IR 05000369/1991008

Insp Repts 50-369/91-08 & 50-370/91-08 on 910408-12.No Violations Noted.Major Areas Inspected:Areas of post- Refueling Startup Test for Unit 2 & Routine Core Surveillance Activities for Both Units

ML20138F746
Person / Time
Site:	McGuire, Mcguire
Issue date:	04/17/1991
From:	Breslau B, Burnett P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
To:
Shared Package
ML20138F736	List: IR 05000369/1991008 ML20138F732
References
50-369-91-08, 50-369-91-8, 50-370-91-08, 50-370-91-8, NUDOCS 9610180054
Download: ML20138F746 (10)

Text

{{#Wiki_filter:. - - -- - -.

# Sfo,r   UNITED STATES   I I

f* 'o NUCLEAR REGULATORY cOMMisslON

[[&

3 .j n REGION 11 101 MARlETTA STREET, l l l * f , 5 ATLANTA, GEORGI A 30323 l .

%.....l l

i Report Hos.: 50-369/91-08 and 50-370/91-08 l Licensee: Duke Power Company ' 422 South Church Street Charlotte, NC 28242 Docket Nos.: 50-369 and 50-370 License Nos.: HPF-9 and NPF-17 Facility Name: McGuire 1 and 2 Inspection Conducted: April 8 - 12, 1991 Inspector: ; b / dadhe 4I,I?! 47'P. T. Burnett / Date Signed Approved 'by: / v s "[/ '/9/

  '8. A. Breslau, Chief   /

Date Signed Operational Programs Section f Operations Branch l Division of Reactor Safety i l SUMMARY Scope: , This routine, unannounced inspection addressed the areas of post-refueling startup tests for Unit 2 and routine core survei' lance activities for both unit Results: Startup tests for Unit 2, cycle 7 were pecformed from December 23, 1990, to . January 8, 1991. Test methodolos y and results were acceptable. Concerns were identified regarding test acceptance criteria (paragraph 2.c) and the schedule ' for performance of an incore and excore nuclear instrument calibration after reaching full power (paragraph 2.d).

Surveillances of core power distribution, end-of-life moderator temperature coefficient, and incore-excore nuclear instrument correlation, were performed l with acceptable frequency and methodology for both units for the period ' reviewed (paragraph 3).

Hot leg streaming and the concomitant problems of determining true hot leg I temperature and reactor coolant flow have increased on Unit 2 since the baffle flow direction was changed to upward during the last refueling outage. Key ' . 9610180054 910422 PDR ADOCK 05000369 G PDR ,

--- _ . _ _ _  _ . . _ _ _ . _ _ . - . _ . _ _ . _ . . _ _ . _ _ _ _ . . - _ . ._. _ _ _
.
       !

I

l parameters are being recorded and trended onsite, and the offsite design i engineering organization is analyzing and attempting to correlate any changes j in the parameters. Additional testing is planned for Unit 1 both before and after undergoing this same modification during the next refueling outage (paragraph 4).

No violations or deviations were identifie l l l l l l ! l l ! ,

!

. -. .-- . . - _- - - . . - _ - - -. _, .--

,

     !

REPORT DETAILS l I

     , Persons Contacted     l Licensee Employees
*D. Bumgardner, Unit 1 Operations Manager
*J. Day, Engineer, Performance    i
*G. Gilbert, Superintendent of Technical Services  l B. Hamilton, Superintendent of Operations   l
*M. Hatley, Maintenance Engineering Supervisor   !

A. Hinson, Maintenance Engineering  ! ,

* Kitlan, Jr., Reactor Engineer l *L. Kunka, Engineer, Compliance T. McConnell, Station Manager

' l

*H. Nazar, Engineer, Performance G. Pelzer, Production Support - Training
*J. Pope, Superintendent of Maintenance
* Sample, Superintendent of Maintenance j *R. Sharpe, Compliance Manager l *G. Small, Safety Review Group

!

*J. Snyder, Performance Manager Other licensee employees contacted included engineers, technicians, and office personnel.

l NRC Resident Inspectors P. K. VanDoorn, Senior Resident Inspector l *T. A. Cooper, Resident Inspector

* Attended exit interview on April 11, 199 Acronyms and initialisms used throughout this report are listed in the last paragrap . Unit 2, Cycle 7, Post-Refueling Startup Tests (72700, 61708, 61710) Precritical Activities l  The following procedures completed prior to criticality were reviewed by the inspector:

1) PT/0/A/4150/38 (Approved August 24,1990), Controlling Procedure for Transition to Next Fuel Cycle, was performed in the period from October 31, 1990, to January 9, 199 The purpose of the test was to assure that cycle-dependent constants were updated in the computer programs used to analyze and predict core performance. The affected programs

,

were the OAC programs: XENON FOLLOW / PREDICT, SAMARIUM . FOLLOW, GENERAL 67, and RAM. Off-line computer programs that were updated were: RHOBAL, LDFLW, EQUILXE, EQUILSAM. XENON,

and SAMARIUM.

,

- _ _ - ,  _ __ _ _ _ _ __.. _ _ _ _
. . - _  .
.
      ;
.

All enclosures to the procedure appeared to have been performed properly and completely, with one exception: Enclosure 13.3, step 13.3.1, requires that a data sheet be completed for five randomly selected burnup ranges to compare calculated RA0C W(z)s with predicted. Only-three comparisons were mad Discussions with plant personnel confirmed they were aware of the problem. Only three sets of burnup data are provided, and additional ranges would have to be created by interpolation. They will change the procedure to either require the interpolations or to reduce the required number , of comparisons to three.

I l 2) PT/0/A/4600/78 (Approved January 11,1990), Prestartup Nuclear Instrument System Realignment Following Refueling, was completed in the period October 10-31, 199 Comparisons of E0C 6 power distributions with those predicted for B0C 7 led to reducing the expected full power currents from the PRNIs by about eight percent and from the IRNIs by about thirteen percent. These corrections assured conservative power indications during startu ) PT/0/A/4600/77 (Approved August 30,1989), Procedure for full Length Rod Control Cluster Assembly Drop Timing, was performed in December 1990. The slowest rod drop time was 1.44 seconds, and the fastest rod drop time was 1.24 seconds. The drop time limit of 3.3 seconds from beginning of decay of stationary gripper coil voltage to dashpot entry, of TS 3.1.3.4, was satisfied in all case The average rod drop time for cycle 7 was significantly , faster than previously measured by about 0.1 second.

' Licensee personnel suggested that the reduced drop time was the result of reducing the withdrawal limits from 230 steps to 225 steps. The measurement system used for the cycle 7 tests was the same as usea in cycle 6 measurements; thus, a systematic error in measurement is unlikely, Initial Criticality Tests The inspector reviewed the completed test procedures discussed below: 1) PT/0/A/4150/28 (Approved December 13,1989), Criticality Following a Change in Core Characteristics, was performed on December 23-25, 1990. C was reduced from 1925 ppmB to 1521 . l

ppmB, and criticality wa,s achieved upon withdrawing control ' banks in overlap to 0-bank at 44 steps. This procedure was l performed again on December 27-28, 1990, for rod withdrawal j only, to continue the interrupted startup test program.

i l !

 ..  - -. . . - . . . - - . . .

l

-

l

3 l t The licensee is reviewing the benefits and safety l significance of diluting and then pulling rods to ! criticality relative to pulling rods and then diluting to criticality. The latter method would appear to have some benefit in assuring SDM is preserve ) PT/0/A/4150/21 (Approved December 19,1990), Post Refueling Controlling Procedure for Criticality, Zero Power Physics, and Power Escalation Testing, was performed in the period from December 23, 1990, to January 8, 1991. Some of the test activities

controlled by enclosures to this procedure were

a) Shutdown Margin, which confirmed excess SDM at the rod I insertion limits for zero power operatio i b) Reactivity Computer Checkout Data Sheet, which l

confirmed that the reactivity computer solutions were ' in acceptable agreement with manual solutions of the t inhour equatio c) Nuclear Heat Determination Data Sheet, which established a testing range below the heating level l for conducting the zero power test c. Zero Power Physics Tests a The inspector reviewed the completed test proceiures discussed ' l below: 1) PT/0/A/4150/10 (Approved September 11,1989), Boron Endpoint Measurement, was performed on December 26, 1990. The measured C, of 1600 ppmB at ARO was in agreement with the predicted value of 1583 ppmB.

i 2) PT/0/A/4150/11 (Approved September 13,1989), Control Rod l Worth Measurement, was performed on December 26, 1990, for ' shutdown bank B. Rod insertion was used to compensate for dilution at a rate of about 300 pcm/h The measured worth of shutdown bank B was 826 pcm, which was in agreement with , the predicted value of 860 pcm. The inspector independently ! analyzed the reactivity computer traces from the measurement and obtained an integral worth of 828 pcm. The differential worth curve showed no unusual structur The individual rod motions were too large to resolve the internal grids in the plotted differential worth curve. However, throughout the test. reactivity was kept within the calibrated range r established in PT/0/A/4150/21, Enclosure 1 ) H/0/A/4150/11A (Approved December 20,1990), Control Rod Worth Measurement: Rod Swap, was performed from December 26 to December 29, 1990. All results satisfied the acceptance i

   -__ _ -. . . . .
.

l- l i i

     '

! 4 l criteria of ANSI /ANS-19.6.1 as well as those of the l procedur l l 4) PT/0/A/4150/12 (Approved August 23,1989), Isothermal Temperature coefficient Measurement, was performed on l December 26, 1990. The predicted and measured ITCs were

+1.90 pcm/*F and +1.7 pcm/*F, respectively. Agreement

' within 2 pcm/*F satisfied the acceptance criterion of step 1 The resulting measured MTC of +3.1 pcm/*F was within i

     '

the limits of TS 3.1.1.3 and COLR2-7 item 2. The inspector independently analyzed the licensee's data and arrived at essentially the same result The procedure calls for temperature changes of at least 3*F, but most similar facilities require at least a 4*F chang The licensee had already identified a need to increase the temperature span and is planning to change the procedure before the Unit I startup for cycle Within limits, the precision of the test will increase with increasing l temperature spa The results of the control rod worth measurements were acceptable with respect to the criteria of ANSI /ANS-19.6.1.

The acceptance criteria within the licensee's procedures i were different from the standard and inconsistent between the two procedures and with the SPTP. The licensee agreed to review the procedures for deviations from the standar Power Escalation Tests 1) PT/0/A/4600/02E (Approved January 15,1990), Incore and Nuclear Instrumentation System Recalibration: Post Outage, was performed in the period from December 31, 1990, to January 3, 1991, to obtain data to determine the relationship between the incore and excore quadrant A0s and AFDs. Twelve quarter-core flux maps were obtained as power was increased from 52% RTP to 76% RTP. The incore A0 varied from + 13% to -5.8% during that period, with an almost straight line relationship between power and AO.

I The licensee used the INEXCAL program to analyze the data obtained. The inspector used a least-squares spreadsheet l with the SUPERCALC3 microcomputer program to perform an independent evaluation of the data. The results were virtually identical for the zero-offset currents and correlation coefficients for each of the eight excore ion chambers. INEXCAL reports a negative correlation coefficient l when the slope is negative. That is not correct, but it was judged to be an output editing problem rather than an analytical problem. None of the four top chambers satisfied the acceptance criterion that the correlation coefficient be i i

,

_ _ _ _ _ . _ . _ . . _ . _ _ _ _ _ _ . _ . _ _ . _ _ _ _ . _ _ . _ _ __

-

l

at least 0.985; the actual values ranged from 0.966 to 0.977. Nevertheless, the licensee decided to proceed with i power escalation. This would not have been a poor decision, ' if the correlation had been performed igain shortly after . reaching full power. However, a complete recalibration was I not performed for nearly three months. As discussed in paragraphs 3.a and 3.b, the later results fo'r slope and zero-offset current were quite different.

The licensee is considering several options, including j relaxing the acceptance criterion for the correlation j i coefficient and scheduling PT/0/A/4600/02G for performance l soon after reaching full power.

! i Summary , The tests performed during the Unit 2, cycle 7 startup and the , ! method of their performance conform to the recommendations of l l ANSI /ANS-19.6.1, Reload Startup Physics Tests for Pressurized

Water Reactors. Some of the licensee's acceptance criteria l differ, nonconservatively from those of the standard, but all of

! the test results of this series of tests do satisfy the criteria of the standar All of the procedures reviewed showed positive ev;dence of l effective peer review of the entries and calculations. From the i number of corrections observed, this quality of review was very necessar No violations or deviations were identified.

! Core Performance Surveillance Tests (61702, 61705, 61708) I ' The following completed core performance surveillance tests were reviewed by the inspector: PT/0/A/5150/27 (Approved December 8, 1989), Moderator Temperature Coefficient Determination at End of Cycle by the Boration/ Dilution C i Method, 292 ppmB to was 301 performed ppm for Unit 2 on June 1,1990.The MTC measured emperature i decrease was -18.4 pcm/* During the temperature increase, the measured MTC was -16.3 pcm/* Since the range of results was less than 20% of the mean (-17.4 pcm/*F); the performance of the test was acceptable. The result was satisfactory by virtue of being less negative than the limits of TS 3/4.1.1.3 and COLR2-6 item 2.1.2 (-32 pcm/*F).

( PT/0/A/4600/02A, (Approved April 13,1989), Incore and Nuclear < Instrumentation Systems Correlation Check, was reviewed for timely r completion for Unit 1, cycle 7 and for Unit 2, cycle 7. For both ! units performance has been at the required frequency and has

identified the need for additional surveillance activities, which

!

.

     . . . .

.

  '6 were performed as required, PT/0/A/4600/02G (Approved June 26,1989), Incore and Nuclear Instrument System Recalibration: RA0C, has been performed three times, at the required intervals, for Unit 1 during cycle 7. All results reviewed were acceptabl This procedure has been performed once in cycle 7 for Unit 2, in the period from March 25 to April 9,1991. The correlation coefficients were all greater than 0.99 and satisfied the acceptance criterion. The inspector independently analyzed the data and obhined essentially the same results for zero-offset currents and correlation coefficients as reported by the license However, the zero-offset currents were all significantly larger than those obtained during startup when PT/0/A/4600/02E was performe In addition, all of the slopes of current versus A0 appeared to be significantly changed. The differences raise the question whether the startup test, although necessary, is sufficient for full power operation. (See also paragraph 2.d) PT/0/A/4600/02F (Approved June 21,1989), Incore and Nuclear Instrumentation System Interim Calibration, was performed on February 5-13, 1991, for Unit 2, in response to problems identified during the performance of PT/0/A/4600/02A. Performance of this procedure did increase the zero-offset currents to values close to those obtained later in the full-scale recalibration (PT/0/A/4600/02G), but it could not correct the error in the slope of current versus AF j PT/0/A/4150/02A, (Approved May 5, 1990), Core Power Distribution, was reviewed for both units for their respective cycle 7. The results of 15 full core flux maps were reviewed for Unit 1. In i all cases, the intervals between flux maps were less than the 31 l
     '

EFPD specified in TS 4.3.2 and the measured hot channel factors were less than the limits of TS 3.2.2, 3.2.3, and 3.3. Similar inspection results were obtained from the review of six full core flux maps performed for Unit No violations or deviations were identifie . Hot Leg Streaming and Reactor Coolant System Flow (61706) TS 4.2.3.5 requires that the NC total flow rate be determined by precision heat balance at least once per 18 month PT/2/A/4150/13 (Approved January 23,1991), NC Flow Calculations, was performed for Unit 2, cycle 7, between January 28, and February 19, 1991. Good consistency was observed among three one-hour, precision heat balance test runs, and an acceptable flow rate through the reactor vessel was determined. However, it was noted that the hot leg RTDs, three per loop, were not giving uniform temperatures. Within a loop, the n:aximum differences between RTDs ranged from 3.3 to 6.7'F. The

. .

- - . - - - . - - - . . - - - . _ . . - - - . _ - _ - . _ -
*
,
       '
.

l 7 L licensee described these differences as being larger than observed

    ~

l l previously and postulated that the change results from a modification ! performed before the start of the cycle: The flow in the core baffle area was changed from downflow to upflow to' eliminate jet impingement damage to the fue Similar hotleg streaming effects have been observed at other facilities, , and usually have been ascribed to low-leakage cores, which produce low ' coolant outlet temperatures from peripheral fuel assemblies. Both McGuire units employ low-leakage cores. Based upon the evaluation of one of the three precision heat balances, the inspector estimated that the streaming effects contribute 2.4% uncertainty to the measurement of core flow. The licensee stated that their current analysis includes 1.7% uncertainty from streaming effects.

L The licensee is continuing a program of trending the parameters which l illustrate or may correlate with the streaming phenomenon. They plan to ! ! take a series of precision heat balances on Unit 1 before and after performing the baffle flow modification during the next refueling . outage. They also plan to install 12 external RTDs on each Unit 2 ! hotleg, downstream of the thermowells, during the next refueling outage ! for that unit. From the additional RTDs, the licensee expects to gain , insight into the temperature distribution within the hotlegs. They are l ! currently gaining experience with a similar temperature measuring system

on the pressurizer surge lin l

! l No violations or deviations were identifie , Exit Interview (30703) The inspection scope and findings were sumarized on April 11, 1991, 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. The licensee did not identify as proprietary any materials provided to or reviewed by the inspector during this inspectio ; i

l

       ;

l i I

. - . _  . ..  . - -_ ..

- -. -. . - . _ _ -
    -

l .' l-

6. Acronyms and Initialisms Used in This Report AFD axial flux difference ANS American Nuclear Society ANSI American National Standards Institute A0 axial offset ARO all rods out l B0C beginning of cycle C boron concentration in the NC system CblR(n-m) Core Operating Limits Report (unit-cycle) F da nuclear enthalpy hot channel factor F nuclear heat flux hot channel factor EEPD effective full power days E0C end of cycle FSAR Final Safety Analysis Report gpm gallons per minute

IAE instrumentation and electrical department

' IP instrument procedure IRNI intermediate range nuclear instrument ITC isothermal temperature coefficient MTC moderator temperature coefficient  : ! NC nuclear coolant system l NIS nuclear instrument system ' OAC operator assist computer pcm percent millirho ppmB parts per million boron PRNI power range nuclear instrument PT periodic test , RAOC relaxed axial offset control l RTD resistance temperature device l RTP rated thermal power j SDM shutdown margin l SPTP Startup Physics Test Program (a reference document) l TS Technical Specification l l ! l

:

l l l l }}