ML18092A681
| ML18092A681 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 07/28/1985 |
| From: | Bettenhausen L, Cheh U, Phelan P, Vito D, Wen P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML18092A679 | List: |
| References | |
| 50-272-85-01, 50-272-85-1, 50-311-85-04, 50-311-85-4, NUDOCS 8508050215 | |
| Download: ML18092A681 (11) | |
See also: IR 05000272/1985001
Text
- ,
U.S. NUCLEAR REGULATORY COMMISSION
REGION I
Report No.
50-272/85-01 and 50-311/85-04
Docket No.
50-272 and 50-311
License No.
DPR-70 and DPR75
Priority ----
Licensee:
Public Service Electric and Gas Company
P.O. Box 236
Hancock's Bridge, New Jersey
Category ----
Facility Name:
Salem Nuclear Generating Station, Units 1 and 2
Inspection At:
Hancocks Bridge, New Jersey
Inspection Conducted:
January 21-25, and February 4-7, 1985
Inspectors:
Engineer
U. Cheh, Reactor Engineer
~a]t~eer
Approved by:
-;;zd~
L. H. Bettenhausen, Chief,
Operations Branch, DRS
Inspection Summary:
dalle
7/.;r /& ~
~date
-Lf./2-S/fls
date
<-t/zy, / EJS
date
date
Inspection on January 21-25 and February 4-7, 1985, (Report No. 50-272/85-01
50-311/85-04
Areas Inspected:
Special,
unannounced
inspection of the
System
leakage measurement
program,
including procedure and test results
review and independent verification using the NRC RCSLK 8 program for PWR RCS
leakage
measurement.
An
evaluation and
independent calculation of core
thermal power was also performed.
The inspection involved 162 hours0.00188 days <br />0.045 hours <br />2.678571e-4 weeks <br />6.1641e-5 months <br /> onsite by
four region-based inspectors .
Results:
One violation was identified relating to procedural inadequacies in
the RCS water inventory balance procedure (SP(0)4.4.6.2(d), Revision 5).
( . 8508050215 850730
I
ADOCK 05000272
G
DETAILS
1.
Persons Contacted
Public Service Electric and Gas Company
W. Bacon, I&C Supervisor
- J. Bailey, Systems Analysis Group
T. Calamito, Engineering Dept. (Newark)
L. Catalfamo, Operating Engineer
J. Davis, Engineering Department (Newark)
- D. Dodson, Licensing Engineer
- L. Fry, Operations Manager
J. Jackson, Technical Engineer
F. Mekulsia, I&C Supervisor
L. Miller, Technical Manager
W. 0
1Brien, Senior Operations Supervisor
- R. Patwell, Licensing Engineer
- D. Perkins, Station QA Engineer
W. Schlegelmilch, Associate Engineer, Operations
- J. Zupko, General Manager, Salem Operations
R. Vanderdecker, I&C Supervisor, Planning
U.S. Nuclear Regulatory Commission
J. Linville, Senior Resident Inspector
- R. Summers, Resident Inspector
- Denotes those present at exit meeting on February 7, 1985.
2.
Reactor Coolant System Leakage Monitoring
2.1
Documents Reviewed
Surveillance Procedure SP(0)4.4.6.2, Unidentifed/Now Identified
Leak Rate Determination, Revision 2, January 6, 1983.
Operating
Instruction
DI
II-1.3.5,
Reactor
Coolant
Leak
Detection, Revision 5, January 13, 1983.
Surveillance Procedure SP(0)4.4.6.2(d), Reactor Coolant System -
Water Inventory Balance, Revision 5, January 13, 1983.
RCS Water Inventory Balance Results for Sa 1 em 1 for November
- 1984 and December 1984.
RC Drain Tank Pump Stop-Start data (from DI II -
1.3.5) for
Salem 1 for November 1984 and December 1984 .
2
RC
Drain Tank level strip charts (from RCDT level recorder
in Sa 1 em 1 contra l room) for November 1984 and December 1984.
Drawing No. 106425, Volume Control Tank, Revision 2.
Drawing No. 106710, Pressurizer Relief Tank, Revision 6.
Drawing No. D-69-161, Reactor Coolant Drain Tank, Revision 9.
Salem Units 1 and 2 System Description SD-R250, Reactor Coolant
Leak Detection System, Revision 0, 7/15/76.
Salem Units 1 and 2 System Description SD-N530, Waste Disposal
- Liquids, Revision 1, October 1973.
NUREG-0986, RCSLK8:
Reactor Coolant System Leak Rate Determina-
tion for PWR
1 s, June 1983.
Salem Generating Station Units 1 and 2 FSAR -
2.2 Scope of Review
Table 5.1-1 (RCS volume)
Table 5.2-4 (Pressurizer, PRT)
Table 9.3-6 (VCT)
Table 11.2-4 (RCDT)
The documents listed above were reviewed for technical adequacy and
for conformance with Technical Specification requirements.
In ad-
dition, independent measurements were performed to verify the ade-
quacy of the licensee 1 s calculational technique.
2.3
Procedure Review
The inspector reviewed the procedures listed in Section 2.1 to deter-
mine their technical accuracy and their ability to assure the proper
detection,
quantification, and identification of Reactor Coolant
System Leakage.
The review revealed several procedural
problems
which are discussed in the following sections.
2.3.1 Leakage Identification
Review of procedure SP(0)4.4.6.2(d), Revision 5, January 13, 1983,
RCS Water Inventory Balance, revealed a leakage value designated by
the licensee as Unclassified Intersystem Leakage.
This leakage iden-
tification was derived from a licensee interpretation of NRC Regula-
tory Guide 1.45, Reactor Coolant Pressure Boundary Leakage Detection
Systems. The licensee determined that RCS
leakage which occurred
outside of the
11
was
not addressed in the
Regulatory Guide
as either identified or unidentified
leakage .
3
Therefore,
the licensee reported unidentified
outside the reactor containment building in a
talled Unclassified Intersystem Leakage.
leakage occurring
separate category
The inspector informed the licensee that other than controlled leak-
age from the Reactor Coolant Pump seals, all RCS leakage shall be
classified as either identified or unidentified.
There can be no
11other category 11 which does not come under either cl ass ifi cation.
The procedural action statement in SP(0)4.4.6.2(d) which corresponded
to an Unclassified Intersystem Leakage of greater than 1 gpm stated
that
11 *** additional efforts should be made to further identify the
source of i ntersystem 1 eakage
11 *
Independent ca 1 cul ati ans performed
by the inspector of RCS Leakage Rates for Salem 1 during the months
of November 1984 and December 1984 (Attachment A) indicated a few
occasions where unidentified leakage would have been slightly greater
than 1 gpm
had Unclassified Intersystem
Leakage
been
included.
Although the licensee did investigate the excessive
11 Unclassified
Intersystem Leakage
11
( 1 etdown 1 i ne va 1 ve -
1CV346) and eventually
reduced it, there was no procedural action statement relating the
leakage to the unidentified leakage classification and its related
Tech. Spec. Limiting Condition for Operation (3.4.6.2).
This pro-
cedural inadequacy, along with others identified later is designated
as a Violation (272/85-01-01).
Since the same procedures are used
for
Salem
Unit
2,
the
violation
applies
to
Unit
2
also
(311/85-04-01).
Prior to the end of the inspection on February 7, 1985, the licensee
had
temporarily
changed
Surveillance
Procedure
SP(0)4.4.6.2(d),
Revision 5, January 13, 1983, to reclassify Unclassified Intersystem
Leakage as unidentified leakage.
A procedural action statement had
also been included to provide for entrance into the RCS Leakage Tech-
nical Specification Action Statement should the unidentified leakage
exceed 1.0 gpm.
The resolution of the other procedural inadequacies
identified by the inspector are being pursued by the licensee as dis-
cussed below.
2.3.2 Measurement of Identified RCS Leakage into the Reactor Coolant Drain
Tank (RCDT)
One of the inputs to the calculation of RCS identified leakage is the
1 eakage into the RCDT.
This 1 eakage comes from the pressurizer re-
1 i ef tank drain, RC Pump #3 seal leakoff, RC Pump #2 seal stand pipes
(head tanks), reactor flange leakoff, and cold leg drains.
The li-
censee measures this leakage using RCDT Pump stop-start information.
The RCDT pump starts when a particular water level (100% level) is
reached in the tank.
Within 2-3 minutes, the RCDT is pumped down to
a minimum water level. The licensee divides the tank volume (from 0
to 100% level) by the time it took to fill the tank to determine a
leakage rate into the RCDT.
The inspector noted that while this
method provided a close approximation of leakage into the RCDT during
an RCS leakage calculation, it did not provide correct results under
some circumstances.
4
To illustrate the procedural deficiency, the inspector noted an ex-
ample discovered in the NRC
independent RCS leakage calculations.
On November 3, 1984, the licensee performed an RCS leak rate calcu-
lation from 1743 to 2109 (3.43 hours4.976852e-4 days <br />0.0119 hours <br />7.109788e-5 weeks <br />1.63615e-5 months <br /> duration).
The gross leak rate
was 2.22 gpm (NRC calculated gross leak rate was identical) and the
measured identified leakage was 1.32 gpm, leaving 0.9 gpm of uniden-
tified leakage.
The RCDT portion of the identified leakage was deter-
mined by the licensee to be 0.63 gpm (from the last calculated RCDT
pump stop-start leakage for a fill period from 0737 to 1207 on
11/3/84).
The RCDT pump stop-start calculated leakage rate for the
next fill period (1209 to 2127 on 11/3/84) was 0.5 gpm.
The inspector
noted that since the latter fill period encompassed the time frame of
the actual RCS leakage calculation, its leakage value should have been
that applied to the i dent ifi ed leakage determination.
Had an RCDT
identified leakage of 0.5 gpm been applied instead o~ 0.63 gpm, the
calculated unidentified leakage would have been 1.03 gpm.
In addition, a review of the Salem 1 control room RCDT level chart
for the period from 1743 to 2109 on 11/3/83 indicates approximately
a 12% increase in RCDT level over the duration of the test.
Using
the RCDT tank curve from the control room (the accuracy of this tank
curve was verified by the inspector via independent calculation), the
12% level increase during the test converts to approximately 0.3 gpm
for the actual test period.
This presents an even more conservative
result.
The licensee acknowledged the inspector 1 s concern and stated that an
attempt would be made to remedy the problem.
Current plans are to
change the RCS Leakage computer calculation in the control room (CRT
39) to include a direct input of RCDT level for the purpose of leak-
age calculation.
Until this can be accomplished, the licensee has
agreed to apply the RCDT pump stop-start leakage for the period during
which the
RCS leakage calculation is performed.
This procedural
inadequacy contributes to the violation (272/85-01-01).
2.3.3 Computer vs. Hand Calculation
The
RCS Water Inventory Balance Procedure (SP(0)4. 4. 6. 2( d)) stated
that either the computer RCS leakage calculation (CRT 39) or a hand
calculation could be used to generate the test result. The inspector
noted that this would allow the licensee to select whichever result
was more favorable.
In addition, the computer-generated values for
RCS average temperature and pressurizer level were more precise than
the hand calculation values which were read from meters on the control
board. The NRC independent calculations (discussed in Section 2.4 of
this report) revealed the leakage rate calculation to be highly sus-
ceptible to variation with a very small Tave change (+/-0.2°F) ac-
companied by a pressurizer level change.
~~~--
5
The licensee acknowledged the inspector's concern and has changed
SP(0)4.4.6.2(d) to state that the computer calculation of RCS leakage
rate is the preferred method.
The hand calculation will be used only
when the computer is unavailable for use.
Also, the licensee has
added a statement which requires a computer calculation after the
computer has been returned to service.
The inspector was satisfied
with the licensee's immediate corrective action in response to this.
concern and !lad no further questions.
This procedure inadequacy is
a contributor to the violation (272/85-01-01).
2.3.4 Calculational Limitations
The licensee's RCS leakage computer (and hand) calculations do not
account for the difference in temperature between water which leaves
the pressurizer ( ~650° F) and water entering and 1 ea vi ng the drain
tanks (RCDT and PRT) and the Volume Control Tank (~100°F).
This temperature correction is accounted for in the NRC RCS leakage
independent calculation computer program (RCSLK8) so that conserva-
tion of mass is maintained in the calculation.
The inspector's in-
dependent calculations for the months of November 1984 and December
1984 indicated a minor problem as a result of the lack of temperature
correction in the licensee's calculations. This problem is discussed
in detail in Section 2.4 of this report .
2.4
NRC Independent RCS Leakage Calculations
The inspector performed independent calculations of the Salem 1 RCS
leak rate for the months of November 1984 -and December 1984 using the
NRC RCSLK8* program.
A comparison of total RCS leak rates, unidenti-
fied RCS leakage, and identified RCS leakage is tabulated in Attach-
ment A to the inspection report.
Prior to performing the independent leak rate calculations, the in-
spector independently calculated the tank curves (slopes) in lbm/%
level for the pressurizer, RCDT, VCT, and PRT for input to the leak-
age calculation.
The inspector used drawings supplied by the li-
censee for the tank dimensions and the formulae in Appendix D of
NUREG-0986 to calculate the tank slopes.
The independent leak rate
calculations indicated that the licensee's tank curves are quite ac-
curate.) This was evidenced by the comparison of the leak rate cal-
culation results where temperature correction had no effect on the
test result.
In these cases, the licensee's and the NRC calculations
were almost identical (+/-0.01 gpm).
After calculating the tank curves, the inspector input this and other
required system information to the RCSLK8 program.
The independent
- NUREG-0986, RCSLK8:
Reactor Coolant System Leak Rate Determination for PWRs,
USNRC, June, 1983.
6
leak rate calculations were made using the same numbers as those used
by the licensee.
In addition, the inspector did calculations using
actual RCDT levels for the beginning and end of the test period (ac-
quired from the control room RCDT level chart) to include the RCDT
portion of the identified leakage in the calculated results.
The
inspector made comparisons of the gross leak rate, identified leakage
and unidentified leakage to determine the calculational accuracy of
the licensee's method.
The comparisons resulted in the following
observations:
1.
With no change in Tave, Pressurizer level, or both during the
test, the Salem and
NRC calculations are almost identical.
2.
When there is a change in Tave with a corresponding pressurizer
level change, the calculated gross leak rate is different.
An
empirical comparison revealed that a Tave change as small as
0.1°F,was enough to effect a gross leakage difference of up to
0.15 gpm.
This difference can be attributed to the lack of tem-
perature correction in the licensee's calculation.
3.
Looking at Item 2 from a procedura 1 acceptance criteria stand-
point, the inspector discovered one instance where, if the tem-
perature-corrected value for gross leak rate had been applied,
unidentified leakage would have been slightly above 1.0 gpm in-
stead of slightly below as
reported by
the licensee.
On
November 22, 1984, the gross RCS leak rate calculated by the
licensee was 1.61 gpm.
The NRC temperature corrected gross leak
rate was 1.75 gpm.
The identified leak rate from both calcu-
lations was approximately 0.65 gpm.
Therefore, the Salem 1 un-
identified leakage was 1.61 - 0.65 = 0.96 gpm.
The unidentified
leakage using the temperature corrected gross leakage would have
been 1.75 - 0.65 = 1.1 gpm.
The
licensee acknowledged the findings of the inspector and
stated that an effort would be made to either account for the
effect of temperature changes in the RCS leak rate calculation
or assure that temperature would not have an effect by including
a procedural requirement that Tave and/or pressurizer level will
be the same at the beginning and end of the test. The inspector
stated that this corrective action should be included in the
licensee's response to the violation (272/85-01-01).
2.5 Conclusions
As a result of the documentation review and the NRC and Salem 1 RCS
leakage test results comparison, the inspector made the following
conclusions:
Of the 41 independent calculations performed, only 3 cases ex-
hibited questionable results due to the inadequacies in the
licensee's procedure (SP(0)4.4.6.2(d)).
r
7
In one case (11/20/84, First Test), where unidentified leakage
was 1.45 gpm, the licensee re~onded appropriately (although not
instructed by the procedure to enter the action statement for
Tech Spec. 3.4.6.2) by performing an immediate retest which re-
sulted in an unidentified leakage of 0.89 gpm.
Leak searches
were initiated by the licensee and a leak in a valve on the
letdown line (1CV346) was discovered and repaired.
The
licensee's
presently
used
leakage
sufficiently accurate if certain procedural
maintained.
These procedural limitations include:
calculation is
1 imitations are
a)
Use of the proper input for the RCDT portion of identified
leakage.
Proper 1 eakage rate to the RCDT can be obtained with the
RCDT pump stop-start records or more accurately by reading
the RCDT 1eve1
change from the contro 1 room RCDT 1eve1
strip chart and converting to a leakage rate.
b)
Temperature Correction.
One way to account for the temperature correction is to
use conservation of mass instead of volume.
The other
method is to include procedural limitations to make sure
that temperature does not affect the test result, e.g.,
requiring that Tave and/or pressurizer level are the same
within prescribed to 1 erances at the beginning and end of
the test period.
c)
Adherence
to
the existing
procedural
instructions for
maintaining steady state conditions during the test.
The 1 i censee stated that efforts wi 11 be made to change
the procedure and the computer program to respond to the
deficiencies identified by the inspector.
3.
Core Thermal Power Evaluation
3.1
Documents Reviewed
Heat Balance Code, Salem Units 1 and 2, Revision 1, August 8,
1984
Salem Unit 1, Thermal Power Calculations -
December 1-31, 1984
I
4.
3.2
8
Scope of Review
The inspector reviewed the documents listed above for technical ade-
quacy and for compliance with Technical Specification requirements.
The inspector also performed an independent calculation to verify the
results generated by the licensee.
3.3
Findings
The inspector developed a simplified model based on the laws of energy,
and heat and mass transfer and made four (4) independent calculations.
These calculations and the'comparable Salem 1 computer calculations
are summarized below:
Date
12/2/84
12/9/84
12/23/84
12/30/84
Plant Computer Thermal Power
(BTU/Hr,
% RTP)
9.481Xl0 9 ,
83.24%
ll.045Xl0 9 , 96.97%
ll.430Xl0 9 , 100.35%
5.386Xl0 9 ,
47.29%
NRC Calculation Difference
(BTU/Hr,
% RTP)
9.479Xl0 9 ,
83.22%
10.909Xl0 9 , 95.78%
ll.422Xl0 9 , 100.28%
5.386Xl0 9 ,
47.29%
0.02%
1.19%
0.07%
0%
The inspector concluded that the licensee's core thermal power calcu-
lation is acceptable and in accordance with Technical Specification
requirements.
The inspector also witnessed the calibration of pro-
cess instrumentation used for the core thermal power calculation. No
unacceptable conditions were identified.
Quality Assurance Involvement
Onsite QA monitoring of normal testing and surveillance activities is per-
formed by the Operations Section of the QA Department.
The inspector veri-
fied this through conversations with the station QA Engineer and by review
of several completed monthly Operation and Maintenance Surveillance schedules.
The inspector concluded that the QA coverage of normal testing and surveil-
lance activities was timely, comprehensive, technically adequate, and ap-
propriately documented.
5.
Tours
The inspector made several tours of various areas of the facility to
observe testing activities, equipment calibration, other work in progress
and general housekeeping.
No unacceptable conditions were identified.
6.
Exit Meeting
A meeting was held on February 7, 1985, to discuss the scope and findings
of the inspection as delineated in this report.
At no time during this
inspection was written
information
provided to the licensee by the
inspector.
ATTACHMENT A
RCS LEAK RATE TEST RESULTS COMPARISON
NOVEMBER 1984 AND DECEMBER 1984
TEST
TOTAL LEAK RATE
IDENTIFIED LEAKAGE
DATE
NRC
SALEM 1
NRC
SALEM 1
NRC
SALEM 1*
11/1/84 1.38
1. 38
0.60
0.60
0.78
0.78
11/2/84 1.37
1. 38
0.88
0.88
0.49
0.50
11/3/84 2.22
2.22
1.19
1. 33
1. 03
0.89
See
NOTE
1
11/4/84 2.53
2.54
2.09
2.10
0.44
0.44
11/6/84 1.84
1. 90
1.63
1. 58
0.21
0.32
See
NOTE
2
11/7/84 1.05
1.12
0.38
0.38
0.67
0.74
11/10/84 0. 69
0.67
0.00
0.00
0.69
0.67
11/11/84 1. 31
1. 31
0.56
0.56
0.81
0.81
11/12/84 1. 46
1.46
0.67
0.67
0.79
0.79
11/18/84 3.94
3.95
3.17
3.22
0. 77
0.73
11/20/84 2.05
2.05
0.60
0.60
1. 45
1.45
See
NOTE
3
11/20/84 1.68
1. 68
0.79
0.79
0.89
0.89
11/21/84 1. 87
1. 75
0.92
0.93
0.95
0.82
11/21/84 1. 65
1. 66
0.69
0.69
0.96
0.97
11/21/84 1. 08
1.08
0.69
0.69
0.39
0.39
11/22/84 1.75
1. 61
0.65
0.65
1.10
0.96 See NOTES 2 & 4
11/23/84 1. 31
1. 31
0.66
0.66
0.65
0.65
11/24/84 1.52
1. 52
0.86
0.86
0.66
0.66
11/25/84 1. 57
1. 58
1. 05
1.05
0.52
0.53
11/26/84 1.48
1. 48
0. 72
0. 72
0.76
0.76
11/26/84 1.58
1. 58
0.65
0.65
0.93
0.93
11/27/84 1.92
1. 78
1.19
1.10
0.73
0.68
See
NOTE
2
11/28/84 1.17
1.18
0.86
0.82
0.31
0.36
11/30/84 1. 44
1. 44
0.44
0.44
1.00
1.00
12/1/84 0.63
0.49
0.75
0.70
-0.12
-0.12
See
NOTE
5
12/2/84 1.24
1. 25
0.86
0.81
0.39
0.44
12/3/84 0.83
1. 25
0.36
0.43
0.47
0.89
12/4/84 0.66
0.88
0.42
0.43
0.24
0.45
12/6/84 1.38
1. 30
0.53
0. 72
0.85
0.58
12/7/84 1.27
1. 27
0.76
0. 77
0.51
0.50
12/10/84 1. 45
1. 60
0.36
0.38
1. 09
1.22
See
NOTE
6
12/11/84 1. 62
1. 60
0.55
0.54
1.07
1. 06
See
NOTE
6
12/12/84 1.33
1. 50
0.47
0.40
0.86
1.10
See
NOTE
6
12/13/84 1.51
1. 50
0.52
0.56
0.99
0.94
See
NOTE
6
12/14/84 1.25
1. 25
0.44
0.37
0.81
0.88
See
NOTE
6
12/15/84 1. 49
1. 50
0.57
0.57
0.92
0.93
See
NOTE
6
12/16/84 1.48
1. 50
0.60
0.55
0.88
0.95
See
NOTE
6
12/17/84 1.48
1. 48
0.55
0.56
0.93
0.92
See
NOTE
6
12/18/84 1.50
1.54
0.43
0.44
1.07
1.10
See
NOTE
6
12/19/84 0.67
0.67
0. 77
0.68
- 0.1
-0.01
See* NOTE
5
12/21/84 0.79
0.85
0.53
0.47
0.26
0.38
f
NOTES
- These values are the unidentified RCS Leakage values which would have been
calculated by the licensee had they employed the proper definition of uniden-
tified leakage.
NOTE 1 -
The NRC value for unidentified leakage (1.03 gpm) was calculated using
a more accurate measurement of the RCDT portion of the identitified
leakage total. This is discussed in Section 2.3.2 of the inspection
report.
NOTE 2 -
The difference in the gross RCS leakage rate is a result of the lack
of temperature correction in the licensee's RCS leakage calculation.
This is discussed in Section 2.3.4 of the inspection report.
NOTE 3 -
The unidentified leakage in this sample illustrates the inadequacy of
the licensee's previous definition of identified leakage.
The li-
censee did, however, perform an immediate retest which generated an
unidentified leakage of 0.89 gpm.
Leak searches were initiated by
the licensee and a leak in a valve on the letdown line (1CV346) was
discovered and repaired.
Si nee the retest was performed within 4
hours of the initial test result, no action statements of Technical
Specification 3.6.4.2 were violated.
This is discussed in Section
2.3.1 of the inspection report.
NOTE 4 -
In this case, the use of temperature correction in the RCS leakage
calculation would have indicated an unidentified leakage of greater
than 1.0 gpm.
This is discussed in Section 2.3.4 of the inspection
report.
NOTE 5 -
The input data for these calculations included large changes in pres-
surizer level (>5%).
Since the RCS leakage calculation is to be cal-
culated under "steady state
11 conditions, and since steady state opera-
tion in a Westinghouse PWR is indicated by little or no change in
pressurizer level, the test results indicate that the test probably
should not be performed if such changes in pressurizer level are tak-
ing place.
Such a procedural precaution was noted in SP (0)4.4.6.2(d)
but was evidently not followed in these cases.
NOTE 6
During this time period (12/10/84 - 12/19/84) the licensee had identi-
fied, via leak searches, leaks ranging from 0.8 to 1.2 gpm.
Identify-
ing these leaks extensively reduces the unidentified leakage values
given in the table .