IR 05000250/1989032
| ML17347B308 | |
| Person / Time | |
|---|---|
| Site: | Turkey Point  |
| Issue date: | 08/21/1989 |
| From: | Belisle G, Burnett P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML17347B307 | List: |
| References | |
| 50-250-89-32, 50-251-89-32, NUDOCS 8909080135 | |
| Download: ML17347B308 (21) | |
Text
~PP AE0I
~d'Ip0 Wp*~4 UNITED STATES NUCLEAR REGULATORY COMMISSION
REGION II
101 MARIETTASTREET, N.W.
ATLANTA,GEORGIA 30323 Report Nos.:
50-250/89-32 and 50-251/89-32 Licensee:
Florida Power and Light Company 9250 West Flagler Street Miami, FL 33102 Docket Nos.:
50-250 and 50-251 Facility Name:
Turkey Point 3 and
Inspection Conducted:
Ju 26 - 30, 1989 License Nos.:
Approved by:
P.
.
urne
/
/
~. /'
.:l=7-I i l'.
A.
elisle, Chic Test Programs Section Engineering Branch Division of Reactor Sa ety Date Signe
~///~
Datk S gned SUMMARY Scope:
This routine, unannounced inspection addressed the areas of post-refueling startup tests for Unit 4 and core performance monitoring, nuclear instrument calibrations, and thermal power monitoring for Unit 3.
Results:
The Uni,t 4, cycle 12 initial criticality was per ormed in a conservative, well-controlled manner.
Five potential improvements to the procedure used were identified.
(Paragraph 2.b)
All zero power physics tests met the numerical acceptance criteria, and the basic test methods were good and yielded convincing results.
The tests could have been improved by better annotation of reactivity computer chart traces, independent evaluation of test results, and by addition of an acceptance criterion for internal agreement of among ITC measurements.
The other accep-tance criteria invoked were consistent with ANSI/ANS-19.6-1985, Reload Startup Physics Requirements for Pressurized Water Reactors.
(Paragraph 2.c)
The plant reactor engineering staff response to NRC initiatives has been excellent.
Past observations in NRC inspection reports on the performance of zero power physics tests both at Turkey Point and St. Lucie have been incorpo-rated into the current test procedures.
Response at the corporate level to NRC initiatives has been poor.
The plant was made aware of the excessive post-trip cooldown and concomitant reduction in shutdown margin at Sequoyah
~H'QI.I'VI.)I-sg)j '=l=g PDR I-"IDIICK i>~
o5'vJt ~
LI~<<?'0 I'.I~~hI,I
nearly a year ago.
Corporate Fuel Resources was requested to provide analysis and guidance for Turkey Poi,nt, and is now four months overdue in its response.
The plant has yet to demand action, but plant management agreed to pursue the issue.
Neither unit will have much extra shutdown margin at the end of its current cycle.
(Paragraph 2.d)
The Unit 3 surveillances of hot channel and hot spot factors, quadrant power tilt, and reactivity anomaly were conducted at acceptable frequencies and with acceptable results.
The calibration of the power range nuclear instruments (PRNI) against the incore axial power distribution, the incore-excore correla-tion, was performed with acceptable frequency.
However., the inspector's independent evaluation of the data did not yield the same results as the licensee's analysis, which is performed by a leased computer program.
The plant staff had no knowledge of the inner workings of the program.
Further-more, the program did not provi de sufficient output, for example, a correla-tion coefficient, to confirm that an acceptable correlation had been obtained.
At the exit interview management made a commitment to establish an acceptance criterion for the test that would require a correlation coefficient of at least 0.98.
Subsequent to the exit interview, when the difference between the inspector and licensee calculations could not be readily resolved, the licen-see was informed that their justification of their results would be an unre-solved item for them to address.
(Paragraph 3)
Routine thermal power calculations and surveillances are performed on a
new plant computer, ERDAS.
Logged data were extracted from ERDAS for independent evaluation by use of the NRC program TPDMR2.
Relative to experiences at other facilities, collecting the data was a most difficult process.
A sufficient number of parameters can not be accessed and printed out on a repetitive basis without excessive manual intervention.
This computer can not be the diagnos-tic engineering tool at Turkey Point that similar computers are at other facilities.
The operator interface was not inspected.
The independent analysis of the thermal power data using TPDMR2 confirmed that the power calculation in ERDAS is adequate for surveillance of the licens'e limit.,
(Paragraph 4)
No violations or deviations were identifie REPORT DETAILS 1;
Persons Contacted Licensee Employees
~L.
W. Bladow,-Plant guality Assurance Superintendent
- J.
E. Cross, Plant Manager-Nuclear
- T. A. Dillard, Maintenance Manager A.
R. Dyches, Reactor Engineer
- R. J. Larl, guality Control Supervisor
- R. J. Gianfrencesco, Maintenance Superintendent
- K. N. Harris, Site Vice President
~J.
P. Hendrickson, Reactor Engineer
- D. W. Herrin, Regulatory Compliance Engineer C. A. Lenhart, Reactor Engineering Computer Coordinator
- E. L. Lyons, Regulatory Compliance Supervisor, Acting
- G. L. Marsh, Reactor Engineering Supervisor
- L. W. Pearce, Operations Superintendent
"A.
T.. Zielonka, Site Engineering Supervisor Other licensee employees contacted included engineers,'perators, and office personnel.
NRC Resident Inspectors R.
C. Butcher, Senior Resident Inspector
- T. F. McElhinney, Resident Inspector
- G. A. Schnebli, Resident Inspector
- Attended exit interview on June 30, 1989.
Acronyms and initialisms used throughout this report are listed in the last paragraph.
2.
Post-Refueling Startup Tests of Unit 4 (72700, 61708, 61710)
a.
Introducti on Startup tes'ting of Unit 4 for operating cycle 12 began on May 19, 1989.
In addition to the approved procedures discussed below., the following documents were pertinent to the tests, and were reviewed by the inspector:
( 1)
Reload Safety Evaluation, Turkey Point Plant Unit 4 Cycle 12, Westinghouse, December 1988;
e (2)
'WCAP-12010, The Nuclear Design and Core Management of the Turkey Point 4 Nuclear Power Plant, Cycle 12, Westinghouse (Proprietary),
January 1989 (This is the source of the predict-ed values of the startup test measurements.);
and, (3)
Digital Reactivity Computer User's Manual, Westinghouse.
b.
Initial Criticality Initial criticality for cycle 12 was controlled by operating proce-dure 0204.3 (March 24, 1989), Initial Criticality after Refueling, which was completed, on May 20, 1989 and approved by management on June 6, 1989.
The inspector's review of the completed procedure confirmed that criticality had been reached in a cautious and well-controlled manner.
The SRMs that guided the process had been demonstrated to be operating reliably by use of statistical tests.
However in plotting ICRR to obtain continuing predictions of the final critical confi guration, only one SRM (N31) response was plotted instead of using both.
One IRM was plotted, but provided no information; since neither IRM came on scale until after criticali-ty.
The inspector plotted the response of the second SRM (N32) and confirmed a response that paralleled that from N31.
The procedure required that ICRRs be plotted first against rod position as rods were withdrawn and then against. dilution water added once that part of the process began.
The common practice of plotting ICRR against C
was not implemented by the procedure.
However, C
was measured'eriodically during the test, and the plot could have been performed by the test personnel.
Other activities controlled under this procedure included determina-tion of the flux level at which nuclear heating was detectable by a change in average RCS temperature.
The upper limit for low power testing was then set below that flux level to assure that measure-ments using the reactivity computer were not compromised by the fuel doppler effect.
Finally, the reactivity computer was checked out by comparing solu-tions from the computer with those obtained by measuring the reac'tor period and manually solving the i nhour equation.
No acceptance criterion was specified for this test, but the actual performance satisfied the usually invoked criterion of 4~ agreement between the manual and computer solutions.
The following improvements to the initial criticality procedure were discussed with and are being considered by. the licensee for incorpo-ration in the procedure:
( 1)
Increase the number of counts per observation to at least 1000 for both Chi-squared tests and ICRR determinatio (2)
Add a plot of ICRR against boron concentration during dilution to the c'urrent practice of.plotting ICRR against dilution water added.
(3)
(4)
Plot the response of both SRMs instead of just one =during rod withdrawal and RCS dilution.
Formalize the confirmation of acceptable overlap between SRMs and IRMs.
(5)
Add acceptance criteria for the agreement between digital reactivity computer solutions and inhour equation solutions.
Low Power Physics Tests Operating procedure 0204.5 (March 24, 1989), Nuclear Design Check Tests during Startup after Refueling, was used to guide the remain-der of 'the startup test activities.
The following appendices to the procedure provided the detailed test instructions for specific measurements:
( I)
APPENDIX A, Boron Endpoint Measurement, was used to establish the ARO CB at HZP.
The result of 1538 ppmB was in acceptable (+50ppmB) agreement with the predicted value of 1572 ppmB.
)
(2)
APPENDIX B,'sothermal Temperature Coefficient, was performed for the ARO configuration.
The measurements were done well in that each temperature swing was at least 5'F and the difference between heatup and cooldown measurements was much less than
pcm/'F.
The average ITC was -0.88 pcm/'F.
The corresponding MTC was +0.92 pcm/'F, which was acceptably less than the TS 3. 1.2. la limit of 5 pcm/"F.
However, the procedure does not include an acceptance criterion specifying the agreement among the measurements used in computing the average result.
Common industry practice is to limit the span of accepted results to
pcm/'F.
(3)
APPENDIX F, Rod Worth, was performed on control bank C from the ARO condition by initiating a slow dilution of the RCS and periodically inserting the bank to compensate for the reactivi-ty increase.
The inspector independently analyzed the reactiv-ityy computer chart traces to determine the differential worth distribution and the integral worth of the bank.
It had been the responsibility of the test personnel to annotate the traces with rod position as the traces were drawn.
The inspector found that interpreting the traces was made difficult because of sparse and sometimes ambiguous annotation.
The inspector's differential worth curve is plotted with that obtained by the licensee in Attachment I.
Two points of significant difference were referred to the licensee for resolution.
The inspector obtained an integral worth of 1305 pcm; the licensee's result
i
was 1329 pcm; and both were within 105 of the predicted worth of 1272 pcm; thus satisfying the acceptance criterion.
The procedure does not require that a second, qualified person make an independent evaluation of the reactivity computer traces.
That is a
common practice within the industry.
(4)
APPENDIX D, Rod Worth by Rod Swap, was performed after APPENDIX F,
and used the results of the latter to determine the worth of
, all of the other rod banks.
For all of the rod banks, the individual differences between measurement and prediction ranged from -7.3X to +2.7X.
~ The measured total worth was 5861 pcm, which was 3.5X less than the predicted total worth of 6072 pcm.
The acceptance criterion was
+10% agreement.
(5)
Data Sheet 10, Differential Boron Worth, was used with data from the appendices to calculate a differential boron worth of 9.6 pcm/ppmB, which was the same as the predicted value.
Except as noted above; all test results satisfied the acceptance criteria specified in ANSI/ANS-19.6-1985, Reload Startup Physics Requirements for Pressurized Water Reactors.
d.
Response to NRC Initiatives Affecting Core Performance and Testing The plant reactor engineering staff response to NRC initiatives has been excellent.
Past observations in NRC inspection reports on the performance of zero power physics tests both at Turkey Point and St.
Lucie have been incorporated into the current test procedures.
Response at the corporate level to NRC initiatives has been poor.
The plant was made aware of the excessive post-trip cooldown and concomitant reduction in SDM at Sequoyah nearly a year ago.
Corpo-rate Fuel Resources was requested to provide analysis and guidance for Turkey Point, and is now four months overdue in their response.
The plant has yet to demand action, but plant management agreed to
- pursue the issue.
Neither unit will have much extra SDM at the end of its current cycle.
Current operating procedures do not provide guidance on limiting and responding to excessive cooldown following a reactor trip.
No violations or devi ations were identified in the performance of the initial criticality and zero-power test procedures.
3.
Unit 3 Core Performanc'e Monitoring and Nuclear Instrument Calibrations (61702, 61705)
Full power operation of Unit 3 for cycle 11 started in January 1988.
The surveillance procedures discussed below were reviewed for performance from cycle startup to the date of the inspections
e a
~
b.
c ~
OP12404.1, System'and Power Distribution Surveillance, had been performed at or near full power using full-core flux maps on
occasions.
The intervals between maps were all less than
EFPD and at least 75K of the installed instrument thimbles were used in each map.
The measured values of hot channel enthalpy rise, hot spot factor, and gPTR all satisfied technical specifications in each map.
OP1004.3, Reactivity Deviation from Design Calculations, had been performed 16 times at intervals of less than 31 EFPD.
The maximum deviations from design calculatedranged from -259 pcm to +528 pcm and satisfied the acceptance criterion of +1000 pcm'.
OP12304.4, Power Range Nuclear Instrumentation Channel Check.and Calibration, had been "performed four times during the cycle.
The interval between measurements did not exceed
EFPD.
The procedure determines the correlation between incore and excore measured axial power distributions and calibrates the instrument channels associat-ed with each excore chamber to produce a uniform voltage output as a
function of AFD.
The most recent measurements were performed on March 10, 1989 using the data from four quarter-core flux maps with measured incore axial offsets of -8.6 to + 1.5%.
The inspector independently analyzed the correlation between axial offset and the currents from the individual chambers of the PRNIs using a
least-squares spreadsheet with the microcomputer program SUPERCALC3.
Seven of the chambers gave good results with correlation coeffi-cients greater than 0.98.
The eighth, N43-bottom, had a correlation coefficient of 0.89, which, in the inspector's judgement, indicates either poor data or marginal performance of the channel.
That judgement is based upon performing similar analyses at many other facilities.
The procedure has no acceptance criterion for the correlation coefficient, and that information is not provided in the output of the computer program EXCAL'Revision 2, June 13, 1985),
which is used by the licensee to analyze-the raw chamber c'ur rent and axial offset data.
EXCAL is accessed through a remote computing service, and there appears to be little onsite knowledge or documen-tation of the calculations and error analysis performed by it.
At the exit interview, the licensee made a commitment to add= an accep-tance criterion to the procedure to require a correlation coeffi-cient of at least 0.98.
(Inspector followup item 50-250 and 50-251/89-3.".-01)
Subsequent to the exit interview, the inspector continued to review procedure 12304.4 and the EXCAL output.
The independent analyses discussed above yielded chamber currents and channel voltages at zero axial offset that were in good agreement with the EYCAL output.
However, the EXCAL-predicted channel voltages at non-zero offsets could not be duplicated.
The magnitudes of the differences are shown graphically in Attachment 2.
Prior to leaving the site, the inspector informed the licensee that EXCAL channel voltage dete'rmi-nations must be justified to resolve the acceptability of the
calculation, and that*the issue would be tracked as an unresolved item.
(UNR 50-250 and 50-251/89-32-02)
No violations or deviations were identified in the, review of these sur-veil lance activities.
Thermal Power Monitoring (61706)
The licensee has installed a new plant computer, ERDAS, which is used to do thermal power calculations and surveillances, among other activities.
With the expert help of a licensee computer engineer, logged data were extracted from ERDAS for independent evaluation by use of the NRC program TPDHR2.
(See NUREG-1167, TPDWR2: Thermal Power Determination for West-inghouse Reactors, Version 2, October 1985.)
Relative to experience at other facilities, collecting the data was a most difficult process.
A sufficient number of parameters could not be accessed and printed out on a repetitive basis without excessive manual interventi'on.
This computer can not be used as a diagnostic engineering tool at Turkey Point with the ease plant computers provide at other facilities.
The operator interface with ERDAS was not inspected.
The independent analysis of the thermal power data was completed in the Regional o'ffice following the onsite portion of the inspection.
As is common in such inspections, the raw data required considerable manipula-tion and adjustment before they could be used in TPDWR2.
The plant parameters used to customize TPDWR2 for use at Turkey Point 3 and the data used in the calculation are given in Attachment 3.
Results of the TPDWR2 calculation are given in Attachment 4.
The agreement between the ERDAS and TPDWR2 calculations was within 0;2A and was acceptable, al-though the calculations models used were quite different.
The ERDAS model is of a single reactor'oolant loop and steam generator using averaged values of all of the loop-specific parameters.
In TPDWR2, each loop is analyzed separately.
The good agreement between the two methods is probably a result of the near-identical feedwater flows in each loop.
If the feedwater flows should start to differ, possibly as the result of differences in steam generator tube plugging, the averaged model used in ERDAS might not be as acceptable.
No violations or deviations were identified.
Exit Interview (30703)
The inspection scope and findings were summarized on June 30, 1989, with those persons indicated in paragraph I above.
The inspector described
'he areas inspected and discussed in detail the inspection findings.
No dissenting comments were received from the licensee.
Proprietary materi-al information was reviewed in the course of the inspection, but is not included in this repor e
.
Inspector followup item 50-250 and 50-251/89-32-01:
For incore-excore nuclear instrument correlation measurements, establish an acceptance criterion that the correIation coefficient be at least 0.98.
(Paragraph 3.c)
Unresolved item 50-250 and 50-251/89-320-02:
Review EXCAL calculations of channel voltages as a function of AFD for acceptability.
(Paragraph 3.c)
Acronyms and Initialisms Used in This Report AFD ANS ANSI-ARO Cc)s-DDPS-dp EFPD-gpm HZP ICRR-IFI IRM ITC.
MCC MTC Mwth-OP pcm ppmB-PRNI-QPTR-RCS RTP SDM SRM TS UNR axial flux difference, expressed in percent American Nuclear Society.
. American National Standards Institute all rods out RCS boron concentration counts per-second digital data processing system differential pressure effective full power days gallons per minute hot zero power inverse count rate ratio inspector followup item intermediate range (neutron) monitor isothermal temperature coefficient motor control center moderator temperature coefficient megawatts thermal operating procedure percent millirho,-a unit of reactivity parts per million boron power range nuclear instrument quadrant power tilt ratio reactor coolant system rated thermal power shutdown margin
.
source range (neutron) monitor technical specification unresolved item of full power Attachments" l.
2.
3.
Differential Worth of Control Bank C, Turkey Point 4 Excore Chamber Calibration Voltages, Turkey Point
Heat Balance Data, Turkey Point
TPD!!R2 Heat Balance Results, Turkey Point
Cl
TURKEY POINT 4, CYCLE 12 CONTROL BANE C
.
Attachment
E3Inspector Licensee 100 150 200 850 Steps II'tbdravrn
TURKEY POINT 3: N41 Attachment
8.8 (Cl
~ 83 Q
A Excore Chum,her Cakihro.Non I
I I
I I
I I
I I
I I
I.
I I
I I
VoLtuges~ Insp-Top
~ Insp-Bottom
~ LicTop
~ Lic-Bottom 7.8-30-20
0
-
=10 Axial Offset (r.')
Ci
Pl hNT PhRk5KTXRS:
HEhl BhLhNCE Dh!h TORKET POIN! 3 86-29-89 Attachment
REhCTOR COOLhNT STSTK5 Puap Power (NN each)
4.1 Puap Efficiency (E)
98.8 Pressuriser Inside Diaaeter (inches)
84.8 REFLECTIVE INSOI hTION Inside Surface hrea (sg ft)
11,122 Heat Loss Coefficient '(BTOs/hr sg ft)
55.89 STKh5 GENERhTORS Doae Inside Diaaeter (inches)
Riser Outside Diaaeter (inches)
Nuaber of Risers Noisture Carry-over (E) in h Noisture Carry-over (E) in B
Noisture Carry-over (E) in C
159.98 19.88
$ 113 8.125 8.125 8.125 LICBNSED THERNhL PONER (5Nt)
2288 NONREFIEC!IVE INSOLk!ION Inside Surface hrea (sq ft)
7,851 Thickness (inches)
4.9 Theraal Conductivity (BTOs/hr ft f)
II.835 DkTh:
715E STEh5 GENKRhTOR h Steaa Pressure (psia)
feedwater flow (E6 lb/br)
feedwater Teaperature (F)
Surface Blowdown (gpa)
Bottoa Blowdown (gpa)
Mater Level (inches)
SKT
SK! 2 1215; 1229 841.6 841.3 3.173 3.176 428.7 428.7 S.s 8.8 45.2 45.2 538.1 538.1 1'I5E STEk5 GENERhTOR B
Steaa Pressure (psia)
Feedwater flow (B6 lb/hr)
feedwater !eaperature (f)
Surface Blowdown (gpa)
Bottoa Blowdown (gpa)
Nater Level (inches)
SKT
1215 1228 829.9 827.7
"
3.195 3.198 426.7 428.7 8.8 8.8 45.2 45.2
= 537.6 537.3 STKh5 GENKRhTOR C
Steaa Pressure (psia)
feedwater flow (E6 lb/hr)
feedwater Teaperature (f)
Surface Blowdown (gpa)
Bottoa Blowdown (gpa)
, Nater 1 evel (inches)
814.4 889.8 3.236 3.248 428.7 428.7 8'.8 e.s 45.2 45.2 533.4 533.2 LE!DONE LINK Flow (gpa)
Teapereture (f)
55.1 55.2 546.7 546.7 CHhRQING LINE Flow (gpa)
Teaperature (f)
43.8 42.7 475.8 476.8 PRKSSORIZKR Pressure (psia)
Nater (evel (inches)
I'262.2 2261.9 299.4, 288.8 RKhCTOR T ave (F)
T cold (f)
573.3 573.2 546.7 54 Attachment
Page l,of 2 DATA SET
OF
1215 hours0.0141 days <br />0.338 hours <br />0.00201 weeks <br />4.623075e-4 months <br /> STEAM GENERATOR A HEAT BALANCE TURKEY POINT
86-29-89 ENTHALPY FLOW (BTUs/lb)
(E6 lb/hr)
POWER POWER (E9 BTUs/hr)
(MWt)
Steam Feedwater Surface Blowdown Bottom Blowdown Po~er Dissipated
'TEAM GENERATOR B Steam Feedwater Surface Blowdown Bottom Blowdown Power Dissipated STEAM GENERATOR C
Steam Feedwater Surface Blowdown Bottom Blowdown Power Dissipated OTHER COMPONENTS Letdown Line
" Charging Line Pressurizer Pumps Insulation Losses Power Dissipated REACTOR POWER 1197.4 486.9 517.8 468.3 1.197. 7 486.8 515.8 459.4 1198.1 486.8 512.3 458.1 543.1 468.8 782.7 3. 154-3.173 8.88888 8.81822 3. 178-3
~ 195 8. 88888 8.81824.
3.218-3.236 8.88888 8.81825 8.82881-8 '1757-8.88886 3.777-1.291 8.88888 8.88839 2.4946 3.886-1.388 8.88888 8.88838 2.5145 3.856-1.317 8 '8888 8.88836 2.5477 8.81138-8.88888-8.88861-8.83788 8.88188-8.83418 738.6 736.4 746.2-18.8 228 Cl
Attachment
Page 2 of 2 HEAT BALANCE TURKEY POINT 3 86-29-89 DATA SET
OF
1228 hours0.0142 days <br />0.341 hours <br />0.00203 weeks <br />4.67254e-4 months <br /> STEAM GENERATOR A ENTHAIPY (BTUs/lb)
FLOH POHER POHER (E6 lb/hr)
(E9 BTUs/hr)
(MHt)
Steam Feedwater Surface Blowdown Bottom Blowdown Power Dissipated STEAM GENERATOR B Steam Feedwater Surface Blowdown Bottom Blowdown Power Dissipated STEAM GENERATOR C
Steam Feedwater Surface Blowdown Bottom Blowdown Power Dissipated OTHER COMPONENTS Letdown Line Charging Line Pressurizer Pumps Insulation Losses Power Dissipated REACTOR POHER 1197.4 486.9 516.9 468.3 1197.8 486.9 514.6 459.2 1198.3 486.9 511. 4.
457.7 543.1 461.2 782.7 3.157-3.176 8.88888 8.81822 3:173-3.198 8 '8888 8.81824 3.222-3.248 8.88888 8.81826 8.82884-8.81743-8.88886 3.781-1.292 8.88888 8.88839 2.4967 3. 888-1.298 8 '8888 8.88838 2.5185 3 '61-1.318 8.88888 8.88836 2 '587 8.81132-8.88884-8.88861-8.83788 8.88188-8 '3412 731.2 735 '
747.8 2283.5