ML20043G174
| ML20043G174 | |
| Person / Time | |
|---|---|
| Site: | Mcguire, Catawba, McGuire  |
| Issue date: | 06/07/1990 |
| From: | Tucker H DUKE POWER CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| TAC-75138, TAC-75139, TAC-75140, TAC-75141, NUDOCS 9006190142 | |
| Download: ML20043G174 (21) | |
Text
,,
- J o
q
'o r
i Duke her Cwviony Hu B Tuan l
PO ika33195 nce Prrsident Charlotte, N C 2S242 Nuclear Prvauctiora (704)373 4.U1 l
1 4
DUKE POWER V
)
June 7, 1990 i
1 U. S. Nuclear Regulatory. Commission ATTN: Document Control Desk
' Washington, D.C.
20555 3
Attention: Document Control Desk
.i
Subject:
McGuire Nuclear Station Docket Numbers 50-369 and -370 Catawba Nuclear Station Docket Numbers 50-413 and -414 Response to Request.for Additional Information Regarding BAW-10174 (TACS 75138/75139/75140/75141) l By letter dated March 27, 1990, the NRC staff requested information on Topical Report BAW-10174, " Mark-BW Reload LOCA Analysis for Catawba and McGuire." Attached are responses to 10 of the 29 Round 1 questions. The remainder will be forwarded when they are developed.
I If there are any questions, call Scott Gewehr at (704) 373-7581.
T Very truly yours, t
gfb.
b)T llal B. Tucker SAG /222/lcs 9006190142 900607 t
PDR.ALOCK 05000369 0
P.
,,b \\
i
a
,s U. S. Nuc1:Or R:geltt:ry Commicsitn Jura 7 1990 Page 2 xc:
Mr. S. D. Ebneter r
Regional Administrator U. S. Nuclear Regulatory Commission Region II 101 Marietta Street, NW, Suite 2900 Atlanta, Georgia 30323 Mr. Darl S. Hood, Project Manager Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, D.C..
20555 l-Dr. Kahtan Jabbour, Projects Manager Office of Nuclear Reactor Regulation U. S Nuclear Regulatory Commission Washington, D.C.
20555 Mr. L. L. Losh 3315 Old Forest Road P.O. Box 10935 Lynchburg, Virginia 24506-0935 Mr. W. T. Orders NRC Resident Inspector Catawba Nuclear Station Mr. P. K.'VanDoorn NRC Resident Inspector McGuire Nuclear Station l-t i
i u
l
~
t REQUEST FOR ADDITIONAL INFORMATION BAW-10174, MARK-BW RELOAD IDCA ANALYSIS FOR CATAWBA AND MCGUIRE UNITS 1
e
\\
s P
i
e, 1.
Justify the values of fuel and cladding conductivities and gap heat transfer coefficient used in REFLOD3B.
Responses As discussed in the response to a question on the REFIDD3B code, the thermal properties of the fuel rod are selected to maximize core heat transfer, thereby resulting in lower core flooding rates.
The conductivities of zirconium and helium (fuel-to-clad gap) increase with increasing temperature.
For use in REF14D3B these properties are based on an upper bound temperature of 1600 F.
The average cladding and gap temperatures are considerably below 1600 F throughout the reflooding transient.
Additionally, the gap coefficient is based on cold gap dimensions to maximize conductance.
The resultant conductivity for zirconium is 13.698 BTU /f t-hr-F and the gap conductance, based on the cold gap thickness of 3.25 2
mils, is 827 BTU /f t -hr-F.
The thermal conductivity of UO, decreases with increasing temperature.
Instead of using an upper bound temperature, a
~
fuel average temperature of 800 F is used to select the fuel conductivity.
The fuel average temperature during most of the reflood transient remains above 800 F, falling below 800 F only when a majority of the core is quenched.
The resultant fuel conductivity is 2.682 BTU /ft-hr-F.
To evaluate the effect of fuel conductivity on core flooding rate, a REFLOD3B calculation was made with a fuel conductivity of 3.6792 BTU /ft-hr-F (400 F).
The impact on core flooding rate was less than 1%.
Because there is no strong sensitivity to basing the fuel conductivity on a lower temperature, the use of a temperature which is below (conservative direction) the time averaged value is sufficient for REFI4D3B calculations.
w
2.
What were the values of Doppler and moderator void and temperature coefficients used in the analysee?
Demonstrate that the values used in the analysis for moderator void and temperature coefficients were conservative values.
Clarify how the reactivity versus fuel temperature and moderator density and temperature curves were calculated and used in the analysis.
Response
The Doppler temperature coefficient is -2.2 pcm/F.
The moderator density reactivity curym is shown in Table 2-1.
The moderator reactivity effect is dominated by the density variation and the moderator temperature effect on reactivity is not used.
For reactor kinetics calculations, REIAP5 calculates average fuel temperatures and fluid conditions for each node modelled within the active core and uses those conditions to obtain reactivities for each node.
The nodal reactivities are then combined by flux-squared weighting and volume weighting to obtain a single value for use in the point kinetics equation.
The reactivity coefficients, Doppler and moderator, used in the LOCA analysis have been selected as a bounding set for the burnup at which the impact of reactivity on the LOCA calculations produces the highest peak cladding temperature.
Since the reactivity change due to void formation in the core dominates reactivity change due to changes in fuel temperature
" -0.20 Ak/k at a relative moderator (for example A pmod o.n density of 0.2 while Apw % = +0.013 Ak/k for a drop of 600 F in fuel tempe:'ature), the selected burnup is beginning-of-life (BOC), which gives the least negative moderator density reactivities.
The moderator density reactivity is based on the most pcsitive moderator coefficient allowed by the technical specifications.
This occurs at BOC conditions. Technical specifications limit the moderator temperature coefficient to O pcm/F, and that value is verified on a cycle-by-cycle basis.
The moderator density reactivity was generated using an infinite array of L
1 s
i assemblies with the NULIF (Reference 2.1) code.
The boron concentration was varied until the moderator temperature coefficient was 0.0 pcm/F for several 61fferent fuel enrichments and burnups. The density was varied to obtain the reactivity dependence.
The different fuel enrichments and burnups had a small effect upon the density reactivity curve j
when the fuel configuration had a zero moderator temperature coefficient.
The least negative values were used.
These values are conservative because radial leakage was ignored.
With lower moderator densities, the mean free path of neutrons increases and the probability of escape from the core is higher, resulting in a more negative reactivity with lower densitics.
l Since fuel temperatures decrease during blowdown, the most negative value for the Doppler coefficient is conservative.
The coefficient of
-2.2 pcm/F is a most negative bounding value for beginning-of-cycle (BOC) conditions.
Several possible fuel cycles were analyzed to determine the range of variation of the Doppler coefficient at BOC, with the result that the most negative coefficient was greater than
-1.9 pcm/F.
The end-of-cycle Doppler coefficient may approach the value of -2.2 pcm/F, but the moderator density coefficient at EOC more than compensates for this variation.
i
).
L V'
f l
Table 2-1 Moderator Density Reactivity Relative Density 4Ak/k 1.0 0.0
.975
.006 j
.950
.028
.900
.122
.800
.535
.700
-1.311
.600
-2.600
.500
-4.628
.400
-7.780
.200
-20.854 i
References i
2.1 W.
A.
- Wittkopf, J.
B.
Andrews, et al.,
NULIF Neutron Spectrum Generator, Few-Group Constant Calculator", and Fuel Depletion Code, BAW-10115, Topical i
Report, Babcock & Wilcox, June 1976.
r
3.
Table 3-1 lists the system pressure as 2250 psia while page 6.9 indicates the system operating pressure to be 2280 psia.
Clarify which pressure is correct and what pressure was used in the Catawba /McGuire analyses.
Response
A pressure of 2250 psia is the correct reactor coolant system operating pressure as identified in Table 3-1.
The system operating pressure is quoted for the steam space of the pressurizer by convention.
The pressure at the core outlet node in the REIAP5/ MOD 2-B&W model is approximately 2280 psia.
The pressure, indicated on page 6.9; should therefore be referred to as the " core operating pressure".
l l
l I
~_
.F 5.
On page 4.4, the statement was made that the upper head model in Figure 4-2b was different from the model in BAW-10168, Figure 4-3.
However, the two figures referenced are the same.
Clarify the meaning of the statement.
t
Response
Figures 4-2 and 4-3 of the RSG LOCA evaluation model topical
- report, BAW-10168, Revision 0,
were inadvertently changed to McGuire/ Catawba figures in Revision 1
of the report.
These figures should have remained characteristic of a non-UHI plant because the model, used to
{
perform the evaluations and sensitivity studies reported in BAW-10168, was for a standard 4-loop 3411 MWt plant.
Figures l
A-1 and A-2 of Appendix A of the evaluation model report are appropriate and correct for the report.
Figure 4-2b of the McGuire/ Catawba report shows an additional node in the upper 2
plenum over Figure A-2 of the EM report.
Although the UHI system has been removed from service (valved out), the upper plenum internals for McGuire/ Catawba remain slightly altered from the conventional 3411 MWt non-UHI design. The additional node allows the representation of these UHI internals structures.
Although the RELAPS/ MOD 2-B&W noding diagrams in chapter 4 do not specifically apply for the report, they are not considered inappropriate as they represent a legitimate application of the evaluation model.
The correct noding is given in Figures A-1 and A-2 of Appendix A which documents most of the studies used by the report.
Therefore, the figures of Chapter 4 are not substantially misleading and need not be revised immediately. They will be replaced with the original Revision 0 figures when the report is next revised.
l
i 6.
On page 4.7, it is stated that appropriate time delays were used to account for delays in the activation of the emergency core cooling system (ECCS).
Provide the time delays used and i
compare them to the actual time delays to verify the conservatism of the delays used in the analyses.
Clarify if the ECC header fill time and hot wall delay were included in the delay times used in the analyses.
Provide justification j
if they were not included.
Provide the same information for other signals used in the analyses such as reactor trip, secondary isolation, etc.
j i
Response: Time delays that effect the LOCA calculation can be considered in two general categories of system and process delays.
System delays refer to the time required for actions (usually electrical or mechanical) within a system, such as the RHR system, necessary for the system to function at the level credited in the analysis.
Process delays account for i
phenomena occurring within the reactor coolant system.
The gravity delay time for ECC penetration into the lower plenum following end-of-blowdown is an example of a process delay.
Process Delavs Process time delays for the evaluation of LOCA are associated with the passage of water from the ECCS through the downcomer to the lower head during the refill phase of the accident.
Because REFLOD3B employs only one node for the downcomer and lower plenum, the transit time through the downcomer must be accounted for independent of the code.
Two effects are present that must be considered: hot wall effects and gravity
' fall time.
The BWFC evaluation model treatment of these two I
items is addressed in the response to Question 1 of the first round of questions on Revision 1 of the evaluation model topical, BAW-10168.
Hot Wall Delav:
This factor is intended to account for possible counter current flow limit effects associated with steam generation during the initial flooding of the downcomer walls with cold ECCS water.
A representative delay, based on
small. scale basically one dimensional tests, is 3 or 4 seconds during which the downcomer wall surface is being cooled and wetted to allow full penetration of ECCS fluids.
The delay is only appropriate to the initial injection of ECCS into the downcomer
- and, once penetration
- starts, should not be reapplied.
(In older evaluation models, developed prior to the availability of experimental data that demonstrates lower plenum penetration considerably before end-of-blowdown, the delay was applied at the beginning of the refill phase by delaying the initiation of accumulator injection in the reflood modelling by 3 to 4 seconds.)
Experimental data from full scale tests (Upper Plenum Test Facility,'see the response to question 1) show that partial penetration of the ECCS water into the lower plenum starts several seconds prior to end-of-blowdown and that full ECCS penetration is occurring 1 to 2 seconds before end-of-blowdown.
These data demonstrate that any hot wall effects will have taken place prior to the end-of-blowdown.
Because the actual injection of ECCS-is a
continuous
- process, uninterrupted, it is not appropriate to apply such a delay at the beginning of refill modelling.
The possibility that a delay might be appropriate for the accumulator injection during blowdown is accounted for within the bypass modelling which artificially bypasses all ECCS water injected prior to the end of bypass time.
Gravity Delav Time:
A gravity delay time appropriate for the fall of liquid from the reactor vessel nozzles to the bottom of the reactor core is applied to the ECCS injection following end-of-bypass.
The travel distance is approximately 16 feet making the delay about i second.
The evaluations in BAW-10174, were conducted with a procedure which set the end of bypass equal to the end of blowdown and j
did not. impose a gravity delay time.
The logic for this
approach was that ECCS water was present in the downcomer at end-of-blowdown
- and, therefore, there was no need to reestablish the downcomer flow column.
That approach was replaced, in the response to question 1, with an alternate CCFL based definition of.end-of-bypass in conjunction with the use of a gravity delay time.
The calculations performed for the topical have been examined for impact wit.h the result that a slight, about 0.5 second, shortening of the adiabatic heatup -
period will occur when the new model is applied.
The end of l
bypass is about 1.5 to 1.7 seconds prior to the end of blowdown and the gravity delay time only 1 second.
Leaving a net improvement of about 0.5 seconds.
Thus, the calculations performed for the report are slightly conservative under the new model.
There are no present plans to repeat these j
analyses and remove the conservatism.
System Delays System time delays appropriate for the LOCA calculations are associated with the sensing, activation, and startup for the 1
emergency injection or emergency protection systems that assist in mitigating the consequences of the LOCA.
These.
delays are discussed on a per system basis in the following paragraphs.
Reactor Trio System:
The reactor trip system acts to trip the reactor causing the insertion of control rods to shut down the nuclear reaction.
Although the system is expected to function, other reactivity ef fects such as void formation, l
which occur as a natural result of the LOCA, act far faster and more effectively to shut down the reaction.
Therefore, no crodit is taken for the reactor trip system during the bloudown phase of the calculations, and no time delay for that systems response is necessary.
1 1
4 -
,4
-r
..a.
Secondarv Isolation:
Secondary isolation is modelled in the calculations as an instantaneous event following the calculation of system pressure below the low pressure trip set point.
The effect of the steam generators in general on the i
results of IOCA calculations is limited.
The impact of varying the time of isolation of the generators is essentially inconsequential.
If anything, the instantaneous isolation is conservative as it would act to litnit the cooling of the primary system by the secondary and inhibit the generation of further trip signals.
Therefore, no delay time is utilized for this function.
The accumulators are a passive system that responds'directly to the decreasing pressure of the primary system.
> J.1 piping between the accumulators and the reactor coolant aptem is fu)?
d water during plant operation.
Therefore, no time delayr, for operability of the accumulators are appropriate or utilir.ed.
Pumned (Hich. Mid, and Low Pressure) Iniection Systems:
There are three pumped injection systems within the ECCS for the McGuire and Catawba plants.
All three are treated with the same delay times within the calculations.
The total time delay assumed from the low-lcw pressurizer pressure setpoint being reached until the safety injection flow at the RCS pressure boundary is credited in the analysis is 37 seconds.
The following events are assumed to occur during this time period:
o Safety injection signal is generated o
Diesel generator starts o
Diesel generator load sequencer actuates o
ECCS valve motors loaded onto emergency bus o
ECCS valves move to correct positions for injection mode o
ECCS pump motors loaded onto emergency bus o
ECCS pumps come up to speed and develop required head
)
?-
n The above sequence of events is currently tested per Technical Specification Table 3.3-5 Item 3.a and is to be completed i
within 27 seconds if off-site power would be unavailable.
This is conservative with respect to the 37 seconds given above.
The ECCS piping from the pump discharge to the RCS injection points is kept full of
ater per Technical Specification 4.5.2.b.1).
Therefore, no ECC header fill time needs to be included in the above delay time.
Other Systems for which Time Delays are Anoropriate:
There are no other system used in the LOCA calculations for which time delays are considered or appropriate.
I I
..m_
F, 7.
On page 4.3, it is stated that the liquid in the vessel at the 7
end-of-blowdown (EOB) was transferred from the RELAPS/ MOD 2-E&W model and placed in the lower plenum of the REFLOD3B model.
If this includes non-lower plenum liquid from RELAPS/ MOD 2-B&W, justify this transfer.
Response
As a result of modeling convention, the " lower l
3 plenum" nodes (REFLOD3B model nodes 3R and 4R) comprise the P
vessel downcomer and bottom head (below the core entrance) and represent the liquid and vapor portions of the combined vessel regions.
Regions of the reactor vessel with measurable caounts of liquid at the end of blowdown include the vessel downcomer (RELAP5 model node 302), bottom head (RELAPS model node 310), and the core bypass nodes (RELAP5 model' nodes 340 through 350).
Liquid inventory in the vessel at the end of blowdown which could be considered "non-lower plenum" liquid resides in the core bypass and the downcomer.
Typically, the f
e volume of liquid mass transferred from the core bypass to the lower plenum for the reflood calculation is small, on the order. of 25 cubic feet.
The volume transferred from the downcomer generally falls between 30 to 40 cubic feet.
5 The end of bypass normally precedes the end of blowdown but the two events were assumed to occur coincidentally in this calculation (further information on end-of-blowdown end-of-i_
l bypass is contained in the response to question number 6 of this set and in the response to question 1 of the first round of questions on BAW-10168, Revision 1).
Any liquid left in the downcomer or the core bypass region at the end of bypass-would drain to t'is lower plenum during adiabatic heatup.
So long as the drain time does not exceed the adiabatic heatup period,. the assumption of instantaneous draining will not affect the time of the start of reflood or any other process in the evaluation.
Adiabatic heatup lasts between 10 and 12
=
seconds, while the drain period for the downcomer and the core bypass is on the order of 1 to 3 seconds.
E
--i----.--m
l'
-Q
,y-l
.4 l
21.
Pages 12.5 and_12.6 identified the axial power profile as one difference between the OFA and Mark-BW fuel assemblies.
Compare the axial profile of the two assemblies.
- Also, compare the axial profile used for the Westinghouse SBLOCA analyses to the axial profile of the Mark-BW fuel assembly to verify the Westinghouse profile is still bounding.
Response
The identification of the axial power profile was to assure a
comprehensive listing of the substantial differences between the OFA and Mark-BW fuel assemblies.
The f
axial power profile at any given time depends more on plant design and operation than on fuel assembly design.
Although the two assemblies will differ in power shape for specified plant configurations, the family of possible power shapes is essentially the same for both.
Furthermore, the shape at which SBLOCA should be evaluated is selected more to assure a i
bounding SBLOCA evaluation than in anticipation that any fuel assembly would experience the shape.
Therefore, a comparison-of the axial power profiles for the two fuel assemblies would be meaningless relative to determining possible differences in response to SBLOCA.
The axial profile under which SBLOCA is evaluated is skewed toward the core outlet to a degree that it would be very unlikely t' hat any real power shape would peak at a higher core elevation.
Double peaked shapes are not allowed and the total peak is pushed to the limit of local power at the elevation of the peak.
The power shape selected for use in the BWFC small break LOCA evaluation model was presented in the response to question' number 46 of the first round of questions on the BWFC
~LOCA evaluation model topical BAW-10168.
This shape is identical to that used in the present McGuire/ Catawba SBLOCA evaluations (see Figure 15.6.5-;T of the Catawba FSAR 1987 update). 'Furthermore, SBLOCA imposed plant operating limits, i
including maximum allowable total peaking, _ are not being changed as a result of replacing OFA fuel assemblies.with Mark-BW assemblies. Thus, there is no difference in the power shapes used for SBLOCA evaluations of the two assemblies.
l I'.
4 n
..z-l t
'26.
Because the hot rod analysis using the BEACH code during reflood includes rod-to-rod radiation heat transfer,-provide.
the'following informations a.
Clarify if the hot rod in the hot assembly is a center rod or a rod on the outside of the bundle.
4 b.
If the hot rod is on the outside of the bundle, clarify how B&W determines the limiting radiation enclosure; how
+
B&W handles radiation heat transfer to bundles of different burnup; and radiation heat transfer to different bundle types (i.e., Mark-BW or OFA).
Response
The heat transfer package used by BEACH does not include rod-to-rod radiation.
Radiation heat transfer is modelled between-the fuel pellet and the cladding of a-fuel pin as part of the gap conductivity calculation. Radiation is also modelled from the cladding surface to the coolant under selected conditions as explained in Section 2.2.2 (specifically the introduction to Section 2.2.2 and Section 2.2.2.11) of the BEACH topical report, BAW-10168.
Because no rod-to-rod radiation or other pin-position-sensitive processes are given credit in the BEACH calculations, the calculations are considered independent of pin position and described as appropriate for the average pin in the hottest fuel assembly in the core.
l i
l l
1
(?
I
\\
I' 27.
Page. A. 4 stated ' the - core inlet' flow and core inventory
. increased by 2 to 5% in the case where the core was assumed to include only OFAs.
Clarify why these two parameters increased
. hen the core-flooding rate decreased in this case, w
u
Response
The important parameters in determining the increase in the core inlet flow and inventory and the decrease in core flooding rate (inches /sec.) for the OFA fuel assembly are the decrease in pin outside diameter and the increase in i,
inlet flow resistance for that design.
The decrease in fuel pin diameter causes the flow area for the OFA core to be about 7 percent larger than for the Mark-BW core.. Therefore, if the flooding rates for the two cores were. identical, the core inlet mass flow, would be 7 percent higher for the OFA core.
Similarly, for the same total carryout and flooding rate, the core inventory would also be 7 percent higher at any given j
time.
J A simplified relationship between the core inlet mass flux, G, and the downcomer elevation head, AP is G3 = bP - FRICTION LOSS
- Thus, for two cases with the same elevation head and essentially the same friction losses, the mass fluxes at the core inlet are the same.
The elevation head is created by the dif ference between the core level and the downcomer level.
For most of the reflood transient, the downcomer is full and the level used in the calculation is constant.
Because the core level is determined by the inlet volume flow less the carrycut, it will be preserved for a core flow area change so long as the carryout is appropriately correlated, i.e.
q decreases appropriately with increasing hydraulic diameter.
When most of the flow resistance is downstream of the core inlet losses, in the steam binding, none of the terms in the mass' flux equation change substantially.
w
. L:
(
- +,.,
For the change from a Mark-BW to an OFA core using the BWFC REFLOD3B code, the above conditions pertain.
The higher fuel assembly inlet resistance causes a slight degradation, about
~
2 percent, of the mass flux that shows up as a slightly
+
. reduced flooding rate.
Tne degree of mass flux change, however, does not offset the increase in core hrea, and the core mass flow and mass inventory are higher for the OFA than they are for the Mark-BW.
The foregoing applies provided that the ECCS capacity is sufficient to keep the downco4ner filled for either core.
For McGuire and Catawba, the ECCS rates are more than sufficient to maintain the downcomer full with either core design since between 15 and 25 percent of the low pressure injection is spilled out of the downcomer to the containment even under minimum ECCS assumptions.
Thus the conditions observed at the bottom of page A.4 are those that are reasonable and expected for the type of core change occurring between the OFA design and the Mark-BW.
The use of the term " mass" 'in the core inlet flow and core inventory would have made the meaning clearer in the original text.
4 t
al,
..j 29._
Clarify if the purge system is used.during power operation at Catawba and McGuirs.
For containment isolation in general and specifically for the purge system if_it is used during power operation, provide the following information:
a.
Verify the actual containment isolation valve closure times are less than or equal to the values used in the FSAR analysis of the reduction in the containment pressure resulting from the partial loss of' containment atmosphere during a
LOCA for-ECC-backpressure determination.
- Also, verify the effect of the containment pressure transient during the 14CA on valve closure times was included in. the analysis.
See'SRP Branch Technical-Position CSB 6-4.
b.
If the purge system is used during power operation, verify the containment pressure analysis assumed the purge valves were initially open.
See SRP Jranch Technical Position CSB 6-1.
Rosponse:
As. described in Technical Specification 3.6.1.9, there are three containment purge systems at Catawba:
Containment' Purge (VP) System Containment Air Release and Addition (VQ) System Containment Hydrogen Purge (VY) System As stated in this specification, only the second of these, the VQ System, may be used during Modes 1-4.
The actual valve closure times for the subject valves, VQ-2 and VQ-3 or VQ-15 and VQ-16, are required to be less than 5 seconds, the maximum isolation time given in Technical Specification Table.3.6-2.
As stated in Supplement 2 to the Catawba Nuclear Station Safety Evaluation Report, NUREG-0954, p.6-1, the staff had concluded that "the effect on containment pressure of air lost l
through open purge / vent lines is insignificant."
Therefore, there was no explicit value assumed for valve closure time in l
the FSAR analysis.
As stated in Item la of Catawba FSAR Table 9.5.10-1, the VQ valve " actuators will close the containment-isolation valves assuming full containment pressure
[
differential and resultant flow."
Since the amount of air lost through these valves at the start of a LOCA is
,t; L
y insignificant, the ' containment pressure analysis ' makes. no explicit assumption about the valves.
i The containment purge system information given_ above is j
generally applicable to McGuire.
The exceptions are:
~
l 1)
The VP system is' allowed to be used up to 250 hours0.00289 days <br />0.0694 hours <br />4.133598e-4 weeks <br />9.5125e-5 months <br /> per j
?
calendar year at McGuire.
A limit on the use of this system is consistent with BTP CSB 6-4.
'In practice the system is not used during power operation since containment pressure is maintained with the VQ system.
2)
The VQ (and VP) system maximum valve isolation time is 3 seconds, which is conservative with respect to the 5 second Catawba time, i
[
i
.