ML18153C454
| ML18153C454 | |
| Person / Time | |
|---|---|
| Site: | Surry  |
| Issue date: | 10/15/1990 |
| From: | Belisle G, Burnett P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML18153C453 | List: |
| References | |
| 50-280-90-29, 50-281-90-29, NUDOCS 9012040211 | |
| Download: ML18153C454 (8) | |
See also: IR 05000280/1990029
Text
UNITED STATES
NUCLEAR REGULATORY COMMISSION
REGION II
101 MARIETTA STREET, N.W .
ATLANTA, GEORGIA 30323
Report Nos.:
50-280/90-29 and 50-281/90~29
Licensee:* Virginia Electric and Power Company
5000 Dominion Boulevard
Glen Allen, Virginia 23060
Docket Nos.:
50-280 and 50-281
Facility Name:
Surry 1 and 2
License Nos.:
Inspection C~onducted:
0 tober 15-19,
Inspector:~~,,,....,.,"--'=,,...._~~'6-'-~~~,a,..;._--------- -~IP>~
.
. Burnett
oiliSlgned
1990
- .j~--J
.
',,/,,/, *Jc
/' . .f), ,' /
- .--
' / .
Approved by:
LG-f.. t<~ C::::f'/ I L2
--=-G7.,,.. """A ..... -B~e.,,.l ..,.i """si?.=---,.-1~"""C.,...h.,.i e-f-=-------------
T est Programs Section
Engineering Branch
Division of Reactor Safety
SUMMARY
.Scope:
This routine, unannounced inspection addressed the areas of surveillance of
core power distribution and hot channel factors, surveillance and calibration
of nuclear instrume~ts, and thermal po~~r monitoring.
Results:
Hot channel factors were controlled within limits for all cases reviewed, but
in one instance the interval between surveillances exceeded 44 effective full
power days, which may be a violation of the Technical Specifications, but is
currently listed as a unresolved item pending an interpretation of the language
of the specifications. (Paragraph 2.b)
All other surveillance activities reviewed were conducted at the required
frequencies and with acceptable results.
End-of-cycle operations at reduced power and temperature appear to have been
well controlled.
(Paragraph 3.c)
Procedures used for routine survei 11 ance of therma 1 power have not been
compared with the beginning-of-cycle precision heat balance, which is a common
industry practice.
(Paragraph 4.a)
No violations or deviations were identified.
REPORT DETAILS
1.
Persons Contacted
Licensee Employees
- W. R. Benthall, Supervisor, Licensing
R .. M. Berryman, Manager of Nuclear Analysis and Fuel
- R. E. Bilyeu, Licensing Engineer
D. D. Dziadosz, Supervisor of £ore Design
- D.S. Hart, Supervisor, Quality Assurance
- J. W. Henderson, Lead Reactor Engineer
- M. R. Ka~sler, Station Manager
- R. W. Orga, Quality Assurance
- J. A. Price, Assistant Station Manager
- E~ R. Smith, Site Qual_ity Assurance Manager
T. B. Sowers~ Superintendent of Engineering
Other .licensee err.pl oyees contacted included engineers, techni.c i ans, *
security force members, and office personnel.
NRC Resident Inspectors
W, E. Holland, Senior Resident Inspector
S. G. Tingen, Resjdent Inspector
- J. W. York, Resident Inspectof
- Attended exit interview on October 19, 1990. *
Acronyms and initialisms used throughout this report are listed in the
last paragraph.
2.
Surveillance of Core Power Distribution Limits (61702, 61707)
a.
Procedures and Other Licensee Documents
)
l/2-PT-28,2 (Approved February 1, 1990), Reactor Core Flux Maps, is
used to checkout the ~ovable incore detectors, coll~it axial flux
distribution data in specified incore thimbles, record PRNI chamber
c~rrents and AFD meter readings during the flux mapping process, and
to submit the raw data for analysis.
The procedure was basically sound~ but it was noted that the checkout
of the MDs prior to mapping was somewhat superficial.
Each MD was,
in turn, placed at midcore; excitation voltage was decreased until a
reduction in output was noted; then, excitation was increased until
an increased signal, beyond the original, was observed.
Finally the
excitation voltage was set at the midpoint of the two observations.
More common practice is to obtain and plot the relationship between
....
2
voltage and current at regular intervals and to confirm :that a
reasonable plateau exists where current is not strongly dependent on
voltage.
The operating voltage is determined by inspection-of the
plateau.
The plotted plateaus are retained for trending MD
performance.
The licensee appeared receptive* to the inspector's
comments on this supject.
The flux map analysis is performed at an off site computer using the
INCORE computer program. A summary of the INCORE results is provided
in a POWER DISTRIBUTION SUMMARY SHEET, which is prepared from the
INCORE output.
Heat flux and enthalpy rise hot channel factors are
.compared with TS limits.
The NUCLEAR CORE* DESIGN MANUAL, USER'S COPY, PART.VII, CHAPTER I,
FLUX MAP. ANALYSIS (Revision 0, May 1990) (written by the Nuclear
Analysis and Fuel Group of Virgiriia Power) provides methods for both
manual and computer based checks of the validity of the input data
and for review of the output of the INCORE code ~rior to issuan~e of
the POWER DISTRIBUTION SUMMARY SHEET.
Other documents reviewed to evaluate the licensee
I s performance in -
this area included: -
(1)
TECHNICAL REPORT NE-657 (Revision 1), SURRY UNIT 2, CYCLE 10,
DESIGN REPORT.
(2)
TECHNICAL REPORT NE-757 (Revision 0), SURRY UNIT 2, CYCLE 10,
STARTUP PHYSICS TESTS REPORT.
(3)
MEMORANDUM (Dated October 10, 1990) SURRY POWER STATION, CORE
PERFORMANCE CHARACTERISTICS FOR SEPTEMBER 1990~ which applied to
both units.
b.
Surveillance Activities
Review of surveillance records for both units confirmed that accept-
able surveillance intervals and results were maintained throughout
cycle lOA for Unit 1. *
However~ the U~it 2, cycle 10, records revealed that the interval
between surveillances was apparently too long in one iristance.
The
surveillance on July* 18, 1990 was conducted at a core burnup of 8016
MWd/MTU,
and the succeeding sutveillance was condticted on
September 4, 1990 at a burnup of 9525 MWd/MTU~
With 33.8 MWd/MTU
equivalent to 1.0 EFPD, this interval is 44.8 EFPD or 1.44 EFPM.
TS
4.lOB requires that the hot channel.factors of TS 3.12 sha*ll be
determined every EFPM.
TS 4.02 allows a 25% tolerance on
surveillance intervals, or a maximum of 1.25 EFPM, in this case. The
licensee's position is that the language of the specification
requires the surveillance in each full-power month, but does not
limit the interval, which might then be nearly 60 EFPD.
They further
I **
3
claim that the NRC has found _this interpretation .and implementation
of the surveillance requirement satisfactory in the past, *but
provided no documentation of that claim.
This cl ass of power reactor has no* capability for continuous --
m*onitori ng of the i ncore power di stri but ion, unlike a 11 other
- classes.
Only gross power distribution parameters, such as QPTR and
AFD, can be monitored continuously* by the excore PRNis.
This
extension of the surveillance interval does not appear to be prudent.
- Nevertheless, pending an NRC management determination of the inter-
pr~tation of the TS surveillance interval, this item will be treated
as unresolved.
(UNR 50-281/90-29-01:
The interval between
surveillances of hot ch~nnel factors exceeded 1.25 EFPM.)
This item
is similar to UNR 50-280 and 281/90-14-02, which will be addressed
by NRC management.
The inspector also noted that all 50 flux mapping thimbles were
rarely, if ever, used (or available) in performing the flux maps in
either unit. Typically, 38 to 41 thimbles were used in the full core
flux maps. Thirty-eight is the minimum number allowed by TS.
Document (3) contain~d * ~ summary of all of the- re~ctivity anomaly
calculations performed for both units for their current cycles.
The
~urveillance frequencies were satisfied, and the observed ahomalies
were well wiJhin the limits of TS.
c..
F~ture Activities
Discussions with plant personnel revealed that, starting with cycle
12 on both units, core loadings will be designed to reduce*the fast
flux exposure of both the beltline and the longitudinal welds. This
added complexity in the number of fuel material regions will necessi-
tate a change in the computer program used to analyze flux maps.
The
INCORE program wi 11 be replaced by the CE-COR program, and efforts to
qualify the program for use at Surry are currently underway.
No violations or deviations were identified.
3.
Calibration of Nuclear Instrumentation Systems (61705)
a.
Procedures and Other Licensee Documents
NUCLEAR CORE DESIGN MANUAL, USER'S COPY, PART VII, CHAPTER E, Power
Range Detector Calibration Versus Burnup, (Revision 0, May 1990)
(written by the Nuclear Analysis and Fuel Group of Virginia Power)
describes the method used for i ncore-excore nuclear instrument
correlation.
The internally generated computer program INEXC is used
for data analysis and determining instrument setpofnts.
It is
described in TECHNICAL REPORT NE-764, VIRGINIA POWER INCORE/EXCORE
INSTRUMENTATION CALIBRATION CODE MANUAL.
- ,
b.
.4
l/2-PT-28.8 (Approved September 5, * 1989), Power Range Nuclear
Instrumentation Calibration, is performed to collect data for the
incore-excore correlation and to recalibrate the F(delta I) function*
and PRNI channels as defi~ed in TS 4.1. The procedure requires that
a minimum of three flux maps (quarter cbre or full core) b~ obtained
over a range of 5 to 10% -in AFD units *. These specifications are
- truly the minimum to accomplish the correlation.
Better and more
consistent results could be obtained by increasing both the number of
flux map_s and the span in AFD.
This observation was discussed with
the licensee.
Surveillance Activities
Completed incore-excore nuclear instrument calibrations were reviewed*
for Unit 1, cycle lOA; and Unit 2, cycle 10; * In all cases, the
frequency of* test performance was satisfactory and test results
satisfied the acceptance criteria established by the licensee.
The . inspector independently analyzed some of the incore-excore
correlations of PRNI.chamber current with incore axial offset using a
least-squares analysis spreadsheet with the SUPERCALC3 microcomputer
program.
Zero-offset currents * and the slopes of current versus
offset agreed with those obtained by the licensee.
In the inspec-
tor1s analyses, the quality of the correlation was determined from
the calculation of a correlation coefficient.
Correlation coeffi~
cients equal to 0.98 or greater, the maximum value is 1.0, were
accepted as i ndi cati ng that no unaccounted for variables were
i nfl uenc i ng the results.
A 1 imi t of O. 98 on the corre 1 ati or:i .
coefficient is consistent with standard statistical practice.
The licensee's computer program, INEXC, calculates a SEE, rather than
a correlation coefficient, and places an upper limit on the SEE of
0.8.
By perturbing data used in both analyses, the inspector was
able to demonstrate that the SEE was not nearly as rigorous a deter-
mi.nant of the quality of the fit as the correlation coefficient. The
perturbed value of the correlation coefficient dropped to 0.90, but
the SEE i,:icreased only to 0.74.
Thus, in the case _where the
correlation coefficient would identify the results as suspect, the
SEE would.accept the results without question. This observation was
discussed with licensee personnel during the inspection.
c.
Unit 1, Cycle lOA, End-of-Cycle Power Coastdown
For cycle lOA, Unit 1 reached the end of full-power capability (zero
boron and ARO) at about 14,000 MWd/MTU.
To extend tycle burnup to
the design value of 15,900 MWd/MTU, the licensee ~lected to operate
at reduced average cool ant temperature and reduced power.
To
evaluate and support the change in operating characteristics,
EWR-90-251,
RC COASTDOWN SETPOINTS/SURRY /1&2, was initiated and
completed *. The EWR identified and evaluated four setpoints that
would require changing:
(1)
Pressurizer control
(2)
st~am Dumps
5
(3)
Coolant Averag~ Temperature
(4)
High T-average Alarm.
No special procedures were issued for the transition. According to
- engineering personnel, the control room manipulations were considered'
the skill of the craft and the operators received refresher training
on the simulator in controlling the xenon transient that would result
from shifting. the axial power distribution. by reducing coolant
temperature.
The planned 8°F reduction in T-iverage was introduced over about a
day of operation, by a slow and intermittent boration process.
The
colder water in the downcomer reduced the leakage flux to the PRNis,
and several interruptions of the process occurred to perform recali-
brations of the PRNis against the reactor *heat balance, which is
discussed in paragraph 4.
During the process,.AFO ch~nged from -2%
to +9%.
The cited instrument setpoints were changed once after the completion
of the temperature reduction.
In the view of the reactor engineers, who supported the transition,
the entire evolution and the subsequent coastdown went smoothly *. The
inspector did not interview the 6perators to obtain their
perspectives on the evolution.
No violations or deviations _were identified.
4.
Core Thermal Power Evaluation (61706)
a.
Procedure~
l/2-PT-35.0 (Revision 1)~ Re~ctor Power Calibration Using Feed Flow,
is used when the unit computer is in service for data logging, and a
hand.calculation of reactor heat balance based upon feedwater flow is
to be performed.
l/2~PT-35.1 (Revision _1), Reactor Power Calibration Using Fe~d Flow,
is used when the unit computer is not available for data logging, and
a hand calculation of reactor heat balance based upon feedwater flow
is tb be performed.
l/2-PT-35.2 (Revision 1), Reactor Power Calibration Using Steam Flow,
is used when the unit computer is in service for data logging, and a
hand calculation of reactor he~t balance based upon steam flow is to
be performed.
Unlike most facilities, the Surry units have
calibrated steam flow venturis.
That feature is discussed in more
detail in Inspection Report 50-280 and 281/88-29.
I
- I
6
In review1ng the reference ~6pies of the Unit 2 versions of the above
procedures, the inspector noted numerous typographical and fixed data
errors.
For each steam generator loop, in each calculation, there is
a constant datum for line loss and conversion of gauge pressure to
absolute pressure.
Although a var,ation from loop.:..to-loop is
expected, there was also a variation from procedure-to-procedure for
loop C~ which was not present in the other ~wo loops.
By reference
to plant calculational files, the licensee was*able to establish
which of the three values was correct, but could not explain or
justify the errors in the procedures.
One of the errors listed as
typographical included a failure to include all of a precautionary
statement in one procedure.
It appears that bettet effort at proof
reading and peer review is required-when protedures are first issued
or changed.
l/2-PT-35.3 (Revision 1), Reactor Power Calibration Using CALCALC
Computer Program, is the most commonly used of the therma 1 power
surveillance procedures.
Other than to initiate the program, no
hum~n interaction is required to enter data into the program or to
perform the analysis.
It performs power calculations based upon
steam flow, feedwater flow, and primary flow and differential
temperature.
ENG-35,0, Calculating Reactor Power, Delta-T Setpoints, and Reactor
Coolant System Flow, is used at the start ,of an operating cycle, in
conjunction with temporarily installed precision instruments. It has
three purposes:
(1)
To calculate reactor power using steam flow *
. (2)
To calculate 100%-power, delta-temperature. protection and
control values.
(3)
To calculate reactor c,oolant flow by equating primary and
secondary side heat balances.'**
However, this procedur-1= does not require any comparison of .results
with the procedures used to perform routine heat ba 1 ances or set
1 imi ts on how much those procedures may differ in results from the
precision calculation.
This observation was discussed with licensee
personnel during the inspection.
b.
Surveillances Activities
Selected surveillances of PRNI. indication ~ersus heat balance were
reviewed for Unit 2 for the month of October 1990,
ln all cases, the
procedure used was 2-PT-35.3; so there were neither data entries nor
hand calculations to review.
The surveillance frequency exceeded the
TS requirements, and it was clear that the operators performed this,
convenient surveillance whenever they had any question about power
level or the PRNI calibration.
Hence, for every surveillance
reviewed, the agreement between thermal power and PRNI indications
was well within TS limits.
7
ENG-35.0 was last performed on October 6, 1989, for Unit 2~
All of
the test objectives were met, but there were no record~ of routine
heat balances in the test package for comparison of those results*
with the precision results.
The lice,nsee does not have a TS
- requireme-nt to measure primary side. flow, but has been measuring it,
with this* procedure, since the steam generators were replaced.
Licens~e engineers stated that the results were being trended ~nd
that two of the loops were unchanged, but that loop A had exhibited a
- slight downward trend, which might still be within experimental
uncertainty.
The experiences of other facilities, at which hot 1 eg
streaming had complicated or invalidated the measurement of primary
flow, were discusied with licensee enginee~s.
No violations or deviations were identified.
5~
Exit Interview (30703)
The inspection scope and findings were summarized on October 19, 1990,
with those persons indicated in paragraph 1 above.
- The inspector
described. the areas inspected and discussed in detail the inspection
findings ... No dissenting comments were received from the licensee.
Proprietary material was reviewed in the course of this inspection, but is
not included in this report~
UNR 50-281/90-.29-01:
The interval between surveillances of hot channel
factors exceeded 1.25 EFPM - paragraph.2b.
6.
Acronyms and Initialisms Used throughout This.Report
AFD
ARO
dP
EFPM
ENG
EOL
HJR
MWd/MTU
NE
pcm
ppmB
RC
SEE
TS
axial flux difference
all rods out
different1al pressure_
effective full power days
effective full power months
engineering.procedure
end of (core) life
engineering work request
movable detector
megawatt-days per metric tonne of uranium
nuclear engineering
percent millirho, a reactivity unit
parts per million boron_
power range nuclear instruments
periodic test procedure
quadrant power tilt ratio
standard error of the estimate
Technical ~pecifications
unresolve:d