ML20044G992
| ML20044G992 | |
| Person / Time | |
|---|---|
| Site: | Mcguire, Catawba, McGuire  |
| Issue date: | 05/26/1993 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20044G989 | List: |
| References | |
| NUDOCS 9306070155 | |
| Download: ML20044G992 (6) | |
Text
e p**>D GEG UNITED STATES 4
[
j NUCLEAR REGULATORY COMMISSION I
2 WASHINGTON, D.C. 20h65-0001
%,...../
SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION DUKE POWER COMPANY. ET AL.
CATAWBA NUCLEAR STATION. UNITS 1 AND 2 MCGUIRE NUCLEAR STATION. UNITS 1 AND 2 DOCKET NOS. 50-413'. 50-414. 50-369 AND 50-370
1.0 INTRODUCTION
In NUREG-0954, " Safety Evaluation Report Related to the Operation of Catawba -
l Nuclear Station, Units 1 and 2," Supplement No. 6, Para 6.2.5, '(SSER-6) dated i
May, 1986, the staff approved the licensee's April 25, 1986, plan.for closure of the remaining issues related to combustible gas control. The three issues-addressed in the plan are the subject of Catawba Nuclear Station Facility Operating License Condition No. 14. They are:
(1) Thermal response of the containment atmosphere and essential equipment for a spectrum of accident sequences using revised heat transfer models, (2) Effects of upper compartment burns on the operation and survival of air return' fans and ice condenser doors, and-(3) Operability of the glow plug in a spray environment typical of that expected in the upper compartment of the containment.
Catawba SSER-5 provided a detailed discussion of the issues.
By the time of preparation of SSER-6 the staff had completed its review of (3) above and was able to report that the issue of glow plug operability in a spray environment-l has been resolved. This safety evaluation addresses (1) and (2) above, the two remaining hydrogen control issues.
Although this evaluation specifically refers-to Catawba, the information and conclusions are appl.icable to McGuire.
l 4
2.0 DISCUSSION AND EVALUATION 2.1 Thermal Response of the Containment Atmosphere and Essential Eouipment~
for a Spectrum of Accicents Usina Revised Heat Transfer Models Backaround (Recao from SSERs-5 & 6): At the time of issuance of the first Catawba unit operating license, the licensee had performed numerous analyses of the containment atmosphere pressure and temperature response during degraded core accidents with associated hydrogen release and combustion.
Calculations were performed, using the CLASIX code, to determine the 9306070155 930526 PDR ADDCK 05000369 p
PDR l
\\
492.
l i
sensitivity to variations in assumed combustion parameters, assumptions regarding availability of safety systems, and variations in the hydrogen and steam release to the containment. Although many analyses were performed, for i
purposes of determining environmental conditions for equipment survivability requirements only the S D (2-inch hot leg break LOCA with ECCS failure) i z
sequence was considered. The licensee stated that this sequence represented a reasonable upper bound scenario. The staff accepted this position on an interim basis (Ref: McGuire SSER-7) with the condition that licensee pursue this issue further.
The staff had concerns with the use of the CLASIX code.
l One problem was that it did not predict burning in the upper containment t
compartment, whereas,.Sandia investigations had. The staff also determined that the licensee's CLASIX code contained errors in the heat transfer models which would cause underprediction of containment temperatures.
Discussion:
In response to the NRC staff's concerns and in conformance to its l
April 25, 1986, plan, the licensee conducted additional analyses of accident sequences. A spectrum of accident sequences that envelop a wide range of reactor vessel pressures was analyzed. Also,.because of staff concerns regarding ECCS recirculation failure, a recirculation failure sequence was added. The sequences include:
f 1.
S D 6-inch hot leg LOCA with ECCS failure, 3
l 2.
S D 2-inch hot LOCA with ECCS failure, 2
3.
S H 2-inch hot leg LOCA with ECCS recirculation mode failure, and 2
4.
TMLU Loss of main feedwater with failure of auxiliary feedwater l
and ECCS.
These sequences were examined with the Modular Accident Analysis Program (MAAP) Versions 2.0B and 3.0B to determine steam and hydrogen mass / energy release histories.
Similar analyses were performed using the MARCH Code on D.C. Cook, Sequoyah, and McGuire and the results compared. Code input assumptions for the licensee's MAAP/HECTR analyses regarding (a) fan operation and timing, (b) ignition and propagation criteria, (c) combustion completeness, (d) flame speed, (e) heat transfer coefficients, (f) ice condenser drain temperature, and (g) containment compartmentalization are best estimate values and are identified in the April 25, 1986, and March 25, 1993, letters. For redundant equipment, both trains are assumed operable. Hydrogen burn parameters were based on a study of experiments carried out under the sponsorship of NRC, the ice condenser owners, and EPRI. These experiments include the large-scale tests at the Nevada Test Site as discussed in Catawba SSER-5.
For each sequence, ECCS recovery was delayed until hydrogen production had peaked. Hot leg breaks were chosen in order to minimize hydrogen holdup time.
None of the MAAP sequences resulted in clad oxidation greater than 25.2%.
Since the Hydrogen Rule (10 CFR 50.44) specifies a 75% zirconium / steam reaction capability, and the hydrogen production resulting from these sequences was considerably less, the licensee also analyzed the events using a non-mechanistic model to extend the hydrogen source term to 75% clad i
oxidation. The non-mechanistic model used to extrapolate hydrogen production to 75% clad oxidation was consistent with that used by the Hydrogen Control
3 l
i Owner's Group as described in the related staff Safety Evaluation Report (NUREG-1417).
The steam and hydrogen release histories from the MAAP analyses were used as l
inputs to the HECTR code to determine the containment pressure and temperatu're' responses, globally, and individually for the upper containment, upper plerum, lower plenum, lower containment and dead-ended volumes. The HECTR Code, described in NUREG/CR-4507, is a lumped-parameter containment analysis code j
~
developed for calculating the containment pressure and temperature responses to hydrogen combustion. HECTR uses default hydrogen concentration limits for
. ignition and upward propagation which are lower than the values set in CLASIX.
j MAAP is an industry-developed code which has not-been reviewed by the staff.
Therefore, the steam and hydrogen release histories from MAAP were compared to i
l steam hydrogen release histories predicted by the staff-approved MARCH code for other ice-condenser facilities (i.e., McGuire, Donald C. Cook and Sequoyah).
j i
The MAAP/HECTR analyses resulted in global peak pressures and temperatures below the corresponding CLASIX results for all four sequences. However, the MAAP/HECTR analyses did result in some higher peak compartment temperatures than CLASIX.
The following table presents comparisons of the peak temperature results.
l PEAK TEMPERATURE COMPARISONS (deg.F.)
SS S@0.3 S3B IMLQ CLASIX Upper Compartment 238.1 2r 239.9 284.2 162 Upper Plenum 609.4 851.4 659.7 775.2 1517 Lower Plenum 680.9 887.7 947.5 751.1 228 l
Lower Compartment 678.9 782.8 848.5 637.0 1138 Dead-Ended Compartment 238.1 301.3 328.6 263.7 238 The significant differences between the earlier CLASIX predictions and the i
MAAP/HECTR predictions are the upper containment and lower plenum burns
)
predicted by MAAP/HECTR but not CLASIX. This is attributed primarily to the 1
higher ignition and propagation limits in CLASIX noted above. -Lower plenum j
burns are of little consequence since there is no equipment located there.
i However, the upper containment contains vital air return fans whose survivability could be-impacted by burns.
By letter dated February 3,1993, the staff informed the -licensee that the accident scenarios selected for analysis encompassed an. appropriate range of events consistent with 10 CFR 50.44(c)(3)(vi)(B)(3).
i 1
l l
1 4
2.2 Effects of Upoer Containment Burns on the Ooeration and Survival of Air Return Fans and Ice Condenser Doors Backaround (Recao from SSERs-5 & 6): As noted above, the licensee's CLASIX analyses indicated no burning in the upper compartment. However, In the NTS tests, discussed in SSERS, thermal ignitors reliably ignited hydrogen-steam-air mixtures with hydrogen concentrations as low as approximately 5.5% volume l
percent. Also, Sandia MARCH /HECTR (Ref:NUREG/CR-3912) analyses indicated higher steam fractions in the lower compartment and numerous upper compartment burns.
In addition, fog analyses sponsored by the Ice Condenser Owners' Group indicated an increased flammability limit in the upper plenun.
Based on these findings the staff considered hydrogen burns in the upper compartment likely for many degraded core scenarios. Therefore, in addition to requiring l
upgraded thermal response analyses, the staff required the licensee to address the effects of upper containment burns on the operation and survival of i
certain safety-related equipment (i.e., air return and hydrogen skimmer fans and ice condenser doors).
These fan concerns relate to (1) differential pressure loads causing the fans to windmill thus generating sufficient current to trip on overload, and (2) ice condenser door reverse differential pressure loads.
[The fan temperature profiles for each sequence were determined independently using a different radiative heat transfer model. These profiles were found to be less limiting than the fan and cable profiles used in the environmental qualification of electrical equipment per 10 CFR 50.49].. By letter dated March 25, 1993, the licensee submitted additional information regarding the remaining issues of fan and door survivability.
Fan Survivability: Based on a licensee analysis, a differential pressure of 4 psid lasting 10 or more seconds is sufficient to trip the air return fans.
To i
address the concern of forward-direction fan windmilling, the licensee analyzed the fan and duct acting as a generator under the forward pressure differential.
The S The S D se,D sequence resulted in a peak pressure differential of 2.7 psid.
quenced produced a peak differential pressure of 3.8 2
psid.
The S H sequence produced a peak differential pressure of 2.7 psid, and theTMLUpro$ucedapeakdifferentialpressureof4.25psid.
The latter exceeded the 4 psid value established as the minimum & needed to trip the fan, but only for 5 seconds, half the time period necessary for the trip.
The I
analyses indicate that windmilling during the four ace.ident sequence conditions would not produce the fan motor current necessary to trip or overspeed the fans.
The hydrogen skimmer fans are less limiting than the air l
return fans.
Ice Condenser Door Operability: Upper compartment burning may result in reverse differential loads on the intermediate deck and lower compartment i
inlet doors of the ice condenser. To some extent the loads are relieved by bypass flow through ice condenser drains. The limiting component with respect to the stress loads imposed by reverse pressure are beams up the middle of the l
door frames. The differential pressures that the doors can withstand are 6 0
psid and 7 psid respectively (Ref: April 25, 1986, letter).
The S 0 resulted in reverse & 's less than 2.5 psid. The S D and TMLU sequen. sequence
{
3 g
ces produced reverse &'s of less than 4 psid.
The S,H sequence produced reverse i
&'s of less than 3 psid.
The results of the MAAP/HECTR analyses indicate 1
i I
l J
.y.
l; 5
that in no case do the calculated differential pressures approach the values which were calculated to result in damage to the door beams.
3.0 CONCLUSION
By this Safety Evaluation, the staff concludes that the two remaining-licensing issues identified.in SSER-6 and in Condition 14 of' Facility Operating License No. NPF-35 for Catawba Unit I consisting of (1) thermal response of the contajnment atmosphere and essential equipment for a spectrum j
of accident sequences using revised heat transfer models, and (2) effects of
{
upper compartment burns on the operation and survival of air return fans and I
. ice condenser doors, have been resolved to the satisfaction of the staff.
In consideration of the similarity in design of the Catawba and McGuire plants and the applicability of the supporting documentation and analyses submitted by the licensee to both the Catawba and McGuire Nuclear Stations, the NRC staff concludes that the remaining provisions.of 10CFR 50.44(c)(3) have been i
acceptably responded to for-both stations.
Principal Contributor:
W. Long Date: May 26, 1993 i
i l
l 1
i e
- -.. - -, -,