IR 05000155/1988017

Safety Insp Rept 50-155/88-017 on 880711-0906.Violation Noted.Major Areas Inspected:Problems Encountered W/ wide-range Monitors During Plant Startup from 1988 Refueling Outage & Modified Control Rod Insertion Sequence

ML20154J807
Person / Time
Site:	Big Rock Point File:Consumers Energy icon.png
Issue date:	09/16/1988
From:	Phillips M, Rescheske P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
Shared Package
ML20154J768	List: IR 05000155/1988017 ML20154J764 ML20154J799
References
50-155-88-17, NUDOCS 8809230142
Download: ML20154J807 (7)

Text

-_ _____ _-_____ ____ _____ _ ____

I

. .

l

!

-

..

i  !

! I U.S. NUCLEAR REGULATORY CCMMISSION

REGION III

l l

l 4 Report No. 50-155/88017(ORS) (

) i-l Docket flo. 50 155 License No. OPR-06 j Licen:ee: Consumers Power Company

212 West Michigan Avenue .

Jackson, MI 49201 ,

!

Facility Name: Big Rock Point Nuclear Plant

)

!

j Inspection At: Charlevoix, Michigan  :

a j Inspection Conducted: July 11 through September 6,1988 r

Inspector: Peggy R. Rescheske 8 l Date r

.J #

Approved By: ente P. (1111ps, Chief Operational Programs Section

/[NI i Date l I

Inspection Summary  !

,

Inspection on July 11_through September 6.1988 (Report No. 50-155/88017(ORS))  !

Areas Inspected

Special, announced, safety inspection focusing on the

problems encountered with the Wide Range Monitors during the plant startup from the 1988 refueling outage, and the resultant LER which documented the event (IP 92700). At the request of DRP, the modified control rod insertion f sequence used at the end of the previous fuel cycle was also examined  !

(IP 92701).

'

Results: One violation was identified during the review of the Wide Range Monitor (WRM) problems. As discussed in Paragraph 2, the licensee failed to adequately test the newly installed WRMs. such that, the instruments did not operate as required when placed in servic The safety significance was minor, based nn the reevaluation of plant transients performed by the licensee. The violation does not equire a response. Both the violation (155/88017-01) and the LER (155/88007) are considered close ;

0809230142 a00916 DR ADOCK 050 1]5

,

-

_ .- - -

. . ..

. .

. _ _ _ - _ _ _ _ _ _ _ ___ __ _

. .

..

,

OETAILS 1. persons Contacted

• T. W. E1 ward, Plant Manager

• R. J. Alaxander, Technical Engineer G. R. Boss, Reactor Engineer J. A. Johnson, Senior Technical Analyst R. L. Krchmar, Quality Assurance R. E. Schracter, Electrical !&C Engineering Supervisor The inspector contacted other licensee staff during the course of the inspectio *0enotes persons attending the exit meeting on August 24, 1988.

2. LER No. 88007 Discussion Introduction During the 1988 refueling outage (April 9 through June 27,1988),

at Big Rock Point (BRP), the original intermediate and power range neutron monitoring system was replaced with General Electric Company (GE) NUMAC DC Wide Range Monitm s (WRMs). The design of the new WRMs is discussed below. During power escalation on June 28, 1988, subsequent to the outaae, the WRMs could not be calibrated / adjuste The reactor was shutdown on June 29, 1983, when the licensee determined that Technical Specification (TS) 6.1.5(e) could not be met. LER No. 68007 was issued on September 2, 198 The sequence of events leading up to the reactor shutdown is discussed belo Design and Installation of the WR3 The rep 1;. cement of the original instrumentation was based on the need to have a more stable, reliable neutron monitoring syste The DC WRM, manufacturci by GE, is a micror,omputer based instrument and was selected to replace the Log-N/ Period amplifiers (intermediate range) and Picoammeter chennels (power range), The WRM was designed to monitor bo',h the intermediate and power ranges, utilizing the three existing out-of-core power range cetectors (compensated ion chambers). Monitoring of the two intermediate range detectors was eliminated, although the actual detectors were not remove The WRM measures the input current from a detector; calculates power level, reactor period, and power rate; and performs trip and alarm functions. The intermediate range portion of the WRM provides

_ _ _ - _ _ _ _ _ - _ _ _ _ _ _ _ _-

. .

..

,

neutron flux level coverage from 1E-7 to 14 powsr. Short period protection was changed from one out of two logic to two out of three logie wich the WRhs. Replacement of the power range instrumentation covers neutron flux levels from 1*4 to 150*4 pcwer. The picoammeter range switches were eliminated; alarm and trip functions were '

maintained. Prior to this modification, administrative controls '

limited the rate of power increase, in accordance with TS. Additional '

protection was incorporated in the WRMs, with rate-of-change logi ,

Two automatic reactor trips were added to limit power ascent and ;

enforce the administrative requirements in the T Purchase ano nrocurement of the WRMs extended over a two year period; installation was completed during the 1988 refueling outage; -

and the documentation for the modification was completed in Facility [

Change Package FC-599 The WRMs were designed to operate from 1E-7 -

to 150's power with an input current range of IE-13 to 1E-3 amper l In a memorandum dated April 10, 1986, Consumers Power Company i furnished information to GE regarding the expected datector output in response to questions from GE. The information stated that 100*4 ;

detector output was between 0.74E-4 to 2.51E-4 ampere. This range '

was based on detector design and picoammeter bench testing. The licensee had performed output current calculations for minimum and maximum resistance (at 100% picoammeter setting), using information contained in a nemorandum from GE to BRP, dated September 10,196 The memorandum stated that 125*4 indication was 3.16E-4 amper :

According to the licensee, these measurements were performed on ;

the bench rather than during clant operation, which would require removing the instrun9ent f rom service. The plant specific data was not incorporated in the standard design of the WRMs supplied by G i Further, this information was not included in the purchase or procurement specifications (or revisions thereof). Therefore, the i discrepancy between the WRM design specifications and plant specific ,

requirements was not recogni:ed by the licensee during design checks

~

and reviews. The document reviews, along with the testing of the !

instruments prior to startup, were performed using GE design ;

specification The deficiency was disclosed to the licensee j when attempts to calibrate / adjust the WRMs failed during power :

ascensinn testin !

Sequence of Events l

i On June 12, 1938, heads-off critical was performed with the startup .

channels responding well. When reactor critical was reached, the new WRMs were responding and on rang The plant Revic Committee i (pRC) meeting minutes from June 14, 1988, concluded that the new j instrumentation worked as designe On June 28, 1933, the reactor was t.Len critical on plant startup, i and the WRMs functioned smoothl However, the WRMs were reading '

_ _ _ _ _ _ _ _ _

. .

'

-

low and were unbalanced. The plant was at approximately 20 HWEG (25% rated power) is read off the megawatt electric (MdEG) meter, and the WRMs were indicating about 8*4 (average) power. The licensee did not consider this discrepancy unusual, based on the behavior of the detectors and picoammeters during prsvious startups. For example, on the March 1937 startup, a heat balance calculated an -

expected reading of 75.3*o (where 100% was equivalent to 145 Kdt),

and the actual picoammeter readings were 41.% , 36.54 and 86.5' In accordance with the licensees requirements, a calibration of the instruments was not required prior to 50% rated power. Note:

100*4 rated power is equivalent to 240 Kdt (megawatts thermal) which is approximately 80 KdEG (depending on plant efficiency). A special PRC meeting was held to review the WRM respense. Direction to the operations personnel was given in the daily orders. The first recommendation was to physica'ly reposition the detectors to higher I flux locations and attempt to balance the readings. The as-found l WRM readings were: 10*4, 8's, 7*4; after repositioning, the as-lef+

readings were: 12*s, 9*4, 11%. The second recommendation was to subsequently raise power to about 50*4 power, stabilize the reactor, and perform a heat balance calculatio On June 29, at 35 MWEG, the detectors were repositioned again; however, the moves resulted in all detectors being located as-left at 20 KdF.G. It was noted that in the escalation from 20 to 35 MWEG, the WRMs showed about a 5*4 increase. A heat balance calculation was performed which determined that actual reactor power was 45.3*4 of rated. The WRMs could .nly be adjusted to indicate 21*s, 18%, and 21*4, respectively; thesi readings were approximately one-third of the calculated power. .'n accordance with the operability requirements in TS 6.1.5(e), the lice'see brought the plant to cold shutdow Corrective Actions The inability to calibrate / adjust the WRMs was determined to be due to the inappropriate design input sensitivity range of IE-13 to 1E-3 ampere. A specification change was initiated on June 29, 1988. A change to a pro;rammable memory circuit (PROM) was subsequently performed and validated by GE. The gain constant was changed from 6.67E-4 input amperes for 100*e po er indication to 2.11E-4 input amperes for 100*4 indicatio The ability to adjust detector sentitivity and rated power remained in the range of $20*f, as originally designed. On July 1,1988, the WRMs were returned to service, calibrated, and teste Additional adjustments problems were discovered by the licensee curing subsequent calibrations of the WRMs. A second specification change was initiated on July 8,1983. It was determined that a software modification was needed to increase the range of adjustment for both the detector sensitivity and rated power adjustments

. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _

_ - _ _ _ - - _ _ _ - _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -_ - _ _ _ - _ - _ _ _ -

'

. .

.

l (design was 220%, or 0.8 to 1.2). On July 10, 1988, GE installed ,

new PROMS in the WRMs which increased the range of adjustability to 0.3 to 3.0 for detector sensitivity, and 0.8 to 2.0 for rated powe b. Licensee Evaluation .

On September 2, 1988, LER No. 88007 was issued b/ the licensee documenting the circumstances surrounding the plant shutdown, and the actions taken to resolve the deficiencies. Included in the LER was an assessment of the safety aspects under the conditions c of the plant during the event. The reacter power was assumeu to be ,

108.7 Kdi (45% rated) and the high flux protection was out of '

calibration (or disabitd). Each analyzed plant transient which

, could be affected by out of calibration nuclear instrumentation was l

reviewed b/ the licensee. Th3 result was that only four transients  ;

could be affected: (1) loss of feedwater heating, (2) effects of mispositioned control rod, (3) increase in feedwater flow, and ,

J (4) turbine trip without bypass. The licensee analysis concluded, for the first three transients listed, that the maximum power increase was below the required high flux trip at 120% of rated power. Furthermere, the minimum critical power ratio remained within an acceptable value and no fuel damage was expected. In j the original analysis (July 1981), turbine trip without bypass

! was determined to be tne limiting case in term; of minimum critical  ;

power ratio. The results from reevaluating this transient was that a reactor scram would cccur at 7.3 seconds due to a high steam drum

] pressure. Assuming that the high flux trip was disabled, power would increase rapidly and peak at 2.48 seconds at 375.2 Rdt (156% ,

rated power). The power excursion would turn around due to doppler '

effects and the reactor scram would occur at ?93 Kdt. The licensee -

j concluded that since the power and heat flux levels were less severe

than the original analysis (performed at full power), fuel damage

'

would not exist.

. i 1 c. Inspection Results  !

l  ?

,

The inspector reviewed the available documentation related to l l

the WRM modification (FC-599A), and held discussions with licensee l

) staff knowledgeable of the problems encountered during startu l

'

Documentation reviewed included the following: design, punct.ase,  !

) and procurement documents; 10 CFR 50.59 Safety Evaluation, TS ,

amendment and Safety Evaluation Report; control room legs, daily l orders, and PRC meeting minutes; specification changes, event report, L and engineering analyses; procedures and results of testing; and F 4 other miscellaneous records and corretpondence. Discussions with '

i the licensee focused on procurement specifications a'id *

) post-installation / modification testing. The resu'ts from (

j the inspection were as follow ;

! l I

l  !

! .

l L

!

}

t

.

- - _ _ _ _ _ _ _

.

..

. .-

(1) The plant specific data given to GE m s of a memorandum dated April 10, 1986, should have bt  % ,, orated into the detign specifications for the WR . -.y base been an oversight on the part of GE, the licensee, or bot (2) Review of the design, purchase, and/o* nrocurement documents by the licensee could have identified the discrepancy in design and plant specific data prior to installatio This level of detail would not have been within the scope of quality assurance reviews; however, engineering personnel should have identified the deficienc During the course of the inspection, the licensee revised two plant procedures, adding or clarifying the requirement: for reviewing design changes. Revision of the "Design heview Checklist" added a check of design compatibility with kr mn operating parameters. The "Design Input Checklist" was revised to include a section on compatibilitt requirement This new section requires that the reviewer identify compatibility of the design with bacun operatiig parameters and operating experienc (3) In accordance with 10 CFR 50, Appenuix B, Criterion XI, "Test Control," testing is required to demonstrate that systems per?orm satisfactorily in servic The WRMs were tested prior to startup; however, the testing only rotulted in assu ance that the WRMs would operate as designed. The testin did not demonstrate that the WRMs vould operate as requir6d, and '

perform satisfactorily in service. The failure to adequately test the newly installed WRMs is considered to be a violation (155/88017-01(DRS)). The safety significance was determined to be minor, based on the reevaluation of plant transients performed by the licensee. However, the reactor protectio'1 system was in a degraded condition during the even Due to the corrective actions taken by the licensee, this violation does not require a response. This item is close (4) (Closed) LER 155/S8007, "Technical Specification Shutdown -

Inoperable Neutron Monitoring System." Based on the aoove discussions, this LER is considered to be closed. The event, corrective actions, and safety significance, was adequately addressed by the license . Modified Control Rod insertion Sequence The licensee modified the control rod insertion sequence for the end of Cycle 22 (April 193S), per Specification Change No. SC-8S-006. The purpose for the change was to reduce the amount of time required to

- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -

__

. .

. .

. . ..

. .

,-

. .

insert all rods and bring the unit to celd shutdowo, after the turbine generator had been taken off line. This was accomplished by establishing a sequence which allowed running tt.e rods in when the reactor was still cr!tica The licensee performed at 10 CFR 50.59 Safety Evaluation and an engineering analysi$. The results were that the probability of control rod misoperation (e.g., rod drop) was unchanged and the consequen:es thereof, wore not increased. Up to the point of suberiticality, the TS requiremeats regarding rod worths must be me However, running rods in at that point would not save a great deal of time. Therefore, the licensee developed the insertion sequence for when ,

the reactor was still critical, which would not exceed the rod worth requirements. A shutdown that would take about two hours with a standard red inseri. ion sequence, took about 30 minutes when modified. Not only was the time reduced, but the modified sequence also reduced the probability for out-of-segunce rod pulls, and reduced the challenges to the reactor protection syste The normal insertion aequence (reverse of the withd: awal sequence on a startup), was used up to the analyzed point where the modified sequence was approved for use. At that point, specified rods were run full in, or to some intermediate position ~

l where they were run in later in the sequence.

l -

l The inspector reviewed the available documentation and discussed the l modified sequence with the licensee. This method of reducing the time (

to bring the reactor to cold shutdown was successful, according to the licensee, and may be used for future planned shutdowns. A new analysis would be required for each shutdown, and would probably only be

,

considered af ter about mid-cycle. The inspector had no concerns regarding the ana!ysis and use of this method of rod insertio . Exit Meeting On August 24, 1988, the inspector helo an exit n'eeting with licensee representatives (denoted in Paragraph 1). Region I!! NRC Management also participated via teleconferenc The scope and findings of the inspection were summarized; the li ensee acknowledged the statements made by the NRC representatives with respect to the vio1Ation (Paragraph 2). The inspector also discussed the likely inforaational content of the inspection report with regard to documents or processes reviewed by the inspector during the inspection. The licensee did not identify any potential report material as proprietar i . .

- _ _ - _ - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _a