IR 05000333/1990011
| ML20006E900 | |
| Person / Time | |
|---|---|
| Site: | FitzPatrick  |
| Issue date: | 02/13/1990 |
| From: | Meyer G NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML20006E897 | List: |
| References | |
| 50-333-90-11, NUDOCS 9002260507 | |
| Download: ML20006E900 (10) | |
Text
.
.
.
-.-
.
.
'
)
-
,,
.,o
..
'
. <.
i
.i U. S. NUCLEAR REGULATORY COMMISSION Region I Report No.:
50-333/90-11
.q Docket No.:
50-333
'
License No.':
DPR-59 Licensee:-
New York Power Authority
Post Office Box 41
-
Lycoming, New York 13093
]
Facility:
James A. FitzPatrick Nuclear Power Plant
'
,
.
r Location:
Scriba,. New York
'
' Dates:
January 31 through February 7,1990
!
'
i Inspector:
. Richard A. Plasse, Jr., Resident Inspector i
Reviewer:
Way L. Schmidt, enIor Resident Inspector
!
V 3"I3'TO Approved by:
w-
,
/Gienn W. Meyer, Chief /
.
' ate i
J
'
Reactor Projects Sectibh No. IB
'
,
Inspection Summary:
)
i A special inspection was conducted to review New York Power Authority's hand-
-!
.
L ling 'of feedwater flow measurement and operation in excess of licensed core
thermal power.
Results:
The inspector identified three apparent violations and one unresolved item.
[
An enforcement - conference in Region I is planned to discuss these apparent
'
violations with New York Power Authority. A Table of Contents follows.
.
J, L
L
,
i
?
e 9002260507 900216 PDR ADOCK 05000333
pg
,
.
1
-,
(
L b
.
L -..-
-
.
e<h'
TABLE OF CONTENTS
.Pagg -
1.-
Purpose...................................................
2.
Background................................................
3.
' Chronol ogy o f Events ( 93702)..............................
I 1 -
~4.
Technical I ssues (93702, 71707)...........................
5.
Inspector Fi nding s ( 93702, 71707).........................
6.
Ex i t Inte rvi ew ( 30703)....................................
.
.
l l
!
i l
l
1'
i
,
-
..
.-
._
.-.
.
,
,
.
s
t,:
.
,
.....
DETAILS
'
1.
Purpose
)
The inspector performed this special inspection to review the New York i
Power Authority's (NYPA's) actions with respect to reactor feedwater flow and core thermal power determinations.
'
2.
Background
,
In order for NYPA to operate at full power without exceeding the unit i
operating license limit for core thermal power of 2436 MWth, accurate feed flow measurements are essential.
Differential pressure across the ven-turis in each of the feed lines is measured by a Rosemount 1151 trans-mitter. Electrical circuitry then converts these pressure drops to flow
_
signals.
The process computer combines the two flow signals to form a j
total flow signal. The process computer uses this total flow signal in its heat balance program to determine core thermal power. This calori-l metric calculates core power based on the total flow measurements and the change in enthalpy of the fluids going into (feedwater) and out of (steam)
',
the reactor during steady state ccnditions.
This computed ' reactor power is used to adjust the power level of the reactor to full rated output and to calibrate the nuclear instrumentation (APRMs). The APRMs provided the reactor power inputs to the Reactor Protection System (RPS).
Accordingly,' errors in feedwater flow measurement cause approximately-
.
equal, errors in calculated power.
If feed flow indicates 100% with a transmitter calibration that is 1% low, actual feed flow is 101%. In this case, the unit would actually be operating at 101% of rated thermal power, while the process computer indicates 100% power.
.
The feedwater flow nozzles were purchased from Permutit and installation was completed September 27, 1984.
.
3.
Chronology of Events
1988 October 3 Rosemount feedwater differential pressure transmitters (06-FT-50A and 06-FT-508) were installed by modification.
October 18 NYPA received GE Rapid Information Communication Services Information Letter (RICSIL) No. 30, entitled "Feedwater
-
Flow Element Calibration Error," dated October 4, 1988.
The RICSIL identified an operating BWR/4 (Hope Creek) with a nonconsersative feed flow error of.5%. due to the theimal
.
.
.
_-
-
. -
-
., _ -
.
.
.,
-
-
-
-.
-
-
-
,.
.
i l
'
.
.
.
.
,
.v.
y
>
expansion factor of the nozzle material being inadvertently
omitted from the calculation.
The RICSIL further stated
that the subject BWR/4 voluntarily derated to 98% of rated
!
power while investigating the situation.
This RICSIL was i
assigned to a technical services operating experience
!
review engineer.
October 23 NYPA received Operating Experience (OE) Report 2928, dated
October 12, 1988. This event report identified that PSE&G
operated Hope Creek at greater than 100% core thermal power because of incorrect feedwater flow instrumentation data.
The feedwater differential pressure transmitter (Rosemount
,
1151) did not include a specified correction factor for static pressurization resulting in a nonconservative feed flow indication.
NYPA review was delegated to technical
services and maintenance.
'
December 1 NYPA received GE Services Information Letter (SIL) No. 452, Supplement 1 Category I entitled "Feedwater Element Trans-mitter Calibration," dated November 18, 1988.
The SIL
identified several errors discovered in the calibration of
the feedwater flow element transmitter at an operating BWR/4 (Hope Creek).
The SIL specifically discusses the
'
possible errors, the resultant magnitude of the errors, and
'
the recommended actions to eliminate the errors.
December 12 NYPA delegated review of SIL No. 452, Supplement 1 to the technical services performance engineering group.
Tech-
-
,
nical services sent an informational copy of the SIL to the reactor analyst and the I&C department.
Decembe 15 RICSIL #30 closed out by technical services based on issu-ance of SIL 452 Sup 1.
1989
.
October 5 Reactor was restarted following the 1989 maintenance
.
outage.
,
1:
l.
October 8 Operations noticed that feedwater instruments were reading low during startup.
i October 9 I&C recalibrated feed flow transmitters, when they were found out-of-tolerance low, and zero and span adjustments were performed to place them in tolerance.
l u
l
.
.
--
-
- -
-
-
f.
.
.
...
.
..
October 10 Occurrence Report (OR)89-184 was initiated by operations to document out-of-calibration feedwater flow transmitters during startup.
October 12 Unit achieved indicated 100% thermal power.
Unknown Technical services performance group received copy of Work Request (WR) 06/68756 documenting feedwater problems during startup.
Plant performance snapshot #41 indicated a reduc-tion in overall plant efficiency and a decrease in turbine first stage pressure.
November 5 Reactor scrammed, due to electro-hydraulic control (EHC)
system failure.
November 10 Reactor was restarted.
November 12 Operators noted feed flow indication was low during start-
,
up.
I&C troubleshooting determined that the transmitters were in calibration.
November 12 Reactor scrammed, during SRV testing.
November 13 Reactor was restarted.
Power was limited to approximately 20% for EHC troubleshooting.
.
,
November 14 I&C determined that the transmitter settings had not been calculated to account for the static pressurization effect.
I&C completed the transmitter calibration following the calculation taking into account static. pressurization in accordance with the Rosemount technical manual.
November 17 NYPA requested feed flow nozzle curve basis from supplier (Permutit).-
November 19 Reactor was shut down to correct EHC difficulties.
November 21 Technical services issued Performance Memo 89-52.
This memo discussed the static pressurization effect and other possible errors described in SIL 452, Supplement 1.
This evaluation stated that the other errors required additional information. The information needed was on the design of the nozzles.
Based on this information a calculation was needed to determine the new calibration data.
NYPA concluded that the static pressurization effect had been corrected, and was the major challenge to core over-power associated with feedwater flow as discussed in SIL 452, Supplement 1.
NYPA further concluded that the other errors identified in the SIL would not represent restric-tions on startup.
>
f; T~~
...
- .3
\\
.
m,-
,
i I
j
' November 22 Reactor restarted.
i
'
November 25 Plant reached 100% power; generator output decreased 1.7%.
{
Technical services evaluation determined that 1% was due to
'
static pressure adjustment and the remaining.7% was due to l
2 leaks (turbine bypass valves).
)
,
December 4 NYPA informed by Permutit that ASME ideal curves could be used for nozzle calculations.
December 8 OR-b9-240, issued by technical services, stated that the feed flow transmitters had not previously been calibrated with - the adjustment for static pressurization.
Further
,
stated is that this oversight resulted in underpredicting feed flow.5'to 1.5%, which would lead to core overpower in i
excess of the licensed thermal power. NYPA did not make a
,
10 CFR 50.72 report on this event, based on the conflicting
'
indication of power from the generator output.
Unknown NYPA decided to purchase two new Rosemont 1151 transmitters and to have them calibrated in the factory. The calibra-
.
tion was to inc_lude the static pressure effect and the other effects documented in the SIL.
Then these trans-mitters would be installed into the system.
.
December 12 NYPA contractor verified technical services methodology for-
>
C determining new calibration settings for the unaddressed
,
SIL items.
December 21 Technical services completed ' calculation (JAF 89-045) of new calibration settings per the SIL.
The engineer who performed this calculation initially believed that these
<
i
'
errors were in the conservative direction.
1990 L
January 4 The technical services superintendent reviewed and approved JAF 89-045.
This calculation indicated that with the transmitters calibrated at the revised settings the actual feed flow - would potentially indicate 0.5% nonconserva-tively.
This data was provided by NYPA to Rosemount to allow calibration of the new detectors.
>
l l
,,
.. _
--r
-- -.--
- - -
-
['
l q
.-
,
o
.
,
'
l January 5 The_ superintendent of power discussed the new calibration data with technical services and I&C based on calculation u
89-045. The.5% nonconservative effect was also discussed.
{
NYPA concluded that this.5% would be offset when the
!
Rosemount-calibrated transmitters were installed.
!
January 25 NYPA received the vendor calibrated Rosemount 1151 trans-mitters and performed a bench test to verify required
tolerances.
!
l January 29-NYPA installed the new Rosemount transmitters.
When the
.I detectors were placed in service NYPA found that core
thermal power was at 2453 MWth (approximately.7% above the i
licensed power level).
NYPA reduced power to below the-
.!
license limit and issued OR 90-22 documenting the over-
'
power.
NYPA made a 10 CFR 50.72 notification of this
-
instance.
January 30 NYPA received feed flow nozzle curve basis from Permutit.
i This information validated calculation [[::JAF-89-045|JAF-89-045]].
4.
Technical Issues The inspector determined that GE SIL 452, Supplement 1 identified several
errors in calibration of feed flow transmitters used to determine core
'
thermal power.
These issues were-
.
l a.
Venturi calibration and full span pressure drop. These errors deal
'
with the testing performed by the nozzle supplier to determine the
'
L-flow characteristics.of the actual nozzle for the different pressure measuring location (taps) and the correlation of this data to oper-ating conditions.
The venturi calibration error results from cali-brating transmitters using data from a tap diffetent from the one
.
actually being used.- GE stated that this could have a +/- 0.2%
'
!
effect on feedwater flow measurements. The full span pressure drop resulted because the nozzles were calibrated with water at 68 F.
The
'
effect of water at operating pressure and temperature had to be accounted for.
This factor also accounted for the thermal expansion of the nozzle, from the tested ' configuration to operating tempera-
,-
ture. GE stated that the water temperature effect resulted in about L
0.5% effect and the thermal expansion in about 0.1% effect, both non-l conservative, in feedwater flow measurements.
'
.
.
L
,
.
--
.
--
.
--
-
- -..
t
' '
.6 The inspector reviewed calculation JAF 89-045 and found that it ade-quately addressed these issues.
This calculation was. based on Appendix A of SIL 452, Supplement 1.
While performing this calcula-tion of calibration settings NYPA determined that several of the potential deficiencies pertained to FitzPatrick.
NYPA determined that the original specific volume of the feedwater at rated pressure and temperature was in error.
The specific volume used density at saturated conditions based on the water temperature rather than the conditions for both temperature and pressure upstream of the flow nozzles.
Further the geometry ratio of the nozzle throat to the diameter was also in error.
The net effect of these two issues i
resulted in an approximate, calculated nonconservative 0.44% effect in feedwater flow measurements, b.
Flow transmitter calibration (static pressurization).
This effect accounts for shifts in the transmitter output when the transmitter operates between atmospheric and operating pressure. This occurred because of the design of the Rosemount 1151 transmitters. GE stated that this could lead to about a nonconservative 0.5% effect on measured feed flow.
The inspector determined that NYPA did not properly perform the initial calibration of the Rosemount 1151 detector after installa-tion, since the static pressure effect was discussed in the trans-mitter technical manual. This effect was compensated for when NYPA calculated the proper settings and performed the static pressure calibration on November 14, 1989, c.
Process computer flow measurement. This effect deals with the design tolerance of a resistor used by the process computer to convert the feed flow transmitter current signal to a process computer voltage signal.
GE stated that resistance variations could lead to a +/-
0.25% effect on feedwater flow.
.
L The inspector determined that to eliminate this error the SIL recom-mends a loop calibration from the transmitter to the process com-puter. NYPA had not performed this calibration. The inspector con-
'
sidered this item unresolved pending NYPA completion of the process computer loop calibration.
(UNR90-11-01).
d.
Process computer feedwater temperature compensation.
A temperature l
l value is necessary to determine the enthalpy of the feedwater enter-ing the reactor vessel.
GE states that this effect is minor and conservative resulting in conservative 0.04% error in measured feed
[
flow.
i l-The ir,spector determined that NYPA had taken acceptable action to reduce this error.
I
L
!
f (
e
.
.
.
.-
..
.A
._
.
' NYPA continued to refine their transmitter setting calculation as addi-
tional data was received.
Further, NYPA believed that feed flow is
'
actually lower than was being indicated, based on reduced turbine gener-ator electrical output.
NYPA stated that they plan to perform a radio-active element trace flow test in the future to determine actual flow.
Based on this determination NYPA stated that they would develop a course
of action to address potential nozzle problems.
'
5.
Inspector Findings
.
'
a.
The inspector identified three examples.in which NYPA failed to take timely and effective coerective actions to prevent exceeding their licensed core thermal power.
1.
NYPA did not take timely corrective actions to evaluate the information provide by OE 2928 and GE SIL 452, Supplement 1.
The inspector determined that NYPA had adequate information on December 12, 1988, to determine that the core thermal pewer was being exceeded.
2.
NYPA failed to take effective corrective action based on tech-
.
nical' service Performance Memorandum 89-52. NYPA did not deter-
!
mine the summation of all the other potential nonconservative
!
errors discussed in Sll-452, Supplement I and did not take appropriate action to limit core therma 1 power while these effects were evaluated.
'
The inspector found that NYPA's conclusion that "since compensa-tion and static pressurization have been corrected, the major challenges to core overpower have been corrected. was not cor-rect nor conservative. Technical services Performance Memoran-
.
dum 89-52 documented NYPA's initial evaluation of SIL 452,
!
Supplement 1.
m
-
r 3.
NYPA did not take effective action when calculation JAF 89-045 indicated - that the unit was operating at greater than rated thermal power.
This calculation was completed on December 21 but not approved until two weeks later. On January 5 the per-formance group and the I&C department determined the new cali-bration. settings for the Rosemount transmitter would result in a change in indicated feedwater flow of approximately.5% in the nonconservative direction.
This determination was discussed with the superintendent of power. NYPA failed to take appropri-ate action to prevent an overpower condition once calculation [[::JAF-89-045|JAF-89-045]] was completed, u
-
f I
i '
,;
,
'..
,
,
f
~~
l l
The inspector concluded that NYPA management did not take timely and
!
effective corrective action - to prevent reactor operation over the
,
licensed rated thermal power. The failure to take such timely action
!
is an apparent violation of 10 CFR 50, Appendix B, Criterion XVI,
,
Corrective Action.
(VIO 90-11-02)
.
b.
NYPA failed to adequately perform design modification M1-87-119 which
installed the Rosemount 1151 feed flow instruments. -This modifica-tion referenced the need to update the feedwater flow instrument I
calibration. settings based on Rosemount Technical Manual 4256/4257.
l The inspector found that pages 1-7 and 1-8 of this manual specified that the calibration settings be adjusted to compensate for static
'
pressure.
The inspector found that NYPA's failure to adequately calculate the settings for the Rosemount 1151 transmitter was an apparent violation of 10 CFR 50, Appendix B, Criterion III, which requires that design
.,
control measures ensure that the design basis for components be cor-
.
rectly translated into instructions (i.e.,
calibration settings).
'
(VIO 90-11-03)
'
c.
NYPA operated the reactor at a power level above the ' licensed core.
thermal power.
Errors in the calibration of the feedwater flow transmitters resulted in the following:
approximately 1.7% core overpower from November 1988 until November,1989, when the static pressurization error was corrected; and approximately.7% core over-
-
power from November 25, 1989 untti January 29, 1990, when the recently calibrated Rosemount transmitters were installed.
The' inspector found that operating the unit above rated core thermal
'
power was an t.pparent violation of Facility Operating License DPR-59, Section.2.c.(1),
(VIO 90-11-04)
6.
Exit Interview
-
The inspector held meetings with NYPA plant management during the inspec-tion to discuss his findings and the inspection scope. The inspector held
-
an exit interview with the resident manager and his staff at the conclus-
ion of the period to discuss the sequence of events and the inspector findings.
Based on the FRC Region I review of trils report and discussions held with
,
L NYPA representatives during the exit meeting, this report does not contain
,
information subject to 19 CFR 2.790 restrictions.
'
i I
k
.,
- - -
- - -
.. - -
m
-