ML19354E804
| ML19354E804 | |
| Person / Time | |
|---|---|
| Site: | Vermont Yankee  |
| Issue date: | 01/16/1990 |
| From: | VERMONT YANKEE NUCLEAR POWER CORP. |
| To: | |
| Shared Package | |
| ML19354E801 | List: |
| References | |
| PROC-900116, NUDOCS 9002010380 | |
| Download: ML19354E804 (12) | |
Text
.
ALTERNATE TESTING FOR CO2 FULL DISCHARGE TEST-BY PRESSURE BUILD AND TRACER GAS DILUTION TESTING O
Prepared by: . - h-tC.-09 E. J. T&intor, f)fre Protection Engineer Reviewed by: m' 8 MU D. C. Girroir, Senior QA Edgirieer NA
$ IW( ff984-PORC (/ l Approved by: //- > M I u / A. [a 9 M k H. M. ~ Mete 11~,
R. D. Pagodin, Technital Support Stiperintendent J Engin}er/x/rr ring S'upport Supervisor _ ;
l N LeAM // /& ff4 '
R. J. riczyk, P Manager i
IndependentRevieweri i A.
h Klein' 9 YNSD Fire Protection' Engineer h k* W b< W l hl y gy j raute es., q u m4 '
m , a- & &~- wM -
9% A) 0 M/fAl W E
, d .;
- ..,(-
F 9002016380 900116 rio
-PDR ADOCK 05000271 si P , PDC-
]
ggp /Q /9 B9, A% W r F^ ^ ^&^
- m ._,7 _ <. _ , , .
(A=sp-J& &)A(:%-
. [
Table of Contents. I
- 1. Introduction and Summary *
'2.. Technical Evaluation '
- 3. 50.59 Safety Evaluation j
- 4. Test Procedure l
a
- 5. AP-0017 Calculations - Cable Vault CO2 Free Vent Area
- Calculation of Egivalent Leakage Area in the~
Cable Vault
- Calculation of Peak Pressure During CO2 Discharge in ;
in the Cable Vault .
- CableLVault Air Leakage by Tracer-Dilution Method: )
- Cable Vault Volume _
- Cable Vault CO2 Descending' Interface Model Hold Time .
- Cable Vault Leakage /Holditime for. Stirred' Tank Model
- 6. References !
4 I
5 T
O 1
i ummary ESD completed the prescribed testing and achieved the following results:
- 1. Per the pressure build section of " Cable Vault Enclosure Integrity Test" ,
the calculatr.d maximum pressure (with the exhaust damper open for 200 seconds) is 3.45 inches of H2O with an equivalent leakage area of 250 sq inches. This compares very favorably with NFPA 12, Table 2-6.2.3 ( Ref 3) for light.
construction strength of 5 inches H20.
- 2. Per the SF6 section of the " Cable Vault Enclosure Integrity Test", ESD found that the air inleakages into the Cable Vault were significant enough such that the descending interface model could not be used.
l
- 3. Per the SF6 section of " Cable Vault Enclosure Integrity Test", ESD found !
- the air exchange rate was 640 scfm. This exchange rate, coupled with RPS-MG fens'and CO2-wall temperature gradients provided enough mixing to approach a
" stirred mixed tank" model as compared to the " descending interface" model.
The stirred mixed tank model showed that a total room CO2 concentration.of 50% '
(with no interface) could be maintained for 14 minutes and 50 seconds. 1 Additionally, the descending interface model was calculated for information and comparison purposes. It showed that a CO2 concentration of 50%
could be maintained for 25 minutes.
JD noted that different models required different CO2 distributions. ESD QtilizedNFPA12toprovidetheCO2distributioninformationforthemixed tank model. (Note that this standard is based on NFPA experience).
- 4. Based on the above and attached information, ESD has evaluated that the Cable Vault CO2 system can perform its intended function and should-be declared operable. An attached 50.59 safety evaluation shows that plant configuration is maintained and supported by the alternate test.
Additionally, ESD has collected the required technical and quality '
documentation to support the above results (See attachments),
i O
.l
([3) :
. f
! s Introduction r
- This report summarizes the pressure build /SF6. testing that VY performed on
! the Cable Vault Room to determine the pressure buildup and CO2-distribution
- after a CO2 injection. VY sought this'information to help provide additional l technical basis for considering the Cable Vault CO2' system operable. VY declared;t the system. inoperable after a June 89 VY/NRC telecon. During the telecon the NRC informed VY that_as the " Authority Having Jurisdiction", thecNRC con'aidered-l the 1985 NFPA 12 functional. testing. mandatory <for VY.-After subsequent discus- '
sions, a NRC to VY letter (Ref 11) informed VY that either a functional test-
! or suitable alternative (as agreed to:by the NRC)Eshould be-used--to address ~this l 1ssue. VY proposed alternate testing to the NRC at an October 25,1989 meeting
- i lieu of performing a CO2 full discharge test (Note VY management evaluated no CO2 full discharge test as a small but potential increased risk for a scram ). VY wrote, PORC reviewed and. plant Manager approved the' alternate' test procedure (see enclosure tab). +
i At the October 25,1989 meeting, VY presented =both the test method and the acceptance criteria. The acceptance: criteriaiwere:
- 1. per NRC-VY discussions, that the peak' pressure would: be determined and verified acceptable.
l .
j 2. per engineering practice, that the alternate test method.("Ldescending j interface " model) required verification. Although-not_specifically mentioned ,
at the meeting, VY-ESD had already specified SF6 leakage testingLfor-this verification per ASTM E741-80. 1 4
'3. per SER for Technical Specification Amendment 43, that the testing denonstrate that-the Cable Vault CO21 system be able to sustain'a 50% CO2-concentration for a hold time of 10. minutes. Additionally VY' stated that the 50% CO2 " descending interface" will not' descend to less than a critical- "
height of 7 feet after 10 minutes.
a y
4 g
. - - - _ . . . .. , .. _. ._ . . _ . - ,.. m ,
Tochnical Evaluation I. Standards Used for Developing VY Test Procedure A) Background VY's test procedure provided a method to determine that the Cable vault Total Flooding carbon Dioxide extinguishing system would function properly. The functional criteria for the system are specified by the Safety Evaluation Report for Technical Specifications, Ammendment No. 43. The functional criteria specified are a CO2 concentration of 50 percent maintained for a period of 10 minutes.
At the time of installation, the system was tested to the criteria of the National Fire Protection Association (NFPA) Standard for CO2 systems (NFPA-12, 1977). The standard of record at that time did not require a full discharge test to demonstrate operability. Recently, the NRC, claiming the powers of the Authority Having Jurisdiction, has interpreted that a fuil discharge test is required to prove the adequacy of the CO2 system (ref. 9).
In the NRC response to the VY response to the notice of violation, the NRC stated that they would entertain an alternate to the full discharge test if a conclusive test method could be designed (ref, 11).
B) Pressure Build Test - NFPA-12 The 1989 Edition of NFPA 12A, the standard on Halon 1301 Fire Extinguishing Systems includes an Enclosure Integrity Procedure in Appendix B.
This test was developed by a research team for the subcommittee of the NFPA 12A committee. The subcommittee was charged with the task of developing an tptable ression systems.
alternative to full discharge testing of Halon 1301 gaseous fire ,
Appendix B to the Standard is the result of the efforts o the research team and the subcommittee. This alternative test was presented to the NFPA 12A committee and approved.
of the new edition of the code. This draftThe test was then added to the draft was then reviewed by the membership of the NFPA and approved by them for inclusion in the 1989 edition of the code.
In that the gas densities for Halon 1301 at a 6 percent concentration and CO2 at a 50 percent concentration are nearly identical, the Enclosure Integrity Procedure can be readily applied to CO2 Total Flooding systems. (A 6% Halon concentration has a density of 0.0935 lbs, per cubic foot, while a 50%
CO2 concentration has a density of 0.0945 lbs. per cubic foot.) With this in mind a test procedure was developed to utilize Appendix B of NFPA 12A for the datormination of operability of the cable Vault CO2 system.
C) Tracer Gas Dilution Test - ASTM E 741-80 To further ensure that the results of the Enclosure Integrity procedure are valid, a standard ASTM tracer gas dilution test was included. This test provided a standard method for determining the air exchange rate within the Cable Vault. Utilizing the results of this test and a standard calculation for the decay over time, a determination of concentration and hold time was possible.
O
II.. Major Test Paranoters
/~'1o major-issues were addressed by the procedure with regards to system ;
kr/rability: l I
A) Ability of the enclosure to withstand the pressures generated ,
during a discharge of CO2. , s j NFPA 12, section 2-6.2 addresses the issue of. Pressure Relief Venting.
In this section guidance is provided with respect to the required free vent
l cllowable strength of the enclosure. The CO2 discharge rate is specified by the original system design. The values utilized for allowable strength are based on general construction practices. Due to the nature of the doors in the l
I room, the values for light building. construction were utilized in determining ;
the minimum acceptable free vent area for this system. The actual calculations ,
for the determination of free vent area are included in the calculations section of this report. The value determined by this calculation is a: free' vent area of 130 square inches. This is the minimum amount of leakage area required for this enclosure to guarentee that room integrity is maintained during discharge.
The Enclosure Integrity Procedure quantifies leakage area within the enclosure. The leakage area is. identified as an Equivalent Leakage Area'(ELA) and is providad in unigg of square. inches. The. Enclosure Integrity procedure- ,
neasured an ELA of 250 square inches. Of this' leakage area, 56 percent was identified as that vent area provided by the exhaust damper. This is a best assible configuration in that the vent path is high in the enclosure,'thus jtingexcessairduringdischarge.
! A comparison of the calculated Free Vent area as opposed to the -
m2asured Equivalent Leakage Area clearly shows an acceptable amount of free o vont area. This ensures that the integrity of the room will be maintained l during a discharge of CO2.
I Further calculations were~ conducted to determine the maximum pressure-which would be experienced during a discharge. The data for these calculations was gathered as a portion of-the Enclosure Integrity procedure.
, The predicted- maximum pressure experienced during a CO2 discharge is '3.45 jtr7 inches of water. This value is well within the limit of 5 inches of water established by NFPA 12 for light building construction.
O
B) Distribution of CO2 within tho onclosuro.
ESD conducted a review of the original design criteria and calculations the system. This review revealed that the system had been designed
,]lizingacceptedindustrystandardsandpractices.
(s The placement and sizing of nozzles and distribution piping was consistent with NFPA 12 recommendations and requirements. To further verify the acceptibility of the system, an as built review of the system was conducted. This review further verified the accoptability of the system.
The Enclosure Integrity procedure includes a model for predicting the height of the descending air to CO2 interface with respect to time. The model assumes leakage within the enclosure which is equally distributed between the floor and ceiling levels. It also assumes a reasonably static environment with '
no techanical mixing. Differential pressures across enclosure boundaries are assumed to be negligible. The model provides a conservative prediction of the CO2 concentration and maximum hold time for a specific enclosure.
The results provided by the Enclosure Integrity procedure for the cable i vault, predict a 50 percent concentration at a level 7 feet above the floor for a period of 25 minutes.
Although the leakage paths within the Cable Vault have been determined to be well distributed, other parameters such as mixing (which occurrs within the room) and differential pressures (between the Cable Vault and surrounding rooms) required further evaluation.
Mixing of the CO2 and air in the Cable vault is provided by the RPS MG
! set rotating assemblies, various leakage paths, and convection due to the mperature differentials between the cold CO2 and the warm concrete walls of enclosure.
Pressure measurements were made relative to the pressure within the enclosure. These measurements indicated a variety of differential pressures across the enclosure boundaries. This information further reinforced the nesd for the tracer gas dilution test.
Due to the potential for mixing, and the differential pressures noted, the assumptions made for the Enclosure Integrity procedure became less valid.
It appears that the conditions within the Cable Vault more closely approximate a " mixed tank" model. With this in mind, a standard tracer gas dilution test (utilizing sulfur Hexafluoride (SF6)) was conducted to determine'the actual air exchange rate for the Cable Vault. The results of this test verified the The assumption that the room more closely approximates a " mixed tank" model.
air exchange rate was determined to be 640 scfm with the Cable Vault in a CO2 discharge configuration. Utilizing a standard calculation for the decay over tiae, it was determined that a 50 percent concentration of CO2 will be maintained for a period of 1-3 minutes following discharge.
An additional 1 minute and fifty. seconds of hold time is developed between the establishment of a 50 percent concentration and the end of the discharge period. Therefore, the 50 percent concentration will be maintained for a total of 14 minutes and .50 seconds.
The additional 1 minute and 50 seconds assumes that a concentration of l 65 percent is reached at the end of the discharge period.
l l
l
Support for this ossunption is dovolopod using NFPA 12, Fig. A-2-1(A).
This figure provide a graphical means for determining the percentage of CO2 roceived with tge nelosure. The injection factor is determined by dividing eg y the available CO2. In this case the injection factor is (j, protecteda cubic feet per lb. of CO2.
Three curves are provided on the graph. The lowest curve represents no l efflux, the middle curve represents free efflux, and the top curve represents )
pstfect efflux. Perfect efflux would be the situation where all of the air was '
ovacuated, to be replaced with a homogeneous mixture of CO2 and air. This would be accomplished with mechanical assistance. The free efflux situation would be an enclosure with normal leakage paths and no additional means for venting air or CO2. The situation in the Cable Vault, based on the delayed closure of the exhaust damper, is assumed to fall between these two curves.
Assuming a conservative value for the efflux achieved, midway between free i
efflux and perfect efflux, provides a concentration of 65 percent of CO2 received at the end of discharge.
l l
O .
i t
1 1
1 O
., -, nn
c) Ability of tho onclosuro to caintain the requirod concontration of- '
gas over time.
The Enclosure Integrity procedure allowed for the quantification of Equivalent Leakage Area. This not only demonstrated the ability of the I o losure to withstand the anticipated pressure incursion, but also provided l data which supports the ability of the room to maintain the concentration.. l If we were to assume a descending interface model for this enclosure, a " worst caso" assumption, then a concentration of 50 percent CO2 could be expected-for 25 minutes-at a critical height of 7 feet above the floor. ,
l The Tracer Gas dilution method for. measuring air exchange provided an i actual measurement for the air exchange rate within the Cable Vault. This '
air exchange rate, in conjunction with'a_ standard decay rate calculation, provides the prediction of the maximum hold time at 50 gercent concentration.
Tho use of the tracer gas dilution assumes a "best case scenario-for the CO2 ,
discharge. In this case a 50 percent concentration of CO2 is predicted for 14 i ninutes and 50 seconds evenly distributed throughout the room.
4 O
lO .
l t
III. Conclusion The evaluation of the test results and data gathered as a result of the T st Procedure indicates that the Cable Vault Total Flooding CO2 system will form as designed. The system will inject a concentration of 65 percent CO2
. the enclosure will maintain a concentration of 50 percent for a-minimum
-period of 14 minute,q,and 50 seconds. The peak pressure incursion will be approximately 3,45 %ches of water. These results provide assurance that system design, licensing criteria, and regulatory requirements are satisfied.
I
?
I i
O O
l l
l
50.59 Safety Evaluation Th3 purpose of this safety evaluation is to ensure that the attached rossure build /SF6 testing provides assurance that the Cable Vault 2 system will operate as designed and that no unreviewed safety stion exists. When evaluating this alternate testing ,it is helpful i to examine the design basis of the CO2 full discharge test which achieves l 2 purposes:
- With adequate sensing devices, the test can prov'ide CO2 distribution concentrations, placement of these sensors are based on engineering judgement.
- The test can provide peak pressure buildup within the enclosure.
Typically, this test is performed on a functional basis versus quantifiable basis ,ie if inadequate venting exists , this test will fail the enclosure instead of measuring peak enclosure pressures.
VY considers that the pressure build /SF6 method meets these design criteria as compared to the CO2 full discharge test based on the following:
- CO2 distribution is determined by-combining a very sensative bulk )
concentration test (ie tracer gas) with accepted bulk venting (ie NFPA 12) bracketed by 2 coraervative/ standard distribution nodels (ie " descending interface / plug flow" model and "back-mix, stirred. tank" i model as described in Reference 14).The net result is that the actual CO2 .
concentration is between a model that predicts 50% CO2 concentration for 25 I minutes and a model that predicts 50% concentration for 14 minutes 50 seconds both of which exceed the minimum requirement of 50% concentration for 10 inutes.
The pressure build test as supported by NFPA 12A App B provides an approved method for determining pressure buildup within a room. This is achieved by i determining an equivalent leakage area (ie equivalent resistance to flow) which ;
can be accurately extrapolated to actual test conditions without risk of l
equipment damage. In this way, both room integrity and room venting needs j can be determined. The net result is a non destructive test that provides ;
more information than the CO2 full discharge test.
4
BasOd en the abcv0 ,th? following 50.59 stotocents ccn be made par ,
VYApr 6002.01: '
- a. This test does not increase the probability of occurrence of an accident Joviously evaluated in the FSAR. The CO2 full discharge test is not specifically described in the FSAR or SER to TS ammendment 43 but is referenced vic NFPA 12 - 1985 edition as a required test. The intent of the CO2 full discharge test is met by the pressure build /SF6 testing. In fact, the pressure build /SF6 test did not physically challenge / alter the Cable Vault Structure which further ensures that the system would perform its intended function "
wh3n called upon .
1.b. This test did not increase the consequences of any of the accidents os described in the FSAR for the same reasons described in 1.a.
1.c. This test did not create the possibility of an accident not bounded .
by those specified in the FSAR as described in 1.a.
- 2. This test did not increase the probability that a piece of equipment n2cessary for Nuclear Safety would malfunction. The Cable Vault automatic total flooding CO2 suppression system is a non-safety class system. The CO2 system provides fire protection for safety class electrical cable and equipment and is designated as a Vital fire protection system. The pressure bu11d/SF6 tost provided additional assurance that the system would perform as intended without challenging the Cable Vault room or equipment. It assured that the CO2 distribution, pressure build profile, and Cable Vault venting area were appropiate for the system.
1 This test did not increase the consequences of any malfunction of equipment 9,uipmentintheCableVaultaboveanyanticipatednormalenvironmentiecessary for Nuclear S th2re is reasonable assurrance that all cable vault equipmenc is in the same condition after the test as it was before the test (excepting normal wear).
- 4. This test did not create the possibility of an equipment malfunction not cnticipated in the accidents specified in the FSAR as described in "3."
cbove.
- 5. This action does not reduce the margin of safety defined in the basis of any Technical Specification. The pressure build /SF6 test assures that th2 Cable Vault CO2 system will perform its intended function while providing cdditional engineering information to ensure that the margin of safety is nointained.
Based on the above analysis, there is no unreviewed safety question.
It should also be noted that additional assurance of CO2 Cable Vault system parformance is provided by a) a supplemental 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> fire watch of the Cable Vault Room that will be provided during the initial review of this report by the NRC and b) second shot capability available from the switchgear rooms.
l 0
-