but very small breaks in Big Rock Point, the water level set point is reached before system pressure has reached the 200 psig trip level.
This fact was verified by hand calculations based on the RELAPh-EM prediction of liquid mass in the upper plenum. Therefore, in RELAFL-EM, the pressure trip on 214.7 psia in the upper plenum is appropriately modeled to initiate core spray valve opening, and the level trip is not modeled. Proper consideration vill be given to the effect of the level trip on core sprey actua-tion time for small breaks if results indicate that the level trip vill be reached later than the pressure trip.
2
Request:
4.
Pace 16, 2nd Sentence Describe the modeling of the steam drum vater level used for actuating the diesel driven spray pump.
Resnonse:
The core spray system is modeled as a fill system which is tripped on steam drum mixture level.
The trip is set on a lov steam drum mixture level with a 20-second delay to allow all components to reach rated capacity.
No spray flow is assumed until this condition is reached (at 22.13 seconds) even though it is assumed that the spray valve begins to open on low pressure at 12.76 seconds and opens linearly over the next 15 seconds.
(The actual valve open-ing time is approximately 25 seconds; however, it has been shown that rated spray flow vill be delivered to the ring sparger with the valve 60% open. )
Thus, spray flow is zero until 22.13 seconds, increases rapidly to a value given by the product of the fill system input and the valve area, when the pump rated capacity is reached and then increases approximately linearly until the valve is fully open (at 27.76 seconds). This behavior is shown by Figure 3.13 of XH-NF-78-25 (Rev 1).
Spray flov is calculated to reach the rated flow value at 22.8h seconds but credit is not taken for achieving rated spray until the spray valve is 60% open (27.76 seconds).
Since RELAP4-EM treats all volumes as upright circular cylinders (linear varia-tion of volume with height), the Big Rcck Point steam drum which is a horizontal cylinder, required some model approximations.
The model was developed to assure the correct initial liquid mass in the steam drum, and the liquid level at time of trip was established to give the correct liquid mass remaining in the steam drum at the
_me of trip.
Thus, the input initial mixture level and trip mix-ture level differ frcm the actual plant set peints but the liquid mass in the steam drum is properly modeled in both instances. The differences in the gravi-tational head introduced by making this assumption are small relative to the total recirculation loop pressure drop and can be neglected.
Requent:
5 Pace JB, Ficure 3.10 Discuss the emergency condenser response shown on this figure in the initial stage of blevdown, as well as the water loss and recovery shown for the terminal stage of blevdown.
Response
Figure 3.10 from Ref 1 shows that the emergency condenser tubes remain full of water for the first 15-16 seconds. This indicates that the heat renoval capacity of the emergency condenser, combined with the subcooled (100 F) water initially in the tubes, is sufficient to condense all the steam flowing to the condenser, 3
and the volume remains full and subcooled. At about 15 seconds, the system has depressurized and the enthalpy of the liquid in the emergency condenser has increased sufficiently such that the fluid reaches saturation and begins to vaporize.
The liquid level in the emergency condenser volume is established according to the approved bubble rise model during the saturated decompression.
As the system equilibrates with the containment, vaporization subsides and emer-gency condenser heat transfer condenses steam flowing in at a faster rate than the liquid flow from the condenser volume; hence, the calculated level increases.
Request:
6.
Page 33, Items 1, 2, and 3 Provide further discussion on the change made to the way choked flow is implic-itly related to a pressure change in a given time step.
Discuss the change in treatment of choked flow at junctions with the new procedure relative to the original process.
Discuss the procedure used for making flow estimates over a given time step.
Describe the causes of estimates that are " drastically different" than the existing flow. Provide a calculated comparison between the original and the changed method for a representative junction under choked flow conditions.
Rcauest:
7 Pace 3h, Item h Discuss the manner by which energy balances are considered when setting a volume of saturation ecnditions as volume pressure enters the 110 psi deadband about the saturation value. Provide a calculated comparison between ENC 26A and ENC 283 for a representative volume undergoing a given pressure transient ap-proach to saturation.
Recuest:
8.
Page 34, Item 6 Describe the method of interpolat5cn employed between heat transfer correlations to minimize discontinuities between correlations. Provide a comparison calcula-tion between ENC models for a given slab transient.
Responses to Recuests 6, 7, and 8 :
Eequests 6, 7, and 8 relate to the changes made in RELAFh to reduce computational time.
Reference 3, ceparately forwarded to the URC staff, gives a detailed de-scription of the RELAPk updates and the impact of these updates on blevdown results.
The reference concluded that the updates separately and accumulatively result in changes in the calculated cladding te=peratu" of less than 20 F.
k
Specifically, Request 6 is addressed in Paragraph 2.1 of Reference 3 and Request 8 is answered in Paragraph 2.3 of the reference.
The subject of Request 7 (inclusion of a saturation pressure deadband as the saturaticn value is approached) was considered as an update for some time but was not included in the final PILAPh version (RELAPh/E!C28B). This point was listed erroneously in the Big Rock Point LOCA Analysis report. Thus, no change in RELAPh was made re:ated to the subject considered in Request 7 Request:
9 Pace 35, Item 9 Identify the criterion used for selecting the trip level at which switch-over from phase separation to homogeneous ecolant modeling is used in a control volume as the liquid level approaches the bottom of the volume.
Describe the influence on flows in connectin; piping above the switch-over level, and de-scribe any comparisons to experimental data for fluid behavior in a volume as liquid levels approach the bottom.
Response
The intended application of this trip is to switch from a phase separation model to a homogeneous zodel when the volume is calculated to become essentially void of liquid; hence a very small trip level (typically 0.01 feet) is chosen.
This switch is provided to prevent the code from calculating model associated flow oscillations at.he connection at the bottom of the volume as small liquid levels are calculated to appear and disappear during depressurization.
The trip capability is consistent with the trip capability of RELAPh.
The desired change in fluid models is a user option for any or all volumes in which a separation velocity greater than zero is initially specified.
This capability is essentially that proposed by and used in Reference 5 Rei-erence 5 provides a comparison between Two-Loop Test Apparatus (TLTA) experi-mental data and RELAP h, with and without this trip capability, and shows that this option is necessary for modeling volumes such as the Big Rock Point steam drum.
Reauest:
10.
Page 35, Last Sentence Describe the comparison tests performed between the two Exxon ECCS model ver-sions, and provide a PCT transient comparison between the previous Big Rock fuel relcad analysis using HUXY and GE blowdown calculations and the present calculations using the ETC28B model with ccmparable linear pcVer densities.
Restonse:
Extensive evaluations of results of the approved code version RELAPh-EM/
ETC26A and its update RELAPh-E4/EIC28B vere made on the cede modifications to accure that their only affect was en running time and not on thermal hydraulic 5
results.
The results of these evaluations confirmed that the changes made did not significantly affect the RELAP prediction. Tests performed with the two model versions were already addressed in the attachment to ENC letter (3) from G F Owsley to D F Ross dated October 30,1978 (Description of RELAPh-EM/
ENC 28B) and reference to this letter should be made for details.
Results of previous Big Rock Point fuel analysis using the ENC HUYY code and the GE blowdown calculations were reported in the reload G-3 uranium fuel licensing submittal.(h) HUXY calculations with GE blowdown data resulted in 2
a break spectrum which displayed the limiting break size as 0.25 ft. A MAPLHGR of 6.554 kW/ft has been found limiting for beginning of life for this ECCS analysis.
The related HUXY transient for this case is shown on Figure 10.1.
The present calculations using the RELAP4-EM/ ENC 28B model for blevdown provided results only for the example problem which is the double-ended guillotine break with a break size of 3.52 ft2 and may not be the limiting break. The combined axial-times-radial power peaking for this case corresponds to a MAPLHGR of 8.89 kW/ft. The HUXY transient for this case is shown on Figure 10.2.
Request:
11.
Ficure 4.1 Describe the blowdown and hot channel configuration models used for the compari-son shown on this figure.
Resnonse:
The blowdown and hot channel configurations on which the transients of Figure k.1 are based are reproduced in Figure 2.12 " Blowdown and Hot Channel Core Configurations" of Reference 3.
This figure identifies the heat transfer nodes being compared.
The hot channel heat transfer node is at slightly lover power than its blevdown counterpart because it no longer includes the maximum power fuel rod of the assembly. Thus, the expected result of the blowdown / hot channel comparison is similar time behavior with the calculated hot channel results slightly below the blowdown results as shown in Figure 4.1.
Request:
12.
General Provide experimental verification of Exxon's Non-Jet Pump EWR Evaluation Model through modeling of the Two-Loop Test Apparatus (TLTA).
Tests conducted re-cently to provide experimental data on blevdown dynamics without ECCS (Test 6007), and with ECCS (Test 6h06), or comparable tests, should be used for such verification.
Response
As requested, Exxon Nuclear Company (ENC) vill use appropriate test results from TLTA experiments to verify its Non-Jet Pump EWR Evaluation Model.
The 6
schedule for performing this verification was discussed at a meeting between E:IC and the NIiC staff en January 8,1979 At this meeting it was agreed that this verification would be deferred until after E:IC cc=pletes pretest analysis of the Loss of Flow Test (LOPI) L2-3 test. The L2-3 test analysis will not be completed before about April 1, 1979 ENC will, at a later date, separately address the detailed schedule for performing the requested verification.
References 1.
The Exxon Nuclear Company WREM-Based JNP-EWR ECCS Evaluation Model and Application to the Oyster Creek Plant, XF-75-55, Revision 2, and Supplements 1 and 2, April 1977 2.
Oyster Creek IDCA Analyses using the E:IC NJP-37R ECCS Evaluation Model, XH-NF-77-55, Revision 1.
3.
ENC letter frcs G F Ovsley (ENC) to Denwood F Ross (NRC), dated October 30, 1978.
h.
ECCS Analysis for Exxon Nuclear Company G-3 All Uranium No Cobalt Fuel for Big Rock Point, XN-NF-76-55, Revisien 1, February 1977.
5 Hendrix, C E, and Schults, R R, TLTA Sensitivity Studies and Test h903 Run 16 Eata Comparison, I:TEL Report No RE-S-76-192, November 1976.
7
RELAP4/003 06/22/78 RUN ON 06/23/78 BIG ROCK POIt4T BLOWOOWil CALC. GIL BREAK IN PUMP DISCHARGE LINE, 29JUN78 i
e i
i i
a i
i i
4 "1
Ea is Mo 2
5
'h
. 1 O
LL O
N.
i i
i i
e i
ep 6
8 12 1C to ti 28 32 3C 40 TIME AFTER BREAK ( SEC 1 4
m FIGURE 1.1 f1 ASS FLOW If1TO Tile Ef1ERGErlCY C0flDEllSER
(
l
(
RELAP4/003 06/22/78 rut 1 Ot1 06/29/78 BIG ROCK POIf4T BLOWDOWN CALC. GIL BREAK Ir4 PUMP DISCHARGE lit 4E, 29JUr478 o
a i
i i
i i
i i
i n
"I ma
=
=
a 9
8
{E, a
~
5 l) g
/
g 02 O
LL o
=
M i
i i
i fj Yo e
a 12 sc to es za 32 sc sa TIME AFTEi! DREAK ( SEC )
g FIGURE 1.2 11 ASS FLOW OUT OF Ef1ERGEllCY C0flDErlSER
R 0
E 9 1 T i
0 1 2 S 6
I NNN I I A E
PPC M
M I
T T
3 M 0
G/
f 0
S 7
U
- DK S
DWA R
4 O E
RME V
O R
S L
B E
E0 f
0 R
0 U
R L
6 T
L A
C A
R E
N M
P E
S M
6 E
0 T
7 T I
0 T/F 5
R E
P0 O
)
T 1 S C
S 0
E K
0 S
I l
2
(
t C2 5 A
2 OO2 0 E C
=
RU 0
M 3
4 I
O T
=
A GL 1 L J
3 l
=
O R
i 2 9
5 E 5 9
3 2 1
0
=
0 1
0 I
0 1
2 r : $
1 1
1 S
8 T 2 3
0 2 3 9
5 9
3 2 3
e 1
1 1
1 1
1 1
8 3
1 3
6 F
8 T
6 3 E
S i
1 I
1 1
1 1
S 4
R O
t T
3 0 5 T
9 0
U 5
i i I
1 2
1 t
1 1
5 0
G I
0 I
t 5
8 0
1 0
8 5 l
6 i
1 1
2 2
2 1
1 i
6 2
F 0
t T
3 0 3 T
9 0
5 1
i 1
1 2
1 1
1 1
5 o6 T
8 T
6 3
4 S
1 I
1 I
1 1
4 2
3 9
5 1
3 2
3 e
3 1
1 1
1 1
1 e
I 0
0 0
1 0
1 2
r s S
1 1
1 S
s T
2 i
2 3
1 5
6 5
4 3
2 t
0 0
0 0
0 0
0 0
0 5
0 5
2 1
1
,S v uEE
' @d
9 i 2 3 4 5 6 5 9 3 2 1 3IG ROCK PT.
11X11 ENC-G3 2 r a S to 12 to S r 2 RLL UO2 10/76 0.
MWD /MTM 3
t2 n ig tS iv 13 12 PUMP DIS GIL BR VOL 11 t 8 tJ 16 1T 13 1T 14 13 3 1 5 10 14 IT IS 20 19 tr 14 10 5 e
& 11 15 18 20 21 20 1s 15 11 s 5 10 11 tr 13 20 13 IT 14 10 5 t 3 13 16 tr 18 1T 16 13 5 9 3 8 12 13 14 15 14 13 12 3 3 1
2 r 8 9 10 11 10 S 8 r 2 1 2 3 4 5 6 5 4 3 2 1 2000 1968 PIN 12 (HOT ROD) 4 PIN IS (INERT ROD) b O
CANISTER 3 1500 E
@ 1000 d
/
500 (
.=
=
i i
i l
i l
i 100 200 300 400 500 600 700 800 TINE (SEC)
FIGURE 10.2 R0D AllD CANISTER TE!!PERATURES VERSUS TIl1E
_. _,