ML18142A503
| ML18142A503 | |
| Person / Time | |
|---|---|
| Site: | Surry  |
| Issue date: | 05/21/1985 |
| From: | Jape F, Macdonald J, Mellen L, Whitener H NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML18142A501 | List: |
| References | |
| 50-281-85-11, NUDOCS 8507010239 | |
| Download: ML18142A503 (15) | |
See also: IR 05000281/1985011
Text
Report No.:
UNITED STATES
NUCLEAR REGULATORY COMMISSION
REGION II
101 MARIETTA STREET, N.W.
ATLANTA, GEORGIA 30303
50-281/85-11
Licensee: Virginia Electric and Power Company
Richmond, VA
23261
Docket No.:
50~281
License No.:
Facility -Name:
Surry
Inspection Conducted:
Aprii ~ - 12, 1985
Inspectors: ~
- /'. tJi~
H. L~ener
- ,~*cs.~[, ... 44 ~
(il-ltv L3 µ~~-/jr71A /f~
J. p Macdona 1 d
~-.
-* * -
'--:--'
- Accompanying Personnel:
Jape, Apri 1 11 - 12, 1985
Approved by:=--::--~~t:'A~~~~__:'==1-~4-~----
F. Jape,
ct on C i
Engineering Branch
Division of Reactor Safety
SUMMARY
S - 7- I - ,rs;-
Date Signed
.5/-z.1/;;r-
Date Signed
. t:i'f&n'll
. 'J/'l..1 /vs-
Date Signed ..
Scope:
This rbutine, announced insp~ction entailed 100 inspect6r~hours on site
in the areas of independent analysis of integrated leak rate test results;
review *of local and integrated leak rate test procedures, data, data analysis;
and test reports.
Results:
Multiple examples of one v*iolation were id~ntified:
Violation
(281/85-11-01) - Failure to follow the requirements of Appendix J to 10 CFR 50
(paragraphs 6~a, 9.a, and 9.c).
8507010239 850530
ADOC~ 05000281
G
1.
Persons Contacted
Licensee Employees
REPORT DETAILS
- H. L. Miller, Assistant Station Manager (NS&L)
- D. L. Benson, Assistant Station Manager (O&M)
L. Curfman, Corporate Office
- R.H. Blount, II, Supervisor Performance and Technical
- E. J. Turko, Engineer
Other Organization
Stone and Webster
- R. I. Samson, Engineer
NRC Resident Inspector
- D. Burke, Senior Resident Inspector
- Attended exit interview
2.
Exit Interview
The inspection scope and findings were summarized on April 12, 1985, with
those persons indicated in paragraph 1 above.
The inspectors described the
areas inspected and discussed in detail the inspection findings.
No
dissenting comments were received from the licensee.
The following new
items were identified during this inspection.
a.
Violation
Violation (281/85-11-01) - Failure to follow the requirements of
Appendix J to 10 CFR 50.
The following examples were identified:
(1)
Contrary to the requirements of paragraph III.A.1.(d), the
licensee failed to vent and drain certain penetrations which in a
post accident condition are potential air leakage paths (paragraph
6.a.).
(2)
Contrary to the requirements of paragraph III.A.1.(a), the
licensee failed to restart the Type A test after the isolation of
excessive leakage which prevented meeting the acceptance limit of
paragraph III.A.5.(b) (paragraph 9.a) *
2
(3)
Contrary to the requirements of paragraph III.A.3.(b), the
licensee failed to determine the reason, take corrective action,
and perform a successful supplemental test when the supplemental
test result did not agree with the Type A test result within the
allowable tolerance of+/- 0.25 La (paragraph 9.c).
b.
Inspector Followup Items (IFI)
(1)
IFI (281/85-11-02):
Review rev1s1on of CILRT procedure, PT 16.3
for proper vent and drain requirements {paragraph 6.a).
(2)
IFI (281/85-11-03):
Review licensee action with regard to
measuring local leakage downstream (low pressure side) of
isolation valves (paragraph 6.b.).
c.
Unresolved Item (UNR)
UNR (281/85-11-04):
Review licensee action to qualify the
secondary system as a containment leakage barrier or to locate and
repair the source of containment leakage into the secondary system
(paragraph 10).
The licensee did not identify as proprietary any of the material provided to
or reviewed by the inspectors during this inspection.
3.
Licensee Action on Previous Enforcement Matters
This subject was not addressed in the inspection.
4.
Unresolved Items
Unresolved items are matters about which more information is required to
determine whether they are acceptable or may involve violations or
deviations.
One new unresolved item identified during this inspection is
discussed in paragraph 10.
-5.
Review of Surry Leak Rate Program (61719) (61720)
This inspection was conducted to review certain issues that developed from
the September 11 - 14, 1983, Unit 2 containment integrated leak rate test
(CILRT) and to review the overall leak rate test program.
a.
Issues Ad9ressed During This Inspection
(1) Pass or fail "as-found
11 status of the September 1983 Unit 2 CILRT
(Type A test) .
3
(2)
NRC position on leakage from the containment to the secondary
system.
(3) Manipulations performed to isolate leakage paths during the CILRT.
(4)
Type A and Type C leak rate test procedure valve alignments.
(5)
Method used in performing Type C leakage measurements on the low
pressure side of the isolation valve.
(6)
Method used in performing the Type A supplemental (verification)
test.
(7) Application of Bechtel Topical Report, BN-TOP-1, Revision 1, test
criteria to termination of the CILRT.
(8)
Method of establishing test start time.
(9) Analysis of CILRT data.
b.
Regulatory and Licensee Referenced Documents
(1) Appendix J to 10 CFR 50
(2)
(3)
Bechtel Topical Report, BN-TOP-1, Revision 1, November 1, 1972
(4)
NRC Letter dated January 15, 1973,
11 Evaluation of Bechtel Topical
Report ***
11 which determines that the criteria and procedure
described in BN-TOP-1, Rev. 1, November 1, 1972 are acceptable for
termination of a Type A test in less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
(5) Surry Unit 2 Type A, B, and C Periodic Test Report transmitted to
Harold R. Denton from W. L. Stewart by letter dated December 13,
1983
(6) Attachment to the licensee's letter of December 13, 1983, (item
(5) above) providing analysis of the as-found Type A leak rate
(7) Surry 2 Technical Specification Section 4.4
(8) Surry FSAR Section 5.5,
11Containment Tests and Inspections
11
(9)
CILRT data
c.
Plant Status
At the time of this inspection Unit 2 was in cold shutdown preparing to
go into a refueling outage. A Type A test is scheduled to be performed
during this outage.
4
d.
Summary of Inspection Findings
Based on review of the above documents and analysis of test data the
inspectors concluded:
(1)
The as-found Type A leakage rate represents a failed test in that
the 1 icensee did not quantify the corrected leakages and
demonstrate that the leak rate would meet the acceptance criteria
of 0.75 La.
(2)
The CILRT was not restarted after isolation of excessive leakage
which caused the Type A leak rate to exceed the allowable leakage
of 0.75 La.
(3)
The acceptance criteria of BN-TOP-1, Rev. 1, were not correctly
applied to determine an acceptable termination time for the short
duration test and to determine an acceptable verification test
time.
(4)
Verification test analysis* was a two point mass determination
which failed to meet the 0.25 La limit when the limit is applied
as a 24-hour allowable limit.
(5)
The origin of isolated leakage from the containment to the
secondary system could not be located and repaired subsequent to
the Type A test.
The potential for leakage through this path has
not been evaluated.
(6)
Certain potential leakage paths were not properly vented or
drained during the Type A test.
(7)
Certain Type C tests are performed in a nonconservative manner.
The above items are discussed in detail in subsequent paragraphs of
this report.
6.
Procedure Review (61719) (61720)
a.
Type A Test Procedure
During the review of the licensee Type A test procedure, PT 16.3, the
inspector noted several penetrations which were not vented or drained
in accordance with Appendix J to 10 CFR 50, paragraph III.A.l.(d) as
amplified in ANSI 56.8, paragraph 3.2.1.5.
Appendix J, paragraph
III.A.1.(d) requires that those portions of fluid systems that
penetrate containment and rupture as a result of a LOCA must be vented
and drained.
ANSI 56.8, paragraph 3.2.1.5 further clarifies the intent
of Appendix J, paragraph III.A.1.(d) in that those lines which are
normally fluid filled and which may be drained or have the fluid driven
5
off by the accident shall be drained to the extent necessary to expose
the containment isolation valve seats to the containment atmosphere.
Contrary to this, penetrations 47, 56A, 56B, 56D, 89, 103 and 104 were
not vented or drained during the September 1983 Unit 2 test.
The
licensee has agreed to vent and drain these penetrations during
subsequent Type A tests.
In addition, the licensee has agreed to
review the Unit 1 1983 Type A test to determine the effect on the final
Type A test results.
Failure to properly vent and drain penetrations for the Type A test is
contrary to the requirements of paragraph II I.A.1. ( d) of Appendix J and
was identified as an example of Violation 281/85-11-01 - Failure to
follow the requirements of Appendix J to 10 CFR 50.
The licensee agreed to review the Type A test procedure (PT-16.3) an
make appropriate changes to the procedure. This matter was identified
for inspector followup as Inspector Followup Item (!FI) 281/85-11-02:
Review revision of PT 16.3 for proper vent and draining requirements.
b.
Type C Test Procedure
The conclusion reached as a result of the review of the licensee's
Type C test procedure was that the downstream method, as performed by
the licensee, was not conservative. It would not measure packing, stem
or flange leakage.
In addition, if the seat leakage was relatively
small, and the down stream volume sufficiently large, the test duration
(nominally 15 minutes) was not long enough to accurately measure seat
leakage.
The licensee agreed to review the downstream test method and evaluate
the use of a more conservative test method.
This is identified as
IF! (281/85-11-03):
Review of downstream test method.
7.
As-Found Type A Leak Rate (61719) (90713)
An analysis of the as-found Type A leakage rate was submitted as an
attachment to the CILRT Report (Reference 5.b.6). Three leakage paths were
identified as greater than 40 standard cubic feet per hour (scfh) which is
the limit of measurement capability for the flowmeters used in the 1983
local leak rate test program. These leakage paths were through penetrations
38, Aerated Drains; 46, Charging to main loops; and, 69, Recirculation Spray
system.
The inspectors reviewed plant drawings and safety classifications
of these systems and agreed with the licensee's contention that under
post-accident conditions penetrations 46 and 69 do not represent probable
leakage paths to the atmosphere .
6
The drain system, penetration 38, does represent a potential air leakage
path post accident.
In that the leakage through this penetration is not
quantified, the as-found Type A leak rate is indeterminate.
Based on
40 scfh leakage from penetration 38, the licensee's analysis shows an
as-found leak rate of 0.047 wt% per day from mass point analysis and 0.079
wt% per day from tota 1-time analysis.
The latter exceeds the a 11 owab 1 e
limit of 0.075 wt% per day.
The inspectors' calculations were in agreement
with the values reported by the licensee.
Based on the analysis discussed above and the failure to quantify the
corrected leakage, the inspectors concluded that the licensee did not
demonstrate that the as-found containment leakage meets the Appendix J
leakage limit of 0.075 wt% per day.
At the exit interview licensee manage-
ment was informed that the September 1983, containment integrated leak rate
on Unit 2 is considered a failure in the as-found condition.
The previous
Unit 2 CILRT in December 1981, was also a failed test in the as-found
condition (IE Inspection Report 50-280/81-34 and 50-281/81.:.34).
In
accordance with Paragraph III.A.6.(b) of Appendix J to 10 CFR 50, .Surry
Unit 2 will be required to perform additional integrated leakage rate tests
at each pl ant shutdown for refueling or approximately every 18 months
whichever occurs first until two consecutive tests meet the acceptance
criteria.
The regulatory basis for the Region II position incorporates the following
requirements.
Surry Unit 2 Technical Specification 4.4 states that the
Type A test wi 11 be performed in accordance with the requirements of
Appendix J to 10 CFR 50.
Appendix J, Paragraph III.A.3.(a)., specifies that
Type A tests wi 11 be conducted in accordance with the provisions of
ANSI N45.4, paragraph 4.2, requires that periodic containment
leak rate tests be conducted before any preparatory repairs are made in order
to disclose the normal state of repair of the containment structure, (i.e.,
as-found leakage rate condition).
It further requires that if the
containment leak rate exceeds the specified maximum, local and integral
tests may be performed and any rrecessary work done to bring the leakage rate
within the specified limits. Then another CILRT is performed to demonstrate
that the maximum allowable leakage rate is not exceeded.
Thus, if leakage
has been corrected as a result of Type Band Type C testing prior to the
Type A test, the amount of corrected local leakage must be applied to the
as-left Type A test result to determine the containment leakage rate prior
to making these repairs.
8.
Test Sequence and Description (61719) (90713)
Pressurization of containment was started at 1620 hours0.0188 days <br />0.45 hours <br />0.00268 weeks <br />6.1641e-4 months <br /> on September 10,
1983. After resolving problems with the air dryers and mechanical chillers,
containment pressure of 61.55 psia was achieved at 1549 hours0.0179 days <br />0.43 hours <br />0.00256 weeks <br />5.893945e-4 months <br /> on
September 11.
Temperature stabilization criterion was met at 2008 hours0.0232 days <br />0.558 hours <br />0.00332 weeks <br />7.64044e-4 months <br />.
In the ensuing 14 hours1.62037e-4 days <br />0.00389 hours <br />2.314815e-5 weeks <br />5.327e-6 months <br />, the leak rate decreased and in the last six hours
appeared to be within the limit.
This trend reversed and over the next
7
seven hours the leak rate increased and indicated a mass loss of about 35
pounds per hour.
The allowable mass loss to meet the 0.75 La limit is about
17.4 pounds per hour.
In the period from 1724 hours0.02 days <br />0.479 hours <br />0.00285 weeks <br />6.55982e-4 months <br /> on September 12 to 1300
hours on September 13 the secondary system out to the steam line non-return
valves was filled with water and the lines in this system isolated by
closing double valves in potential leak paths. Test start time was declared
at 1300 *hours on September 13.
However, leakage isolation proved
ineffective and further isolation manipulations were performed at 1800 hours0.0208 days <br />0.5 hours <br />0.00298 weeks <br />6.849e-4 months <br />
on September 13.
The test was terminated at 0500 hours0.00579 days <br />0.139 hours <br />8.267196e-4 weeks <br />1.9025e-4 months <br /> on September 14.
The licensee provided the following table derived from the test log which
provides a brief summary of the test sequence and events.
SYNOPSIS OF UNIT 2 *
TYPE A TEST - SEPTEMBER 1983
1549 (9/11)
2008 (9/11)
2008 (9/11)-1023 (9/12)
1023 (9/12)-1724 (9/12)
1724 (9/12)-0530 (9/13)
0530 (9/13)-1132 (9/13)
1300 (9/13)
1800 (9/13)
0204 (9/14)
0500 (9/14)
Pressurization was secured.
Temperature stabilization criterion was met.
Average leak rate for time period was 18.8
l bm/hr.
Leak rate increased to 35.4 lbm/hr.
Filled steam lines with water.
Average loss
for this time period was 4 lbm/hr.
Average loss was 33.3 lbm/hr.
At 1000 hours0.0116 days <br />0.278 hours <br />0.00165 weeks <br />3.805e-4 months <br />
started double valving the secondary side in
safeguards.
Start of Type A leak rate period.
NRVs manually torqued closed and open isola-
tion valve to PORV closed.
Removed manometer U0963 from program.
Failed
at 0100 ( 9/14).
Completion of Type A Leakage period (16
hours).
8
9.
Test Performance, Results, and Analysis (61719) (90713)
a.
Test Start Time
Manipulations performed to block containment leakage into the secondary
system were completed and time zero for the test was declared at 1300
hours on September 13.
However, i so 1 ati on of the 1 eakage proved
ineffective.
At 1800 hours0.0208 days <br />0.5 hours <br />0.00298 weeks <br />6.849e-4 months <br /> on September 13, the steam line non-return
valves were retorqued and a manual valve in line with the steam line
power operated relief valve was closed.
These manipulations resulted
in a significant reduction in loss of containment air mass through the
secondary system.
The licensee failed to reset the test start time at
or about 1800 hours0.0208 days <br />0.5 hours <br />0.00298 weeks <br />6.849e-4 months <br /> and did not analyze test data to determine the
impact of this leakage correction on test results.
The official test
time reported to the NRC for the 16-hour test was 1300 hours0.015 days <br />0.361 hours <br />0.00215 weeks <br />4.9465e-4 months <br /> on
September 13 to 0500 hours0.00579 days <br />0.139 hours <br />8.267196e-4 weeks <br />1.9025e-4 months <br /> on September 14.
Results of data analysis
over these 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> are actually results of the composite leak rates
from 1300 hours0.015 days <br />0.361 hours <br />0.00215 weeks <br />4.9465e-4 months <br /> to 1800 hours0.0208 days <br />0.5 hours <br />0.00298 weeks <br />6.849e-4 months <br /> and from 1800 hours0.0208 days <br />0.5 hours <br />0.00298 weeks <br />6.849e-4 months <br />, September 13 to 0500
hours, September 14.
Results of this composite leak rate lead the
licensee to terminate the Type A test in 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> in that the total
time upper confidence limit (UCL) appeared to meet the acceptance limit
of 0.075 wt% per day required for termination of the test in less than
24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
Since the exact time that isolation of the excessive leakage
from the secondary system became effective is not known and to allow
some time for the secondary system to come into equilibrium with the
containment, the inspector selected 1900 hours0.022 days <br />0.528 hours <br />0.00314 weeks <br />7.2295e-4 months <br />, September 13 as a
reasonable starting time for the test. Using 1900 hours0.022 days <br />0.528 hours <br />0.00314 weeks <br />7.2295e-4 months <br /> as time zero,
termination of the Type A test at 0500 hours0.00579 days <br />0.139 hours <br />8.267196e-4 weeks <br />1.9025e-4 months <br />, September 14 yields a 10
hour test.
The total time UCL for this test period was 0.106 wt% per
day which does not meet the acceptance limit for termination of the
test in less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
Based on review of the test log, test data, and discussions with
licensee personnel the inspectors concluded that there was no intent to
select a starting time which would allow early termination of the test.
Selection of 1300 hours0.015 days <br />0.361 hours <br />0.00215 weeks <br />4.9465e-4 months <br />, September 13 as the starting time was a valid
selection based on the expectation that subsequent to filling the
secondary system and double isolating associated systems an acceptable
leak rate would be obtained.
However, failure to restart the Type A
test after further isolation of leakage was necessary at about 1800
hours, September 13 was an error which indicates a lack of attention to
Appendix J requirements.
Paragraph III.A.1.(a) of Appendix J requires that if excessive leakage
is identified which prevents meeting the Type A test acceptance limit,
the Type A test will be terminated, repairs made, and then a Type A
test will be performed.
An NRR staff position issued October 25, 1977,
has determined that under the appropriate conditions, isolation of a
leakage path and subsequent adjustment of the Type A leakage rate based
9
on local leakage measurements is within the intent of this regulation.
Consequently, fa i 1 ure to reset time zero and perform a Type A test
subsequent to the isolation of excessive leakage at 1800 hours0.0208 days <br />0.5 hours <br />0.00298 weeks <br />6.849e-4 months <br />,
September 13, 1983, which prevented meeting the Type A test acceptance
criteria is contrary to the requirements of paragraph III.A.1.(a) of
Appendix J.
This item was identified as an example of Violation
281/85-11-01 - Failure to follow the requirements of Appendix J to 10
CFR 50.
b.
Type A Test Results and Analysis
As discussed above, the licensee analyzed a composit of two distinct
leak rates to determine the containment leakage status.
As a result,
an incorrect containment leakage status was accepted by the licensee
and reported to the NRC.
Further, acceptance of the incorrect leakage
rates lead the licensee to an unauthorized termination of the Type A
test in 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> when the total time UCL did not meet the acceptance
criteria. These errors, which are considered to be a direct result of
using an incorrect test start time, raise a concern as to the actual
leakage status of the containment.
The inspectors performed an
independent analysis of the data to determine the as-left containment
leakage rate.
The composite leakage rate calculated by the licensee and
leakage rates calculated by the inspectors for several time spans are
shown in Table 1 below.
The numbers are in wt.% per day; LSQF is the
leak rate obtained from a least square fit to the data; UCL is the
upper confidence limit leak rate calculated in accordance with ANS 56.8
for mass point analysis and in accordance with BN-TOP-1, Revision 1 for
total time analysis.
'
Mass Point Analysis
1300 - 0500 hours0.00579 days <br />0.139 hours <br />8.267196e-4 weeks <br />1.9025e-4 months <br />
LSQF
UCL
0.023
0.028
1300 - 1800 hours0.0208 days <br />0.5 hours <br />0.00298 weeks <br />6.849e-4 months <br />
LSQF 0.109
UCL 0.129
1800 - 0500 hours0.00579 days <br />0.139 hours <br />8.267196e-4 weeks <br />1.9025e-4 months <br />
LSQF 0.005
UCL
0.011
1900 - 0500 hours0.00579 days <br />0.139 hours <br />8.267196e-4 weeks <br />1.9025e-4 months <br />
LSQF 0.009
UCL 0.015
TABLE 1
Total Time Analysis
0.023
0.0596
0.100
0.159
0.046
0.205
0.020
0.106
10
Although the numbers differ for the latter two calculations, the end
result is the same:
the total time UCL does not meet the acceptance
criteria and the test should have been continued.
The inspectors
considered the most appropriate starting time as 1900 hours0.022 days <br />0.528 hours <br />0.00314 weeks <br />7.2295e-4 months <br />.
To
observe the effect of the starting time the inspectors also calculated
leakage rate results at 0500 hours0.00579 days <br />0.139 hours <br />8.267196e-4 weeks <br />1.9025e-4 months <br /> when the start time was advanced at
20 minute intervals in the period from 1300 hours0.015 days <br />0.361 hours <br />0.00215 weeks <br />4.9465e-4 months <br />, September 13 to 0300
hours, September 14.
These data showed that for 43 calculations the
total time upper confidence limit was met in only 11 instances or about
25% of the possible start times.
Further, the grouping of the 11
successful start times indicates cyclic behavior (see table 2).
A
trend analysis of the total time UCL leakage rate was performed by
calculating the leakage rate for a fixed time span as the time span is
advanced through the test data at preselected intervals. Time spans of
two, four, six and eight hours all confirmed the oscillatory behavior
of the total time UCL.
At the low points of the oscillations the total
time UCL came wtthin the acceptable limit but never stabilized within
the limit.
It was observed that the oscillations were trending
downward and decreasing in frequency.
The total time UCL may have
stabilized within the limit had the test been continued.
TABLE 2
LEAK RATE CALCULATIONS ENDING AT 0500 HOURS SEPTEMBER 14
WITH VARYING START TIMES
MASS POINT
TOTAL TIME
START TIME
ANALYSIS (wt%/day)
ANALYSIS (wt%/day)
Leak
Leak
RATE
UCL
RATE
UCL
1300 - Septemer 13
.023
.028
.023
.06
1320
.022
.027
.02
.061
1340
.020
.026
.017
.071
1400
.019
.024
.028
.099
1420
.017
.023
.013
.082
1440
.016
.021
.027
.127
1500
.014
.020
.018
.135
1520
.013
.018
-.02
.079
1540
.01
.015
-.017
.072
1600
.007
.012
-.013
.064
1620
1640
. 005
.010
-.007
.082 .
1700
.004
.01
-.01
.119
1720
1740
.003
.009
.014
.107
1800
.005
.011
.046
.205
1820
..
11
MASS POINT
TOTAL TIME
START TIME
ANALYSIS (wt%/day)
ANALYSIS (wt%/day)
Leak
Leak
RATE
UCL
RATE
. UCL
1840
.009
.015
.015
.076
1900
.009
.015
...* 020
.106
1920
.009
.016
.009
.098
1940
.009
.016
-.016
.091
2000
.007
.014
-.023
.127
2020
.006
.014
-.001
.139
2040
.007
.016
.052
.263
2100
.011
.019
.004
.043
2120
.011
.019
.005
.054
2140
.010
.020
-.007
.063
2200
.009
.020
.0004
- .053
2220
.010
.021
.0002
.074
2240
.012
.024
.023
.097
2300
.015
.029
.013
.071
2320
.017
.032
.003
.091
2340
.018
.034
.020
.091
2400 (0000) Sept. 14 .022
.040
.029
.132
0020
.026
.047
.011
.179
0040
.028
.051
.060
.220
0100
.037
.062
.056
.209
0120
.039
.069
.076
.349
0140
.038
.075
.011
.162
0200
.016
.051
-.010
.178
0220
.003
.044
-.044
.299
0240
.003
.051
.019
.194
0300
.036
.079
.031
.127
- Indicates tota~ time UCL leakage rates which meet the 0.75 La
requirement for terminating a test short of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
From an overall evaluation of the data, it appears that subsequent to
isolation of the leakage into the secondary system at 1800 hours0.0208 days <br />0.5 hours <br />0.00298 weeks <br />6.849e-4 months <br />
the test data exhibited cyclic behavior around a reasonably low leakage
rate. The cyclic behavior is to some extent averaged by the mass point
analysis.
From Table 2, the mass point analysis leak rate appears to
be stabilizing at about 0.01 wt.% per day with an upper confidence
limit of about 0.02 wt.% per day.
Due to the dependence of total time
analysis on the initial data point, the cyclic behavior causes wide
variations in the leak rate results depending on whether the initial point
is at the top or bottom of a cycle.
For these reasons the inspectors
concluded that, excluding the leakage to the secondary system, the
test data indicate that the containment leakage is within acceptable
limits.
& -
12
c.
Supplemental Test
The verification (supplemental) test was performed using a pump back
method to inject air into containment.
In conjunction with the pump
back technique, the acceptance criteria is based on a two point mass
step change.
Mass pumped into containment {approximately equal to La)
is metered.
The metered value is then compared with the difference
between starting and final mass calculated from the CILRT instrument
system.
The metered and calculated values must agree within +/-0.25 La.
In August 1984, the NRC evaluated the technique of the mass step change
supplemental test and concluded that it does not meet the requirements
of Appendix J to 10 CFR 50, Paragraph III.A.3.(b) (See IE Inspection
Report 50-338/84-29). Since this evaluation occurred subsequent to the
1983 Surry CILRT the test was evaluated only for meeting the required
Appendix J limits. The test results are shown below:
TABLE 3
1.
Metered air pumped into containment
in 1. 66 hours7.638889e-4 days <br />0.0183 hours <br />1.09127e-4 weeks <br />2.5113e-5 months <br />
2.
Difference in starting and final mass as measured
by the CILRT instrument system
3.
Difference in metered and calculated mass (1-2)
4.
Allowable error of +/-0.25 La
(LBM)
597.50
572.62
24.88
+/-138.00
The inspectors identified two deficiencies in the test.
First,
BN-TOP-1, Revision 1, requires that for a test less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, the
supplemental test duration shall be about 1/2 of the Type A test
duration or in this case eight hours.
The mass was pumped into
containment in 1.66 hours7.638889e-4 days <br />0.0183 hours <br />1.09127e-4 weeks <br />2.5113e-5 months <br />.
The second problem relates to the misapplication of the allowable
error.
The tolerance is specified as +/-0.25La.
La is defined as wt.%
per day.
Therefore, the allowable error band in terms of pounds is
+/-0.25 (550 lbs) per day which is +/-138 pounds per day or +/-5.75 pounds
per hour.
The test error of 24.88 pounds reduces to 15 pounds per hour
which does not meet the acceptance limit of 5.75 pounds per hour.
The
Type A test is not considered completed or successful until a success-
ful supplemental test is completed.
Due to misinterpretation of the
Appendix J limits the licensee did not recognize that the supplemental test
was not acceptable; consequently, the reason was not determined,
corrective action was not taken and a successful supplemental test was
not performed.
This condition contributed to the conclusion that the
1983 CILRT was not an acceptable test.
..
.. .
13
Appendix J to 10 CFR 50, Paragraph III.A.3.(b) states that, "If results
are not within +/-0.25La, the reason shall be determined, corrective
action taken, and a successful supplemental test performed.
This item
was identified as an example of Violation 281/85-11-01 - Failure to meet
the requirements of Appendix J.
During discussions of acceptable methods for performing the
supplemental test, the licensee pointed out that Technical Specifi-
cation 4.4 states that the makeup air method will be used for the
supplemental test. The Technical Specification further states that the
leak rate test will be performed in accordance with Appendix J to 10
CFR 50 which recommends that the imposed leak rate method specified in
ANSI N45.4 be used for the supplemental test.
In a case where
regulations conflict with Technical Specifications the regulations
being a higher level document will govern unless the Technical
Specification is identified as an exemption, to the regulation.
The
inspectors concluded that the licensee's Technical Specification will
permit the use of the imposed leak rate method for performing the
supplemental test.
10.
Review Of Leakage To Secondary System (61719) (90713)
In the review of an application for a pl ant, NRC considers the secondary
system a qualified closed system.
As such the steam generators and
associated piping become an extension of the containment boundary.
While it
is recognized that there will be steam pressure in the secondary system
immediately post accident, the system is essentially an unregulated system
and there has been no generic review to qualify this system as a containment
leakage seal system.
Consequently, in the absence of a plant specific
evaluation of the secondary system as a containment leakage seal system
which is acceptable to the NRC staff, leakage from containment into the
secondary system is considered as Type A containment leakage and must be
included in the Type A test leak rate measurement.
Further, Appendix J,
Paragraph III.A.1.(a) requires that if the leakage exceeds acceptable Type A
leakage limits, the Type A test is terminated, corrective action is taken,
and the Type A test is performed.
Relative to this requirement, the NRC
staff issued a position October 25, 1977, which states that isolation of a
leakage path in order to pass the Type A test is within the intent of
Appendix J, Paragraph III.A.1.(a) if the isolated leakage can be measured by
local test methods after the Type A test and is used to appropriately adjust
the Type A test result.
As discussed previously excessive loss of air mass from containment into and
out of the secondary system occurred during the CILRT.
To isolate the
leakage the secondary system was filled with water out to the mainsteam
non-return valves and all piping connected with the _system was double
isolated.
This effected a water seal which reduced the containment leakage
and allowed the licensee to perform the CILRT.
After the CILRT the licensee
attempted to locate the leakage path. Since the secondary system would not
hold air, a hydro was performed at 100 psi.
No leakage was detected.
Further, no steam leaks wer~ detected during a subsequent plant heatup.
14
The inspector reviewed with licensee personnel, the action taken to isolate
the leakage path and tests performed after the Type A test to identify and
repair the leakage path.
In that no leakage could be identified, no adjustment
was made to the Type A test result.
Due to the concern that the leakage
path may tend to open under containment pressure and to seal under secondary
system water pressure, the licensee agreed to take the necessary steps to
perform an air test on secondary system during the current outage before any
maintenance is performed on the system.
The fact that the Type A leakage
exceeded acceptance limits contributed to the decision that the 1983 CILRT
is considered a failed test.
During this outage the Type A test will be
repeated.
If the leakage path is shown to be corrected there will be no
impact on plant operations.
This matter was identified as an unresolved item (UNR) 281/85-11-04 -
Review licensee action to obtain approval of the secondary system as a
containment leakage limiting system or locate and repair source of leakage
into the secondary system.