IR 05000341/1986026

Insp Rept 50-341/86-26 on 860729-0929.Violations Noted: Failure to Establish Firewatch as Required in Tech Specs & Failure to Implement Procedure

ML20213E400
Person / Time
Site:	Fermi
Issue date:	11/04/1986
From:	Wright G NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
Shared Package
ML20213E356	List: IR 05000341/1986026 ML20213E353 ML20213E388
References
50-341-86-26, IEB-86-001, IEB-86-002, IEB-86-1, IEB-86-2, NUDOCS 8611130146
Download: ML20213E400 (38)

Text

.

.

.

U.S. NUCLEAR REGULATORY COMMISSION

REGION III

Report No. 50-341/86026(DRP)

Docket No. 50-341 Operating License No. NPF-43 Licensee: Detroit Edison Company 2000 Second Avenue Detroit, MI 48226 Facility Name: Fermi 2 Inspection At: Fermi Site, Newport, MI Inspection Conducted: July 29 through September 29, 1986 Inspectors: R. W. DeFayette C. D. Anderson W. G. Rogers R. A. Becker M. E. Parker D. S. Brinkmann M. J. Farber P. L. Hartmann R. A. Kopriva M. D. Lynch T. S. Rotella L. E. Whitney

'

7{

Approved by: ..G Wr 411ef //

Reactor Projects Section 2C Date' '

Inspection Summary Inspection from July 29 through September 29, 1986 (Report No. 50-341/86026(DRP))

Areas Inspected: Routine, unannounced inspection by resident inspectors and by the augmented restart inspection team of previously identified violations, inspector identified items, allegations, regional requests, operational safety verification, onsite followup of events, emergency response, plant trips, IE Bulletins (IEB), TI2515 - IEB 8601, TI 2515 - IEB 86-02, monthly maintenance observation, monthly surveillance observation, LERs, degraded grid voltage calibration, sustained control room and plant observation, management meetings, and report revie Results: Two violations were identified (Paragraph 6.d - failure to establish firewatch as required in Tech. Specs., and Paragraph 13.e - failure to properly implement a procedure). Two unresolved items were identified (Paragraphs 6.b and 15.b) and four open items were identified (Paragraphs 6.c, 6.e, 13.a, 13.e, and 17.d).

8611130146 861104 PDR ADOCK 05000341 G PDR

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

__

..

.

DETAILS 1. Persons Contacted Detroit Edison Company:

• F. Agosti, Vice President, Nuclear Operations S. Booker, Assistant Maintenance Engineer L. Bregni, Compliance Engineer J. Conen, Licensing Engineer R. Eberhardt, Rad-Chem Engineer

• J. Leman, Superintendent, Maintenance and Modification

• R. Lenart, Plant Manager, Nuclear Production L. Lessor, Consultant to the Plant Manager, Nuclear Production R. May, Outage Management Engineer

• Miller, Supervisor, Operational Assurance S. Noetzel, General Director, Nuclear Engineering J. Nyquist, Supervisor, Independent Safety Engineering Group T. O'Keefe, Technical Engineer G. Overbeck, Superintendent, Operations J. Plona, Technical Engineer

• E. Preston, Operations Engineer W. Ripley, Assistant Operations Engineer, Administrative

• L. Schuerman, General Supervisor, Nuclear Engineering

• F. Sondgeroth, Engineer, Licensing

• B. R. Sylvia, Group Vice President, Nuclear Operations C. P. Sexauer, Nuclear Production Administrator G. Trahey, Director, Quality Assurance

• R. Wooley, Acting Supervisor, Licensing U.S. Nuclear Regulatory Commission:

• D. Brinkmann, Reactor inspector

• R. DeFayette, Regional Project Manager

• Farber, Reactor Inspector

• Parker, Resident Inspector

• Rogers, Senior Resident Inspector

• Denotes those who attended the exit meeting The inspectors also interviewed others of the licensee's staff during this inspectio . Followup on Violations (92702) (Closed) Violation (341/85013-03(DRP)): Contractor Maintenance

, Procedures not reviewed by the Onsite Review Organization. The inspector verified that appropriate administrative procedures were revised to clearly specify the review requirements for contractor proceou es and that these requirements had been communicated to site personnal through a Nuclear Operations Notice. The inspector reviewed the licensee's evaluation of the impact the use of the unapproved procedures had on plant hardware. The licensee concluded

. - . -. - . . - - .

.

.

~

that the unapproved procedures had no hardware impact. The inspector determined that the evaluation was acceptable. This item is considered close (Closed) Violation (341/85037-02(DRP)): Failure to Perform Intermediate Range Monitor Weekly Surveillances in a Timely Manne The inspector verified through observation and discussion with the surveillance scheduler that the surveillance performance forms include the critical completion dates, the plan w the day includes surveillances scheduled, overdue surveillances are placed on a board in the main planning conference room and a general calendar summary report is delivered to the NSS containing pertinent surveillance dates. The inspector determined through record review that the Instrument and Controls (I&C) foremen were informed of the importance of surveillance completion dates. This item is considered close (C1csed) Violation (341/85040-01(DRP)): Failure to Follow Procedures Associated With Pulling Control Rods. The licensee provided revised administrative controls concerning rod pulls and provided additional training on the revised controls and changes to the rod worth minimizer. The inspectors satisfactorily inspected these corrective actions in Inspection Reports No. 50-341/86016 and No. 50-341/8601 (Closed) Violation (341/85040-02(DRP)): Failure to Implement Appropriate Administrative Controls Concerning Control Room Activitie Licensee management provided additional direction to the NSSs, NASSs, SOAs, STAS, and reactor engineers with regard to their duties and responsibilitie Shift personnel were directed to improve log keeping activities. The NASS was assigned to the control room. lurnover requirements were enhanced. More frequent management overview of control room activities was instituted. The inspectors satisfactorily inspected these corrective actions in Inspection Reports No. 50-341/86016 and No. 50-341/8601 (Closed) Violation (341/85040-03(DRP)): Failure to take all the corrective actions associated with a significant condition adverse to quality. The licensee communicated the rod pull error and its consequences throughout the organizatio Key shift personnel received additional direction on their duties and responsibilitie The inspectors satisfactorily inspected these corrective actions in Inspection Reports No. 50-341/86016 and No. 50-341/8601 . Followup on Inspector Identified Items (92701) (Closed) Unresolved Item (341/85013-n2(DRP)): Alarming Standby Liquid Control (SLC) Tank Level Annunciator. This item concerned inadequate implementation of Engineering Design Packages (EDPs) in that the licensee had implemented an EDP to change the SLC tank level alarm setpoints without updating plant operating procedure The inspector also observed that when the SLC tank level alarmed high, the sequence-of-event recorder printed out the reactor vessel level low alarm instead of SLC tank level Hi/ Low alarm. The inspector reviewed Surveillance Procedure No. 24.000.03,

_ - _ _ _ _ _ _ _ _ _ _ _ _ ___________ ____ ___ ___________ _ ________ _

. . .

d

.

" Mode 5 Shiftly, Daily, Weekly, and Situation Required Surveillances," Alarm Response Procedure (ARP) 3D13-SLC, " Tank Level Hi/ Low," and Operating Procedure 23.139, " Standby Liquid Control System," to ensure these procedures were properly updated to reflect the EDP changes. The inspector also reviewed POM 12.000.64, Revision 7, "EDP Implementation Procedure," and POM 12.000.15, Revision 20, "PN-21 (work order) Processing," to ensure the licensee has an adequate program of updating plant procedures and documents prior to releasing plant equipment to service after implementing an EDP. These procedures require the licensee to review Nuclear Production programs and documents to determine if either a revision or posting is required prior to accepting the modification for service. Additionally, these procedures require the listing of documents which are required to be posted or revised in the PN-21 (work order) to ensure they will be updated prior to returning equipment to service. Finally, the inspector observed the successful performance by the licensee of a test to actuate control room annunciator 3D13-SLC tank level Hi/ Low and 3D56 reactor vessel level low to ensure that the control room CRT and sequence-of-event recorder printed out the proper alarm upon actuation of the annunciator. This item is considered close j 1 (Closed) Unresolved Item (341/85013-05(DRP)): Inoperable Standby Liquid Control (SLC) System Testable Check Valve. This item concerned post maintenance surveillance testing being performed to ensure the adjustments made to the limit switches on C41-F007 (inboard testable check valve) were correct. The work order and surveillance testing were initiated to fix dual indication on the actuator after both were performed and cleared, the problem still existe As identified in Open Item (341/86016-01(DRP)), the licensee has completed an engineering design package to remove the testable check valve feature on C41-F007 and C41-F006. This will eliminate the limit switch maintenance problems previously encountered. Concerning the clearing of the work order and surveillance test with the dual indication still existing, the licensee initiated a Deviation Event Report (DER) to document the concern. The DER indicated that the root cause was personnel error and that a memo was generated from operations to emphasize that surveillances of testable check valves should be monitored and that any discrepancies must be investigated promptl This memo was put in urgent required reading to ensure that all licensed operations personnel were aware of the concern. This item is considered close (Closed) Open Item (341/8504;-Ot(DRP)): Communication Within the Operations Department. The inst ctor witnessed appropriate and timely communications t3rr a, hor the operations department staff and shift personnel regardiag ple- aonditions, upcoming plant evolutions and knowledge of planned maintenance activitie Specific actions taken to improve communications included NSS attendance at planning meetings, full shift briefings and better utilization of standing and night orders. This item is considered close . ,

.

.

d. (Closed) Open Item (341/85043-02(ORP)): Team. Effort of the Operating Shift Personnel. The inspector has observed improved team effort as the result of placing the NASS im th'e control room, plant manager / operations superintendent meeting with all NSS/NASS/SOA personnel to explain their roles and responsibilities, and training shift personnel on the simulator as a team. This item is considered close e. (Closed) Open Item (341/85043-03(DRP)): Duty Station for STA and SOA During Major Plant Evolutions. To integrate the SOA/STA/ Reactor Engineer into shift activities, the licensee has discussed the roles and responsibilities of the SOAs/SlAs with the SOAs/ STAS, assigned the SOAs to a particular shift, assigned the reactor engineer staff

, to full shift. coverage and performed simulator training with the

'

SOAs, STAS and reactor engineers oresent prior to the plant startup in August 1986. Based on these actions, this item is considered close f. (Closed) Open Item (341/85043-04(DRP)): Weaknesses in administrative controls. The' licensee directed the advisor to the plant manager and operations superintendent to review logs and shift actions under the Reactor Operations Improvement Plan. The licensee has installed a limiting condition for operations status board in the control room. The licensee has revised Procedure No. 21.000.01, " Shift Operations and Control Room", to include parameter values on check sheet These actions deal with the weaknesses identifie This item is considered close g. (Closed) Open Item (341/85043-05(DRP)): The Control Room Information System (CRIS). The licensee has changed the dot color to orange, relocated the card file to a central location in the control roomland periodically audits the dots on the control board against the files. During the first week of August the inspector audited all the dots / files and found one discrepancy. The e

discrepancy was brought to the attention of the reactor operator and was immediately corrected. The inspector reaudited the system during September and found one discrepancy. The discrepancy was brought to the attention of the reactor operator and immediately corrected. The inspector considers CRIS an adequate system provided the amount of equipment out of service is not excessive. This item is considered close h. (Closed) Open Item (341/85043-06(DRP)): Work Order Processin The latest revision to Administrative Procedure No. POM 12.000.15,

"PN-21 Processing", specifically states the NSO shall review proposed maintenance and sign in the " Released for Repair by" block.

s This matter is considered closed, i. (Closed) Open Item (341/85043-07(DRP)): Region III Concurrence Prior to Exceeding 5% Reactor Power. In a letter dated September 12, 1986, the NRC Region III Regional Administrator authorized the licensee to increase reactor power in excess of 5%

but no greater than 20%. This item is considered closed. With the closure of this item all items contained in CAL-RIII-85-10 are complete and the CAL is considered close . ._ . . _ -_ _ . . _ . . _ .

.

s

.. (Closed) Unresolved Item (341/85048-01(DRP)): Timeliness in-

'

Declaration of System Inoperability. The licensee has revised the DER procedure to require the notification of the NSS within one hour of a DER being approved by the initiator's supervisor. This

~c hange from previous practices should assure timely notification and declaration of system inoperability where appropriate. This new administrative requirement was established with the issuance of Procedure No. 12.000.52, " Deviation and Corrective Action Reporting."

g This item is considered close (Closed) Open Item (341/86016-01(DRP)):

' Standby Liquid Control Testable Check Valves. This item concerned the lack of indication on Testable Check Valve No. C41-F007. The check valve disk position

'

indicator did not indicate "open" during the 18 month surveillance.

'

As a result of problems encountered with the testable check valve feature, the licensee implemented an Engineering Design Package (EDP)-1495.to remove the testable check option on check valves C41-F006 and C41-F007. The purpose of the EDP was to eliminate.the valve packing leaks and limit switch maintenance problems previously encountered. The inspector noted that this EDP has been field completed, and that on July 22, 1986, the licensee implemented L action per Deco Memo VP-86-0081, for relief on the inservice testing program for these valves. This request provides an alternative method of testing which must be approved by NRR. The licensee has also taken action to modify the FSAR by initiating an FSAR Change

'

Notice (FCN)-85-114 to modify applicable steps addressing the testable feature of these check valves. This item is considered

closed.

) (Closed) Open Item (341/86022-02(DRS)): Definition of " annual." ,

'

i This item addressed the clarification by the Office of Nuclear

. Reactor Regulation with regards to the definition of the term

" annual" as it pertained to licensed operator requalification both

by written examination and performance of annual required control i manipulations. By memo from W. T. Russel to C. J. Paperiello dated June 23, 1986, annual control manipulations are expected to be l completed once in each 12 month period as opposed to once per 12 '

! month interval. As a result, it appears that the current practice of permitting as long as 23 months to lapse between performance of required control manipulations stipulated on an " annual" basis is acceptable to NRR. This item is close ,

! Allegations (99204B)

(Closed) Allegation AMS-RIII-85-A-0151: Letter to Public Forum in i- Lafayette, Louisiana, Newspaper. This allegation resulted from a letter to the "Public Forum" in the Lafayette, Louisiana, newspaper, Advertiser, which made several charges regarding four nuclear sites

! including Fermi 2. The letter alleged in general terms the

alteration of certification documents for site personnel, improper i non-destructive examination inspections, and use of non-calibrated

, non-destructive evaluation (NDE) inspection machines. NRC Region III personnel contacted the alleger on two different occasions in an

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

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

.

.

attempt to obtain any specific information pertaining to the Fermi 2 site but the alleger was unable to provide any such informatio The individual indicated his information was not relevant to the Fermi Plant. The individual was requested to provide specific information about the remaining nuclear plants, but was not able to do so. Based on this lack of specificity, this allegation is close b. (Closed) Allegation AMS-RIII-86-A-0082: Personal Papers of Former Deco Vice President. This allegation stated that the personal papers of the former vice president of Detroit Edison Company concerning Fermi I contained evidence pertaining to the credibility of the Detroit Edison Company chief executive officer and management and that papers and notes found in the Monroe County Library substantiated this claim. A representative of Region III reviewed the Fermi 1 LPDR records in the Monroe County Public Library. No personal records or notes were found. Regardless of the above findings the question of management credibility is already under review by the NRC (further information can be found in inspection reports and correspondence related to the inadvertent criticality event of July 1,1985).

.

Based on the above information, this allegation is close c. (Closed) Allegation AMS-RIII-86-0095. Resume With Potentially False Information. This allegation was received by the Senior Resident Inspector who was told by an alleger that his resume may have been sent to Detroit Edison as part of a bid package for work from the alleger's company. The alleger stated that the resume may contain false information overstating his qualification The Senior Resident Inspector investigated whether DECO received a proposal or the resume in question and determined that neither had not been received. The allegation does not apply to Fermi 2 and is close d. (Closed) Allegation AMS-RIII-86-A-0105: Violation of Stop Work Order. The alleger stated that he had seen information that a Detroit Edison Company employee was fired because he had a stop work order on his desk for several days and during that time people were working on valves that potentially could have harmed the Region III reviewed this matter and determined that this was an electrical industrial hazards matter related to potential harm to workers from manipulating the value in question while performing the maintenanc It did not involve a radiological hazard and therefore does not fall within the regulatory authority of the NR Therefore, the allegation is close e. (Closed) Allegation AMS RIII-86-A-0106: Effect of Quarry on Ferm This allegation stated that a large quarry recently had been granted authorization to expand its operations nearer to the Fermi 2 site, and that blasting from the new quarry, which would be about 8000 feet from the site, may have an impact on the safety of the Fermi 2 facilit _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .

, ,

O

- .

A review of the Fermi 2 Final Safety Analysis Report (FSAR)

indicated that the nearest quarry to Fermi 2 at the time of application by Detroit Edison Company (DECO) was about eight miles from the site and that it posed no hazard to Fermi 2. Subsequent to the filing of the original application; however, a new quarry, Rockwood Stone Quarry, was opened in 1982 and is located about three miles NNE of the safety-related structures of Fermi 2. The Monroe County Office of Planning informed DECO that only one other quarry is being planned in the vicinity of Fermi 2, and that would be about six miles NNW of the site. The project, however, appears to be years away from operatio The NRC requested DECO to evaluate the effects of the Rockwood Stone Quarry on Fermi 2. Using NRC guidance, the conclusion was that there would be no effect even considering a worst-case accident which would involve the simultaneous explosion of the maximum potential inventory of explosives on site coincident with the arrival on site of a truck with a full replacement load of explosive DECO's evaluation also considered the nearest approach of the truck to Fermi 2 while transporting the explosives to the site, ground effects of routine blasting, effects on ground water, and effects of the quarry on the site emergency plan. DECO will ,

formalize this analysis and incorporate it into the next update of the FSA The NRC also contacted the owners of the Rockwood Stone Quarry and inquired if they have plans for further expansion of their quarr They confirmed that recently they received permission to expand their operation about 2600 feet in a westerly direction (which might bring them slightly closer to the site since they are located NNE of the site) but there are no plans to expand within 8000 feet of the sit They also stated they always had this legal right and that the recent approval was based on environmental considerations, not legal ones, since they already owned the property. Furthermore, they confirmed that the information used by DECO in its evaluation of the effects of the quarry included this 2600 feet expansio Based on this information, this allegation is close . Followup on Regisnal Requests (92705)

During the inspection period NRC Region III management identified specific areas of review. These areas were designated based on licensee identified or inspector identified concerns at other sites which may have generic significance. The areas and the inspections in those areas are described belo As a result of concerns that were identified at the Monticello Nuclear Plant regarding Low Pressure Coolant Injection (LPCI) loop logic circuit timing, the inspector revitwed the licensee's response to information received by the Fermi 2 General Electric (GE) site staff regarding this problem. The problem was that because of a

. --

.

.

" relay race" between break detection logic and loop selection logic, the faulted primary coolant loop could be selected in a loss of coolant accident. The Fermi GE site staff reviewed this information and concluded the problem with logic circuit relay timing did not pertain to Fermi since the 1/2 second time delay required in relay 10A-40A was presen The inspector concluded the GE staff assessment was correct and the problems associated with logic circuit timing of the LPCI loop select logic identified at Monticello did not apply to the Fermi 2 plan b. As a result of concerns identified at the Perry Nuclear Power Plaat, the resident inspectors were requested to review holddown mechanisms for the control rod drive hydraulic control unit and to determine if torquing specifications were identified in installation instructions. Through discussions with technical group personnel and review of Drawings No. 6I721-2113-6 and 6I721-2113-7, the inspector was able to determine that the licensee had modified the holddown devices to provide additional support. These drawings had been modified by Deviation Disposition Request (DDR) 1400B, which modified the holddown and changed bolt torquing values. Specified torque values were 40 foot pounds. Review of the completed process control sheet for Drawing 919D615 identified that a Quality Coatrol inspector had verified bolt torquing requirements for the holddown devices on the CRD hydraulic control unit c. During the week of September 15, 1986, the NRC Region III Section Chief requested information as to whether the field breakers on the MG reactor recirculation pumps were GE-AKF-25s because of the failure of this type breaker to trip at the Pilgrim Nuclear Plant. The inspector ascertained through discussion with the licensee that the breakers were this type and that no problems had occurred with these breakers failing to tri d. Due to recent concerns regarding the reliability of seismic monitoring instrumentation, the NRC Regional Offices have been asked to conduct periodic inspections of maintenance and testing of such instrumentation. Specifically the resident inspectors were requested to provide a description of the seismic monitoring instrumentation, applicable Technical Specification requirements, preventative maintenance performed, non-Technical Specification surveillance performed, and failure data for the last 24 months. This information was provided to Region III for evaluation by the NRC office of Inspection and Enforcemen e. Part 21 Report on Static "0" Ring Inc. Pressure Switche The inspectors were asked to ensure the licensee had received and reviewed the subject Part 21 Report. The inspector subsequently provided a copy of the Part 21 to the licensee to determine if the Part 21 was applicable to Fermi 2. The licensee, through a previous review of the Master Instrument List (MIL) for Bulletin No. 86-02, was able to determine that the specific SOR pressure switches

.- - - . . . . , . ..

F

.

are not used on systems important to safety at the Fermi 2 facilit In DECO memo, Wooden to File, dated September 30, 1986, the licensee noted that no SOR pressure switches, beginning with part numbers 1, 8, or 9 were listed on the MIL. The licensee also indicated that it has not purchased any SOR pressure switches after January 1983. The Part 21 is applicable to those switches purchased after January 198 The inspector also performed an independent check of the MIL to confirm that the SOR pressure switches referenced in the Part 21 Report were not used at the Fermi sit No violations or deviations were identified in this are . Operational Safety Verification (71707)

The inspectors observed control room operations, reviewed applicable logs and conducted discussions with control-room operators and operations staff during the period from July 29, 1986 through September 29, 198 The inspectors performed walkdowns of the control boards to verify proper safety system valve position, pump status, and expected instrument readings. The inspectors reviewed out of service logs to determine whether the out of service equipment was identified and the appropriate Technical Specification action statements were entered. Tours of the reactor building and turbine building were conducted to observe plant i equipment conditions, including potential fire hazards, fluid leaks, and excessive vibrations and to verify that maintenance requests had been initiated for equipment in need of maintenanc The inspectors, by observation and direct interview, verified that the physical security plan was being implemented in accordance with the station security pla The inspectors observed plant housekeeping / cleanliness conditions and verified implementation of radiation protection controls. During the inspection, the inspectors walked down the accessible portions of the High Pressure Coolant Injection System to verify operability by comparing system lineup with plant drawings, as-built configuration or present valve lineup lists; observing equipment conditions that could degrade performance; and verified that instrumentation was properly valved, functioning, and calibrate These reviews and observations were conducted to verify that facility operations were in conformance with the requirements established under Technical Specifications, 10 CFR, and Administrative Procedures. During these operations reviews a number of situations required additional followup by the inspectors. Those situations and the results of the followups are presented belo On July 28, 1986, at 0300, the licensee, while trying to draw a vacuum in the main condenser, was unable to draw down below 8 - 9 psi The operators also noted that the secondary containment pressure decreased. This led the licensee to believe a path existed between the secondary containment and the condense The licensee

_ _ - _ ._ . _ - - -

.

.

subsequently determined that a four inch valve (G41-F026) and a ten-inch valve (G41-F001) were open, providing a direct path from the fifth floor dryer separator storage pit area to the main condense The licensee closed these valves and main condenser vacuum decreased to 1.2 psig vacuum. The licensee then issued a Deviation Event Report (DER) due to a potential secondary containment LC0 violation. Nuclear Engineering was given responsibility for action of the DER. Initially the licensee determined that these valves were closed during core alterations, but were opened later when the CS and LPCI pump CMC switches were placed on OFF position, thus requiring secondary containment integrity. The licensee took immediate action to close these valves and place permanent tags on these valves and other penetrations with similar impact, cautioning operators that these valves are secondary containment boundary valve Nuclear Engineering has determined that a twenty foot loop seal exists on these penetrations that provides an 8.66 psi seal to the four inch line. Operations indicated that the secondary containment fifth floor blowout panels are designed for 1.5 psi. Therefore, the licensee believes that the loop seal provides adequate isolation of this penetratio The inspector reviewed the situation and concluded that the licensee had not violated secondary containment requirement b. On August 29, 1986, the licensee attempted to perform routine monthly surveillance testing of Emergency Diesel Generator 13. When the EDG was started, the control logic automatically started EDG service water pump B. However, the service water pump's feeder breaker tripped open stopping the service water pump. The EDG was declared inoperable and the appropriate Limiting Condition for Operation action statement was performed. Subsequent investigation into why the breaker tripped open revealed that one of the three primary power fuses associated with the service water pump had blown. Also, the blown fuse's amperage rating was not in accordance with the fuse rating design document for that particular positio Specifically, an 80 amp fuse was installed where a 100 amp fuse was ;

specified. The licensee conducted an inspection of all accessible fuses associated with EDG 13. The inspection was completed late in the afternoon of August 29, 1986, and revealed five other fuse l

'

discrepancies from the design document. Four of these discrepancies were determined to be with the design document and not with the fuses installed in the field. The fifth discrepancy was a 100 amp fuse installed where an 80 amp fuse was specified. In addition to 1 the inspection conducted on EDG 13, the licensee conducted inspections !

on the other three EDG fuses. The inspection of the other EDGs did l not uncover any other field installation problem The licensee reviewed the historical records on EDG 13 maintenanc The records review revealed that on June 16, 1986 the fuses for 14 fuse positions were removed under an abnormal lineup sheet to begin l

11 l l

l

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

.

.

an 18 month maintenance outage for Emergency Diesel Generator (EDG) 1 Restoration of EDG 13 was completed and the EDG monthly start and load test was successfully performed on July 6,1986. The EDG was returned to service on July 7, 1986. During the restoration process the fuses were reinstalled with a 100 amp fuse installed in an 80 amp location for breaker 72EC-2C position SA (electrical feed to the EDG skid for auxiliary equipment) and an 80 amp fuse installed in a 100 amp location for breaker 72EC-2C position IE (electrical feed to the EDGSW pump B). The electrical lineup of the EDG General Operating Procedure (GOP) was used as the independent verification to verify the system was properly returned to service. However, the electrical lineup only checks the EDGSW pump breaker position and not whether the proper fuses were installe The EDGSW pump was run to support EDG 13 four times before the EDGSW pump tripped on high starting amps. This indicates that the EDGSW/EDG may have performed its safety function during certain instances when called upon. The licensee's engineering department analyzed the installation of the 80 amp fuse. The analysis concluded that the fuse should have withstood a normal start of the pump. The basis of the conclusion is this type of 80 amp fuse would take 10 seconds to blow at locked rotor amps of 362A and it takes 3 seconds for the pump motor to reach full speed. Therefore, the fuse should not blow on locked rotor amps. Also, the fuse should not have blown on full load amps since normal running current of the pump motor is 57 The inspector attributed the root causes of this situation to be inattention to detail by the operator during the fuse reinstallation activities and inadequate controls in the independent verification process for returning equipment to servic A determination as to the regulatory impact this situation shall be made by NRC Region III. Until that decision is made, this matter is considered unresolved (341/86026-01(DRP)).

c. During the inspection period the inspector or the licensee identified discrepancies between installed fuse size and the fuse specifications. The events identifying the discrepancies were:

(1) The inspectors compared the fuses actually installed in the circuit breakers of Valves E41-F001, F004, and F012 against Specification 3071-128 EJ, the red-lined design document on fusing in the plant. The licensee has designated certain specifications, drawings, etc., as documents which shall be current without referral to the outstanding engineering changes through a process of red-lining the affected documents at the time of closecut of the work package. The comparison revealed a discrepancy. Specification 3071-128 EJ designates a 20 amp primary power fuse for valve F004 whereas the inspector observed a 15 amp fuse installed. The inspector brought the difference to the licensee's attention. The licensee reviewed

.

.

'

the engineering changes (EDP) outstanding on Specification EJ and identified that EDP 1424 replaced the 20 amp fuse with a 15 amp fuse. The inspector confirmed the change through review of the EDP. The inspector ascertained through discussion with the licensee that Specification EJ became a red-lined document in March of 1986. The actual EDP was implemented in November of 1985. Therefore, when the work package was closed out, the specification was not required to be red-line (2) On September 13, 1986, while performing preventative maintenance tagouts of the reactor core isolation cooling system room cooler, the licensee found a fuse installed which did not agree with Specification E Further investigation by the licensee revealed that the fuse had been changed under an engineering design change and Specification EJ had not been update (3) While performing the walkdown of the fuses associated with the emergency diesel generators (see Section b of this paragraph)

the licensee noted that Specification EJ identified one size fuse for the primary bus pots and another size was installe Further review by the licensee determined that the installed fuse was of the correct siz The inspector requested the licensee to assure that Specification EJ is correc The licensee committed to provide that assurance. To implement that commitment the licensee has reviewed Specification EJ against any outstanding changes. All accessible fuse locations of safety-related equipment shall be walked down with completion of the walkdowns to be by November 16, 1986, and any fuse determinations which will have to be done after deenergization of the panel shall be done at the next maintenance deenergization of said panel. All discrepancies shall be documented on a DER. Completion of these tasks is considered an open item (341/86026-02(DRP)).

d. On September 8, 1986, at 8:00 p.m. while performing POM 24.501.17,

" Sprinkler System Simulated Automatic Actuation Test," the diesel fire pump room wet sprinkler system failed to annunciate. The test was performed to satisfy Technical Specification (TS) Surveillance 4.7.7.2 which requires that the sprinkler system be demonstrated operable at least once per 18 months by performing a system functional test which includes simulated automatic actuation of each system by opening the inspector's test valve and verifying the water flow alarm annunciator alarms. TS 3.7.7.2 action statement requires an hourly firs watch to be established with one or more of the required sprinkler systems inoperabl On September 8, the licensee failed to realize operation of the water flow annunciator in the control room was a TS requirement and as such failed to post a fire watch. On September 9, 1986, during a subsequent review of the surveillance the licensee identified the

- _ _ - - _ _ .

-. _ _ - - - - -

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

.

.

error and at 11:55 a.m., approximately 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> after the diesel fire pump room wet sprinkler failed the surveillance, a fire watch was established. This is considered a violation of TS 3.7. (50-341/86026-03(DRP)).

In reviewing the circumstances surrounding this violation, the inspector determined that per the technical specifications the system was inoperable, due to failure of the system to pass the TS surveillance. However, the malfunction of the annunciator would not actually affect operation of the diesel fire pump room wet sprinkler. The system would still function as intended; however, the control room would not be aware and, as such, would not be able to respond and dispatch the fire brigade to a fire in a timely manne e. The licensee informed the inspector of a potential violation of a Limiting Condition for Operation associated with the Division 1 thermal hydrogen recombiner. DER 86-119, initiated on September 16, 1986, stated that a thermocouple essential to the operability of the recombiner had its leads reversed during restoration activities of 18 month surveillance Procedure No. 44.220.01 on June 29, 1986, and that this condition was discovered during performance of 6 month Surveillance Procedure No. 24.409.01. Upon contacting the cognizant maintenance engineer for the recombiners, the inspector ascertained that the performance of the 6 month surveillance procedure and discovery of the reversed thermocouple leads had been done on July 17, 1986. Therefore, the recombiner was operable when the licensee ascended to a mode requiring the recombiner in August of 198 Further discussion revealed that the DER was initiated due to quality assurance department involvement in the surveillance package closecut. The inspector reviewed the two surveillance procedures and substantiated the maintenance engineer's statements. The results of the inspector's review of this situation were:

(1) The 6 month surveillance procedure was completed on July 17, 1986 and a Limiting Condition for Operation had not been violate (2) The 18 month surveillance procedure does not utilize an independent verification for the restoration of the thermocouple leads disconnected during the testin (3) Since the thermocouple leads are disconnected during the channel calibration, there is not a calibration of the senso Additional calibrations of Technical Specification temperature instruments were reviewed with the same result that sensors are disconnected from the instrument string during calibration activitie (4) The documentation of the thermocouple reversal was untimely and reflects a lack of understanding of the DER process on the part of personnel involve .

.

The inspector informed senior licensee management during the exit that the potential existed that all safety related temperature sensing devices may not be appropriately calibrated. The inspector will followup on this Open Item (341/86029-09(DRP)).

f. On September 16, 1986, the licensee informed the inspector that an invalid hydrogen sample of the off gas system had been taken at 0500 on September 16, 1986. Technical Specification Limiting Condition for Operation 3.3.7.12 requires a hydrogen sample of the offgas delay piping every 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> while the hydrogen monitor is out of servic The validity of the sample was suspect based on the lack of flow through a meter which should have been representative of the flow at the sample point. Flow through the meter had been stopped by the closure of a valve in the offgas system. The valve was closed to support condenser air in-leakage investigations. Later that week the licensee informed the inspector that the sample was actually a good sample and though the meter showed no flow there was adequate gas flow at the sample poin The results of the inspector's review of this situation were:

(1) The sampling point did have an adequate flow path for the 0500 sample to be valid even with the meter showing no flo Therefore, there was not a violation of a Limiting Condition for Operatio (2) The sampling personnel involved should have immediately identified what they thought was an invalid sample to the NSS and then retaken the sampl (3) Followup by the licensre on all the facts surrounding this event should have been more timel (4) Shift personnel should have been more aware of the consequences of closing the offgas valv g. During mid-August 1986 the licensee experienced two inoperable Hydraulic Control Units (HCU) accumulators. The licensee took the Limiting Condition for Operations actions associated with HCU accumulators exclusively. One hour after the second accumulator became inoperable the first accumulator was returned to an operable statu However, a question arose as to whether the Limiting Condition for Operation associated with inoperable control rods also applied when two accumulators are inoperable. The inspector requested the licensee provide guidance to on shift personnel in this area and/or explore changing the wording in the Technical Specifications with regard to the applicable Technical Specification No other violations or deviations were identified in this are .- . ._- - - - _ _ . .-- _ _ - _ - -_ . - __

.

.

'7. Followup of Events (93702)

During the inspection period, the licensee experienced several events which required prompt notification to the NRC pursuant to 10 CFR 50.7 The inspectors pursued the events onsite with licensee and/or other NRC official In each case, the inspectors verified that the notification was correct and timely, if appropriate, that the licensee entered into the emergency plan, emergency systems were reset and returned to standby when parameters permitted such actions, conditions adverse to quality were identified with corrective actions initiated, and activities were conducted within regulatory requirements. The specific events are as follows:

. July 31, 1986 - Reactor Building Ventilation Exhaust Radiation monitor tri . August 6, 1986 - Control Center HVAC shifted into chlorine mod . August 6, 1986 - Fire in motor control center D.C. electrical pane Further response and followup to the fire is discussed in Paragraph . August 13, 1986 - Potential inability of four containment isolation valves to automatically clos Further followup of this condition was performed by a region based inspector as documented in Inspection Report No. 50-341/8602 . August 23 & 26, 1986 - Automatic closure of HPCI valve E41F00 Further followup on these two automatic closures is discussed in Paragraph 1 . August 29, 1986 - Reactor SCRAM on high steam pressure. Further followup of the events leading to the SCRAM are discussed in Paragraph 17. Post-SCRAM response is discussed in Paragraph . August 27, 1986 - Emergency diesel lube oil filter inspection not performed quarterly. Further followup on the missed inspections is discussed in Paragraph 1 . August 31, 1986 - Reactor shutdown due to Limiting Condition for Operation. Further followup of the equipment out of service and the repair activities is discussed in Paragraph 1 . September 15, 1986 - Automatic closure of HPCI valve E41-F06 No violations or deviations were identified in this are . Emergency Plan Response (93702)

e

.

On August 6, 1986, at 5:30 p.m., the licensee declared an " alert" due to a fire in an electrical panel on the third floor of the auxiliary building. Three NRC inspectors were on site at the time and were able to observe the licensee's response closely. The Senior Resident Inspector (SRI) was in the Control Room when the annunciator trouble light came on and followed a reactor operator to the electrical panel where the fire was located. The oparator attempted to extinguish the fire with a CO

portable fire extinguisher, but before he could do so the CARD 0X system activated, forcing him, the SRI, and others who responded to leave the room. The fire brigade quickly responded and assured the fire was ou The licensee's response to the emergency was timely and satisfactor The inspectors noted that the control room staff performed in a calm and professional manner not only in assessing the emergency and activating the emergency plan, but also in maintaining and controlling the reactor which was operating at the time. The Technical Support Center (TSC) was activated and one of the inspectors went to that facility to observe licensee performance and to activate the NRC desk. Licensee performance also was calm and professional at that location. All major positions were staffed promptly, data was collected and disseminated, and proper notifications made to the NRC, Monroe County, Canada, INP0, and the American Nuclear Insurers (ANI). Status boards containing information on the plant, on environmental conditions, and on radioactivity measurements were updated routinely. The Plant Manager was the senior licensee management person in charge of the TSC and he kept the staff of the TSC informed on conditions by periodically making general announcements including one at about 6:50 p.m. that the CO2had been cleared from the

auxiliary building and a team had been assembled to assess the damage.

j The fire was determined to have been in a breaker controlling a minimum flow valve for the HPCI system. Although this system was not required

'

for operational conditions in effect at the time (about 1% power and 135 psig) the licensee decided to shut the reactor down until the cause of the fire was determined. Therefore, a controlled shutdown was begun at approximately 9:30 p.m. The reactor was not in danger and no radioactivity was released as a result of the fir Subsequent analysis determined the fire originated in the closing contractor of the breake . Plant Trips Following the plant trip on August 29, 1986, the inspector ascertained the status of the reactor and safety systems by observation of control room indicators and discussions with licensee personnel concerning plant parameters, emergency system status and reactor coolant chemistry. The inspectors verified the establishment of proper communications and

,

reviewed the corrective actions taken by the licensee.

All systems responded as expected, and the plant was returned to operation on August 30, 198 No violations or deviations were identifie _ _ ___ _ __-___. _-_ ____ ____ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ - _ - - .

.

'

.

10. IE Bulletin Followup (92703)

Each of the following IE Bulletins was reviewed by the resident inspectors to determine if: (1) the licensee's written response was submitted within the time limitations stated in the bulletin, (2) the written response included all information required to be reported, (3) the written response included adequate corrective action commitments based on information presented in the bulletin and the licensee's response, (4) licensee management forwarded copies of the written response to the required onsite management representatives, (5) information discussed in the licensee's response was accurate, and (6) the corrective action taken was as described in the respons . (Closed) IE Bulletin 86-01, Minimum Flow Logic Problems that Could Disable RHR Pump . (Closed) IE Bulletin 86-02, Static "0" Ring Differential Pressure Switche See Paragraph 11 and 12 for additional inspection actions associated with IEB 86-01 and IEB 86-02 respectivel No violations or deviations were identified in this are . Inspection of TI 2515/82 Associated with IEB 86-01, " Minimum Flow Logic Problems that Could Disable Residual Heat Removal Pumps" (25581)

The inspector reviewed TI 2515/82 and determined that the review conducted on IEB 86-01 in the previous Inspection Report No. 50-341/86019, satisfied the inspection requirement No violations or deviations were identifie . Inspection of TI 2515/81 Associated with IEB 86-02, " Static 'O' Ring Differential Pressure Switches" (25581)

IEB 86-02 identified erratic setpoint drift in Static "0" Ring (SOR)

differential pressure switches which could affecting the reactor protection system, emergency core cooling systems, primary containment isolation systems, and other engineered safety feature systems. As a result of concerns identified in licensee responses to the bulletin, Temporary Instruction TI 2515/81 was issued and assigned to NRR and the resident inspectors. The TI requested the inspectors to review specific actions of the bulletin and to review the licensee's responses to assure that the licensee addressed all systems important to safety as defined in 10 CFR 50.4 Review of DECO letter VP-86-0111, dated July 28, 1986, indicated that, based on a thorough review, DECO found no Series 102 or 103 SOR differential pressure switches in use at Fermi The inspector reviewed the Master Instrument List (MIL) for all applications and types of SOR switches in use at Fermi 2. This review identified three SOR

.

.

differential pressure switches in use at the present time of which none were Series 102 or 103. The three switches were Model No. 15R3 used in the offgas syste The bulletin requested the licensee to respond on the extent to which SOR Model No. 102 or No. 103 differential pressure switches are installed or planned to be installed as equipment important to safety. The licensee's response did not address any planned use of these switches. The licensee has revised its response to state that there are no planned uses for SOR Model No. 102 or No. 103 switche As a result of no Series 102 or.103 SOR differential pressure switches in use at Fermi 2, TI 2515/81 is close No violations or deviations were identifie . Maintenance Observation (62703)

The startup team has had a concern with the Fermi maintenance progra This concern developed from, among other things, repeat problems with several systems during the heatup phase testing. For example, there were recurring problems with HPCI, the startup level controller, south reactor feed pump, and steam jet air ejectors. Although the problems were corrected, these and similar problems will undoubtedly continue to occur as unused systems are operated under real conditions for the first tim Although individually the problems may not be significant, collectively they may be indicative of a weakness in the maintenance are The licensee has been advised that additional attention is needed in this area to assure that these and similar problems are effectively resolve Aggressive handling and prioritization will be required in the maintenance area and in the Instrumentation and Control Group during and beyond Test Condition 1 (20% power hold point). The team will continue to monitor this area closel Station maintenance activities of safety-related and important to safety systems and components discussed below were observed by the restart team to ascertain that they were conducted in accordance with approved procedures, regulatory guides and industry codes or standards, and in conformance with technical specification There was an unscheduled reactor shutdown on September 2, 1986, to repair a malfunctioning torus-to-reactor-building vacuum breaker isolation valve (also see Subsection "e" of this paragraph). The problem was determined to be binding of the valve shaft with a fiberglass bearing (bushing). The inspectors witnessed portion of the disassembly and repair of the valve and noted that the local leak rate test was completed successfully. The reactor was restarted on September This was the third valve of this type to have similar problems (one was in September,1984, and another in November,1985).

The licensee has requested it's Nuclear Engineering Department to evaluate a different type of bearing material and/or a possible reorientation of the valve. The vendor also has recommended the use

~

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

- - - - . - _

.

.

of stainless steel components in the valve assembly to minimize corrosion. In the short term, the licensee is trending valve stroke times as an indicator of pending problems. Because of their concern about future failures, the inspectors have requested the Division of Reactor Safety in Region III to review this matter. It will, therefore, be carried as an open item until resolved (341/86026-04(DRP)).

b. On September 16, 1986, Intermediate Range Monitor (IRM) "D" would not retract fully from the core. The inspectors accompanied an Instrument and Controls (I&C) technician into the reactor building to check the fuse panel controlling this IRM. The technician located the panel and identified three fuses that had blown. The inspector verified that all three fuses were of proper size (1 amp).

The inspector then accompanied the technician to a locked QA cavinet where proper replacement fuses normally are kept, but none were available. The technician prepared paper work to obtain new fuse Subsequently, the licensee replaced the fuses and again attempted to operate IRM "D". It would insert but could not be withdrawn fully and the fuses again blew. Even though the mode switch was in the

"Run" position and the IRMs were bypassed automatically, the licensee declared the IRM inoperabl On September 20, 1986, after power had been reduced to about 6%, the licensee made a drywell entry to investigate the cause of the IRM failure. A slight misalignment between the IRM drive mechanism and the under-core platform was detected such that the detector was rubbing against the side of one of the openings in the platfor The platform was rotated slightly to realign the opening with the IRM, the fuses were replaced, and the IRM was satisfactorily reteste c. Extensive main,tenance and repairs were performed on the south reactor feed pump turbine during the Fermi 2 extended outage because of vibrations, which heavily damaged it in 1985 (reference Enclosure 2 in Inspection Report No. 50-341/86005). Consequently, the licenset instituted an extensive " shakedown" period for the repaired turbine after the reactor was restarted and steam was available to the turbine. Much of the testing involved measuring movement of the turbine and monitoring vibrations during operatio The licensee determined that there is vertical movement of between 0.040 inches and 0.050 inches at the low pressure end of the turbine, but that this movement also occurs on the north feed pump turbine which experienced no operational problems last year. Extensive work

-

was performed to reduce or eliminate this vertical lift but it could not be reduced below about 0.043 inches. Both the licensee and the vendor (DeLaval, Inc.) agree that this movement is within the tolerance of 0.100 inches, but that efforts should be continued to reduce it. After reactor restart, the turbine was run with no unacceptable vibrations as measured by several vibration detectors attached to the turbin .

. .

After this initial testing was completed on the south feed pump turbine, it was coupled to the pump. During the week of August 25, ?

1986, while preparing to switch reactor feedwater control from the north feed pump to the south feed pump, the south feed pump tripped on overspeed. The problem was determined to be a malfunctioning tachometer which had to be replaced. The south feed pump then was put on l'.ne where it ran successfully for several days until September I when the reactor operators shifted to the north feed pump due to a 0.0013 vibration alarm on the south feed pump couplin This was determined to be caused by a calibration problem with the

'

vibration detector. The detector, was subsequently reset to the proper setting On September 17, leaks developed in the south reactor feed pump minimum flow line drain valve. The inspector observed the leaks and not"d they were coming from the weld where the vertical drain line was welded to the minimum flow pipe. The licensee changed to the north feed pump and repaired the lea It appears to the inspectors that the vibration problems with the south reactor feed pump turbine have been reduced to acceptable limits. However, the licensee will continue to monitor both feed pump turbines as reactor power in increase d. On August 1, 1986, just prior to reactor startup, the licensee notified the NRC of a leak in a valve (N20-F618) in the bypass line of the minimum flow line from the condensate demineralizers to the condenser. The valve was not isolable; therefore, the system had to be drained to make repairs. Subsequent observation by the licensee and by the inspectors noted that there was a thin 10 inch

'

crack extending about 150' around the bonnet of the valve near the stem. The inspectors also observed after the valve had been disassembled, that the seat of the valve was scoured and the guide pin had broken off. The valve body had considerable pitting. A mechanic was observed making thickness measurements on the valve body to assure the pitting had not eroded the metal to such an extent that it was unusabl The inspectors inquired as to how the valve failed and were told that it was probably caused by vibrations. Furthermore, the Limitorque operator for the valve, which weighs approximately 400-500 pounds, hangs horizontally on the end of the bonnet causing additional stresses. After the valve had been repaired with a new bonnet and disk and the system operated again, the inspectors did note considerable vibration in the line. Much of this may be caused by the design of the line which contains two 90' bends in a very short distance. As a temporary corrective action the licensee added supports to the bonnet and operator which reduced the vibration The licensee has requested the Engineering Department to investigate redesigning the line to reduce the vibrations.

l

.

e The inspectors also were told that part of the reason for the failure may be that the line in question was run for periods of time for which it was not designed. As noted, the F618 valve is a bypass valve which normally is not used because flow is controlled by a throttle valve (F404). However, the throttle valve had been out of service for several weeks, thus requiring the use of the bypass valve. Shortly after the bypass valve was repaired; therefore, the licensee repaired the throttle valve and was able to close the bypass valve. The throttle valve repair consisted basically of replacing the "0" ring sea During the week of September 1, 1986, the inspectors observed repairs being made to the isolation valve for a torus to reactor building vacuum breaker. The valve had been declared inoperable on August 31 when it failed in the closed position, causing the reactor to enter a 72-hour Limiting Condition for Operation (LCO). The inspectors were present when disassembly commenced and noticed that a proper work order had been written and was present at the job site. However, before the final two bolts of the valve were removed, the maintenance personnel waited for a Health Physics (HP)

technician to arrive because a statement on the work order directed that the system should not be breached without the presence of such

<

an individual. Upon arrival at the job site, the HP technician stopped all work and directed that the area be roped off and all workers dress in anti-contamination clothing. His reason for doing this was that a specific radiation work permit (RWP) had been written calling for such protective measures in case the lines and valve were contaminate The maintenance personnel were unaware of the existence of the specific RWP because it was not noted on the work packag Subsequent investigation by the inspectors revealed that, in processing work orders, reviews are required by several organizations, including health physics, as stated in Plant Operations Manual (P0M) Procedure No. 12.000.15, Revision 2 Such a review was performed and acknowledged on the original work order by the HP organization. As the scope of the work changed; however, several modifications were made to the work orde Section 7.1.2.4.C of POM 12.000.15 states; " Changes to Attachment A's (of the work orders) shall be reviewed by QA and Health Physics as applicable and the Nuclear Shift Supervisor. All those who are required to review the change shall sign the Attachment A to signify their review". This was not done completely as noted by the absence of a health physics signature on Revision "0" of the work order package. The inspectors were told by HP personnel that for reasons unknown, the modification was inadvertently not sent to the HP

, department for review and; therefore, the need for a specific RWP was not included in the package. This failure to follow procedure is considered a Level V violation (341/86026-05(DRP)).

. . -. -. _ _. -_

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

. - - - _ - - -

.

.

The inspectors identified their concern to the licensee about the generic implications of this event and whether it could occur routinel For example, the Attachment A does not contain a signature block to trigger the reviews of work order modification The licensee committed to review POM 12.000.15 to determine what changes can be made to resolve this problem, and issued " night orders" to inform the operations staff of the event and to be aware that modifications to work orders must receive proper review. The licensee's action to improve the process permanently will be tracked as an open item (341/86026-06(DRP)).

With respect to the valve itself, the licensee determined that the disk " froze" in the closed position because of deterioration of a fiberglass bushing on the valve shaft. This is the third known case of such a failure at Fermi 2 and the inspectors made known their concerns that other valves at the site also could fail in such a manner. The licensee also was concerned and requested the Engineering Department to formulate corrective / preventive measures, or a program to assure the valves would operate when called upon until such time that permanent corrective action could be undertake The vendor (Jamesbury, Inc.) also has been requested to investigate possible corrective action The inspectors observed the reassembly of the original failed valve and noted that all proper procedures were followed. The purchase order for the new disk and shaft was reviewed and found to meet all QA requirements for traceability. The inspectors also reviewed the qualifications of the person who tack welded the pins which hold the disk to the shaft, noting that he was properly qualified as evidenced by having been tested at Fermi 2 using a proper welding qualification procedur No other violations or deviations were identified in this are . Surveillance Observation (61726)

The inspectors observed surveillance testing required by Technical Specifications and verified that: testing was performed in accordance with adequate procedures; test instrumentation was ca..brated; limiting conditions for operation were met; removal and restoration of the affected components were accomplished; test results conformed with technical specifications and procedure requirements and were reviewed by personnel other than the individual directing the test; and any deficiencies identified during the testing were properly reviewed and resolved by appropriate management personnel. Test activities witnessed in whole or in part were:

. 24.609, Rod Sequence Control System Functional Tes . 24.608, Rod Worth Minimizer Functional Tes . 54.000.01, Shutdown Margin Chec . 54.000.02, Reactivity Anomalies Chec . 24.137.11, SRV/ ADS Valve Operability Test.

l

i

.

,

. 44.030.264, ECCS Reactor Vessel Water Level (Ads Level 3 and Feedwater Level 8) Division II, Channel B Functional Tes . 44.030.252, ECCS Reactor Vessel Water Level, Division II, Channel B Functional Tes Additionally, the inspector performed a record review of completed surveillance tests. The review was to determine that the test was accomplished within the required Technical Specification time interval, procedural steps were properly initiated, the procedure acceptance criteria was met, independent verifications were accomplished by people other than those performing the test, and the tests were signed in and out of the control room surveillance log book. The surveillance tests reviewed were:

. 24.000.02, Shiftly, Daily, Weekly and Situation Required Surveillance . 24.139.01, SLC System Explosive Charge Continuity Verificatio . 24.630, Remote Shutdown Instrument Channel Check . 24.307.14, Emergency Diesel Generator No. 11 - Start and Load Tes . 44.030.263, ECCS Reactor Vessel Water Level (Ads Level 3 and Feedwater Level 8), Division 1 Channel A Functional Tes . 44.010.128, APRM E Channel Functional Tes . 44.010.129, APRM F Channel Functional Test

. 24.000.02, Shiftly, Daily, Weekly and Situation Required Surveillance . 44.020.69, NSSSS Turbine Building Area Temperature

. 44.020.59 NSSSS Condenser Pressure, Division I Channel A Functional Tes . 44.030.254, ECCS Reactor Vessel Water Level (Levels 1, 2, 8),

Division II, Channel D Functional Tes . 44.020.61, NSSSS - Condenser Pressure, Division I, Channel C Functional Tes . 44.030.300, ECCS Drywell Pressure, RHR, CSS, and HPCI Actuation, Division II, Channel B Functional Tes . 44.030.302, ECCS Drywell Pressure, RHR, CSS, and HPCI Actuation, Division II, Channel D Functional Tes . 44.020.238, NSSSS RCIC Steam Line Pressure, Division II, Channel D Functional Tes . 44.020.236, NSSSS RCIC Steam Line Pressure, Division II, Channel B Functional Tes . 44.010.127, APRM D Channel Functional Tes . 44.010.125, APRM B Channel Functional Tes . 44.090.03, Seismic Monitoring, Active Seismic Monitorin . 24.501.13, Yard Fire Hydrant and Hose House Inspectio . 24.206.04, RCIC Flow Rate Test at 150 psi . 24.202.02, HPCI Flow Rate Test at 165 psi The inspector identified two of the surveillance tests which had been completed but not logged as such in the surveillance log book. The condition was brought to the attention of on-shift personnel who immediately logged the completed test ,- . . __ _ .. .-. _

.. -

.

.

No violations or deviations were identified in this are . Licensee Event Reports Followup (92700) Through direct observations, discussions with licensee personnel, and review of records, the following event reports were reviewed by the resident inspectors to determine that reportability requirements were fulfilled, immediate corrective action was accomplished, and corrective action to prevent recurrence had been accomplished in accordance with technical specifications:

(Closed) LER 85001, Loss of Division I Offsite Powe (Closed) LER 85002, Loss of System Service 64 Transforme (Closed) LER 85006, Division I Standby Gas Treatment System and CO

System Inoperabl (Closed) LER 85007, Loss of Circulating Water Reservoir Decant Line Radiation Monito (Closed) LER 85008, Reactor Scram Due to False Upscale Trip of Intermediate Range Monitor (Closed) LER 85009, Failure to Remove RPS Shorting Link (Closed) LER 85010, APRM Scram (Closed) LER 85011, Reactor Scram (Closed) LER 85020, Failure to Meet Sampling Requirements While Exhaust Monitor was Inoperabl (Closed) LER 85023, Actuation of Emergency Equipment Cooling Water Due to Reactor Building Closed Cooling Water

. Valve Controller Failur (Closed) LER 85062, Missed Intermediate Range Monitor Surveillanc (Closed) LER 85070, Missed Containment Atmosphere Samplin (Closed) LER 85080, Operators Fail to Place a Radiation Monitor in Service While Releasing Liquid Effluen (Closed) LER 86010, Tech. Spec. Surveillances Not Performed for Safety / Relief Valve Low-Low Set Caused by Inadequate Procedure for Updating Surveillance Schedul (Closed) LER 86014, Loss of Division II Offsite AC Power Due to Procedural Violations by Personnel Cleaning System Service Transformer 6 (Closed) LER 86017, Emergency Equipment Cooling Water System Cooling Capacity Inadequate for Some Postulated Accident Condition (Closed) LER 86018, Inadvertent Loss of Power to Reactor Building Exhaust Radiation Monitor Results in Radiation Monitor Upscale Trip and Engineered Safety Feature (ESF) Actuation (Closed) LER 86020, Hydraulic Transient During an In-service Leak Test Resulting in an Unusual Event (Closed) LER 86030, Emergency Diesel Generator Lube Oil Filter Checks Misse . - _ _ _ _ _

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

.

.

.

In addition to the review criteria stated above the LERs were reviewed for potential violations of regulatory requirements. The results of that review are presented below:

(1) Violations of Limiting Conditions for Operation were associated

, with LERs 85006, 85007, 85009, 85020, 85070 and 85080. These violations occurred during the same time frame and were of the same type as the violations identified in Inspection Report No. 50-341/85040. As indicated in Paragraph 9.d. of Inspection Report No. 50-341/86019, the escalated enforcement actions of Inspection Report No. 50-341/85040 adequately address these violations and no citations shall be give (2) The inadequate design of the emergency equipment cooling water system explained in LER 86017 was identified by the licensee during the design document reconciliation effort. The inspection and closure of all the other LERs that were produced from the reconciliation effort were discussed in Inspection Report No. 50-341/8601 Based upon the extensive corrective actions, licensee self identification of the under designed cooling water system, and to provide regulatory consistency with regard to the design deficiencies identified during the reconciliation effort, no violation shall be give (3) The event described in LER 86030 was a violation of License Condition 2.C (10). The inspector reviewed the violation and the licensee corrective actions against the criteria of 10 CFR 2 Appendix C.V.A. and determined that the criteria were me Specifically:

(a) The missed filter inspections were identified by the license (b) The failure to perform the filter inspection did not have safety implications. Originally, the filter inspections were to help provide detection of EDG bearing failur This has since been determined by the licensee not to be neces sary. As such, the SER on the EDGs dated July 16,

-

1986, addresses only gap inspection and visual bearing inspection to maintain the EDG reliability envelop (c) The LER was submitted within 30 days of identificatio (d) The licensee's corrective actions were to perform the filter inspections, review all the other license conditions to assure that they were properly identified in the preventative maintenance program, modify the computerized preventative maintenance scheduling calendar to include the filter inspections as a regulatory commitment, and institute a request for modification of the license to delete the filter inspection . . - _ - _ - , . _ - _. .__ _ __ _

- - _ _ _

.

.

(e) No corrective action to a violation in the last 2 years would have corrected the proble (4) All other violations or deviations were previously identified in inspection reports for the LERs listed abov b. The Restart Team members performed a review of one LE The results of their review are provided below:

.

(Closed) LER 86029: Flow Transmitter Failures Causing ESF Actuation of HPCI Valve. On two separate occasions, August 23 and 26, 1986, the inboard containment isolation valve in the steam supply line to the High Pressure Coolant Injection (HPCI)

system isolated. When these isolations occurred the plant was in Operational Condition 2 (startup), reactor pressure was approximately 900 psig and reactor power was less than 5 percen The isolations were the result of a downscale failure of a transmitter that monitors HPCI steam line flow. After the second event, the faulty transmitter was replaced. Two similar failures in other Rosemount Inc. transmitters have occurred, but, per design, these did not result in ESF actuation After the first isolation on the HPCI system, the licensee verified there were no leaks or breaks in the HPCI system piping and investigated the problem. It was determined that the trip unit which activates the auto-isolation logic was in the tripped condition because the flow transmitter failed downscale (this flow transmitter monitors steam flow to the HPCI turbine). While troubleshooting the failed transmitter it began to operate properly for no apparent reason and, since no cause could be determined for the failure, it was returned to service. Three days later this happened again and at this time the transmitter was replaced with an identical mode The two other failures occurred on August 5, 1986, and August 26, 1986. One was in the nuclear boiler system and the other in the RHR syste The inspectors learned that on July 18, 1986, a message appeared on the Nuclear Network Newsletter (an INP0 message system) from the Trojan Nuclear Plant which stated, " Trojan has experienced what may be a generic problem with Rosemount pressure transmitter The transmitters located on our steam lines (Models No.1153 GD8 and No.1153 GD9) have failed either low or high a number of times. When we isolate and vent then repressurize the transmitter functions properly."

On July 22, 1986, Rosemount Inc. sent a letter to Daniel International Corporation of Monroe, Michigan, advising them of a " performance aberration" regarding certain models of its transmitters (Models No.1152, No.1153, and No.1154).

Rosemount stated that it had received reports of a few

_- . . . _ .

.

.

transmitters that demonstrate an intermittent syndrome characterized by an instantaneous output signal shift off scale (around 28 MA high or 2.8 MA low). The shift typically is not accompanied by a shift in the actual process pressure and the condition continues until it is corrected by any or all of the following: fluctuation or removal of input pressure; removal of power supply voltage; or continued operation in the proces Once corrected, the transmitter performs within specification although the offscale output may recur later. Rosemount has not been able to reproduce this phenomena in the laborator This letter is referenced by Detroit Edison Company (DECO) in LER 86-029. DECO apparently was not aware of these letters prior to the first failur The inspectors inquired of DECO personnel what corrective actions were being taken for the long term. The response was that although DECO does not plan any long term corrective action at this time, it has taken several short term action First, DECO contacted Rosemount for possible solutions to the problem but were informed that Rosemount does not have any answers at this time and is not sure where the problem lays- whether it is the instrument itself or in its applicatio Secondly, DECO reviewed General Electric Topical Report No. NED0-21617-A, December 1978, " Analog Transmitter / Trip Unit System for Engineered Safeguard Sensor Trip Inputs," which was submitted to NRR and accepted as a basis for qualifying the transmitters. Based upon the failure rate data in that report and extrapolating the data to the Fermi 2 facility, DECO has determined that 10-15 failures could be expected per year at the Fermi 2 facilit DECO also notes that from information it has received, the transmitters are supposed to annunciate for a gross failure but those at the Fermi site do not. DECO determined that the reason the transmitters do not annunciate is that they are not failing to a " gross failure" position and therefore, do not activate the annunciators. To correct this DECO issued Standing Order Number 86-22 briefly explaining the problem and detailing four steps to preclude a failed detector going unnoticed. Basically the action requires operators to check all the indicators during the shiftly channel checks and to bring any anomalies to management's attentio Although the inspectors are satisfied that Deco is taking a reasonable approach to correct the problem, they are more concerned that this may be a generic problem. It is noted that Rosemount has not issued a 10 CFR 21 report on the proble The inspectors, therefore, have requested further investigation of the problem by the Division of Reactor Safety in NRC Region III and will keep this as an unresolved item pending resolution (341/86026-07(DRP)).

No violations or deviations were identified in this are .

. ..

16. Degraded Grid Voltage Calculation Review During the third party Stone & Webster review of the core spray system the electrical auditor challenged one of the assumptions used for minimum motor starting current. As the licensee attempted to resolve this concern an invalid assumption was identified in the calculations for Division I degraded grid voltage. The licensee reperformed the calculations for Division I degraded grid and submitted a Technical Specification change for a revised setpoint. The Safety Evaluation Report for the Technical Specification change states "Our conclusions regarding the acceptability of the proposed change in the Division 1 degraded grid relay setpoints is dependent on the verification by Region III of the documented undervoltage characteristics of the Division 1 safety-related electrical components." In response to that statement the inspector reviewed: The licensee's documentation from the vendor on the lowest acceptable bus voltage for operation of the Division 1 motor operated valves.

, The calculation output on the minimum bus voltage of the Division I motor operated valve The resolution when calculated valve voltage was less than vendor acceptable valve voltage, The electrical load drawings against the loads specified in the design calculations, The outputs of the appropriate calculations were utilized as inputs into the next set of calculation The inspection revealed the discrepancies listed below: Valve E1150F028A was identified as being fed from Bus 72E-5A in Calculation 968 instead of Bus 728-3 An incorrect bus voltage input was utilized in one of the design calculation The purchase order used to acquire the minimum bus voltage from the vendor was not designated safety relate The licensee revised the calculations with the appropriate inputs to resolve the first two discrepancies. The revised calculations did not affect the degraded grid setpoint. The licensee acquired a letter from the valve vendor stating that the information provided from them was generated in accordance with their quality assurance program and procedures. This resolved the third discrepancy.

.

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

_ _

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

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

..

l

.

'

' 17. RESTART TEAM OPERATIONS OBSERVATIONS (71715)

' Scope An augmented inspection team (e.g., startup team) was constituted to

.

ascertain that all licensee commitments for restart of Fermi 2 had been satisfied after a ten month outage, that actions taken to

! correct the deficiencies stated in the December 24, 1985,

10 CFR 50.54(f) letter had been completed or were being implemented -

and that the licensee was ready and capable to safety operate the reactor. The startup team provided 24-hour inspection coverage during i the initial criticality and observed specific tests during the restart i program. The team members reviewed procedures, interviewed senior management and operations staff personnel, and directly observed i i

operations activities. Particular emphasis was focused on control "

room activities and interfaces.

'

During the heatup test phase (from startup to 5% power), which was

designed to verify the operation of some of the major reactor

-

systems, the licensee completed demonstration runs and surveillance i tests of the RCIC and HPCI systems and declared them operable. The

south reactor feedwater pump, which failed during operation in 1985

'

and had been rebuilt, also was tested, vibration measurements made, and the pump and turbine verified to be functioning properly.

, Similar measurements also were made on the north reactor feedwater

! pump. Vibration measurements were taken on the main steam bypass lines to verify that the modifications made following discovery of cracks in 1985 corrected the problem. At the end of the inspection

! period a regional inspector reviewed the results of

the measurements to date. The inspector's review will be discussed ~

"

! in a future inspection report. The lines will continue to be i monitored as power is increased. Most of the other startup tests i required during the heatup phase, including scram time testing of l all control rods, were completed in 1985 following initial

criticality and startup and were not repeated during the present i heatup testing phase. The NRC startup team has concluded that the

test program has been effective in identifying equipment problems.

l Prestartup Activities i

'

Prior to reactor startup the inspectors walked down accessible l

, portions of several systems to verify operability by comparing system lineups with plant drawings, as-built configurations, or present valve lineup sheets. The specific walkdowns were identified in Paragraph 2 of Inspection Report No. 50-341/86019. Additionally, i the inspectors reviewed the licensee's completed valve lineups for selected plant system i

? l l r

,

i i

j 30 i

l,

- % w"M W O'-*'7WMrd-Pmr-tW Dey% we em---ec-w-*---tr-g+- sv _ &++ e ry v egv ew yr pmm w g prpr e-mm>= ,.

, q Tume -t y p _wirgpe-ty'*-**-J*u---M

. -

'

The inspectors confirmed that licensee commitments and actions required for startup were completed. The inspectors also reviewed the adequacy, completion, and closecut of several modification packages including:

. EDP-5544, Changing the Operation of Drywell Isolation Valve . EDP-5100, Installation of Two Isolation Valves in TIP Purge Line . EDP-5702, Installation of Thermocouples for Reactor Water Cleanup System The inspector reviewed the licensee's corrective action's taken in response to violations identified in Inspection Report No. 50-341/86011 including the status of the surveillance scheduling and testing program revie The corrective actions resulted from an enforcement conference held

'

on May 30,\ 1986 (as documented in Inspection Report No. 50-341/86011)

in which the licensee committed to perform a review of the surveillance procedures and the surveillance scheduling and tracking g program. The licensee had committed to review with their Independent Safety Evaluation Group, (ISEG) all surveillance requirements to insure they were.adequa'?ly addressed by surveillance procedure j , ,

' .'

,

With rega'rd.to"turveillance procedures since the ISEG review was incomplete at the time of the inspection, the inspectors could not

, ascertain the affectiveness of the ISEG review. The inspectors did

! ( notice; however, the apparent thoroughness of the review as

! evidenced oy the large number of accurate findings in addition to the rapid anc." effective resolution of the more significant findings. Subsequent to the inspection the ISEG review effort was completed and resulted in the issuance of Licensee Event Report No. 86022 which documented the discovery of Technical Specification violations caused by surveillance requirements not being met. In the LER the licensee committed to update the LER by November 5, 1986 to add specific detail With regard to the scheduling and tracking program the licensee was usintj primarily the Surveillance Coordinator with some input from the ISEG to review the adequacy of surveillance scheduling and tracking. Thi: review considered the accuracy of the scheduling and tracking program relative to the Technical Specification

surveillance requirements, with emphasis on the accuracy of the schedule's mode requirement The licensee stated it had completed the review prior to the inspection. During the inspector's review, the inspector discovered that the only documented evidence that the Surveillance Coordinator had performed a review was his signature on a licensing action tracking form. The inspector indicated that this was inadequate documentation to close out this portion of the violation. The licensee's Quality Assurance staff also had identified this as a

>

, _ , _ _ .- -

,. ,,,,c -, ,--r- - - --- --- - . - - -. s-- . c.- - - - , - - _ , - - -

,y

_ _ _ _ _ _ _ _ _ _ _ _

.

.

documentation deficiency. In the exit interview the licensee committed to perform another review of the Technical Specification Surveillance Scheduling a'.d Tracking progra At a later date, the inspector verified verbally with the licensee's Licensing Department that a second review had been performe Related to surveillance testing, on July 30, 1986, the following three Engineering Safety Feature, (ESF) actuation occurred: Reactor Water Cleanup Isolation l Reactor Water Cleanup Isolation Group 2 Isolation Item "a" and "b" were attributable to the same isolation valve, but with two seoarate causes. Item "c" was unrelated to the first two in that the cause was a grounded cabl The inspector reviewed the events with the test personnel to determine the root cause of the actuations. The inspector agreed with the licensee that the root cause appeared to be the way in which the testing of the systems was being performed. Specifically, because plant startup was planned shortly, the surveillance tests were being performed quickly and without adequate preparation by the plant staff on familiarization with the system status and the test methodolog As a result of these actuations the licensee began putting by more emphasis by responsible testing personnel on familiarization with system status and on procedural adherence. Licensee action in response to these events appears to be responsible and adequat The inspectors also reviewed potentially generic design deficiency related to the unmonitored failure of a fuse on the negative side of the DC power system for the Intermediate Range Monitors (IRMs).

The unmonitored failure is a potential common mode failure which could cause the IRM's to become inoperable without generating a channel trip. This problem was documented in a General Electric Rapid Information Communication Servic? Information Letter (RICSIL)

Number 7 dated June 26, 1986 and was initially identified at the Monticello facility. At the request of Region III the licensee conducted a test on the IRMs and found they had the design deficiency identified by GE. While Fermi 2 did have the design deficiency the inspector noted that per analyses from G.E. and by design the deficiency had no safety significance at the Fermi-2 facility. After discussions with Region III the licensee voluntarily made the design corrections prior to startu _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ____ _____- _________________ __

. .

.

.

'

The inspector also determined that the same design deficiency existed in the following components:

-Source Range Monitors-Radwaste Liquid Effluent Monitors-Reactor Building Closed Cooling Water Monitors-Fuel Pool Area Monitors-Offgas Process Monitors The licensee indicated these systems would be modified at a later date. This was acceptable to the inspecto c. Plant Startup and Operation Prior to First Shutdown Immediately prior to initial criticality and continuing through the heatup phase of testing (e.g., from "0" power to 5% power) the inspectors spent many hours in the control room observing operations and speaking with control room staff. During the initial startup, the restart team observed control room operations continuously for approximately 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> focusing on communications among shift personnel, operator adherence to procedures, operator recognition of and response to annunciators, involvement of shift supervisors in plant operations, and congestion in the control roo During the initial startup, the control room staff generally functioned in an adequate and professional manner. Staffing levels were satisfactory, procedures were followed, communications were good, turnovers were adequate and logs were kep Communications between the operators on shift and with personnel from other interfacing departments were good. Though not always done formally, acknowledgements were made between the control room operators, the on-shift management, and the nuclear < power plant operators. When an expected alarm was received, the operator would announce this so all persens involved were aware of the situatio The operators were aware of when personnel from other departments were causing alarms. If an operator other than the control room Nuclear Supervising Operator (NS0) acknowledged an alarm, the control room NSO was informed by that operato In general, when an operator, such as the NSO or Nuclear Assistant Shift Supervisor (NASS), left the "at control area" (the area within the bounds of the panels), an acknowledgement was made by the relief operator to verify that he had coverag On August 5, 1986, the inspector noted a slight error as having occurred when, during shift turnover, both NASSs (the off going and

'

on-coming) who are Senior Reactor Operators (SR0s) left the "at controls area" without having the Nuclear Shift Supervisor (NSS) in that area. Less than one minute elapsed prior to the off going NASS returning. The inspector spoke to the NASS who admitted his error of leaving the "at controls area" without the NSS being present as

- - _ - _ _ _

.

.

.

required by Procedure No. 21.000.01, Paragraph 6.3.1 when in Operational Condition 1, 2, or 3 (the unit was in Condition 2 at ,

the time). The licensee was in compliance with its Technical '

Specifications and 10 CFR 50.54(n)(2)(ii) which requires an SRO to be in the control room when above cold shutoown because, at the time, there were at least five SR0s in the control room including the one at the reactor control panel. At the time of this event a procedure change was being processed to delete this requirement (it subsequently was issued on August 20,1986).

A good practice conducted by the operations personnel and observed by the inspectors was frequent briefings by the NSS or another involved individual about upcoming evolutions. For example, prior to starting up or shutting down the reactor, the NSS read through the precautions and limitations section of the procedures to inform all involved control room personnel what to expect. The NSO in charge of performing the control rod manipulations would read aloud or discuss the procedural steps he would be taking in the upcoming evolutions. Prior to major surveillances, the indiv.idual conducting the test would brief those in the control room on what would be required and what would be expected. Also, during the startup the onshift reactor engineer would inform the NSO making the control rod manipulations if a source range monitor (SRM) short period alarm would be expected or if the rod was a high worth ro Another time in which meetings or briefings were used was at each shift turnover. Following the individual turnovers for each position, the NSS conducted a meeting in the control room for all shift personnel including licensed operators, non-licensed operators, reactor engineering, chemistry, and health physics representatives. The NSS gave the limiting conditions for operations that were currently in effect, planned activities for the shift and problems that occurred since their last shift. The NASS and control room NSO also provided input to this meeting. This also served as an opportunity for anyone to ask questions concerning the plant and activities. One potential problem of conducting these meetings in the control room is the possible distraction this could cause to the operators in responding to any alarms during the meeting due to the increased congestion from the relatively large number of people involved (at times in excess of 30). Although the inspectors are not aware of any specific problems this has caused, they believe it is something the licensee should watch closel The shift turnovers were satisfactor Each position turned over individually prior to the gener3! shift meeting, allowing the off going and on-coming personnel.to walk down the control panels together to go over any unusual conditions. This gave the on-coming person ample opportunity to ask questions concerning equipment status or annunciated condition . - . _ . -. . - - . -. -__ . . -

o

.

O On August 6, 1986, members of the startup team observed the response of operations personnel to a real emergency caused by a fire in an electrical breaker. The response was calm and deliberate, the emergency plan was followed, the fire extinguished, and the reactor was controlled at all times. The reactor subsequently was voluntarily shut down. This is discussed in more detail in Paragraph 8 of this repor In summary, the operational activities that were observed by the inspectors during the initial startup were handled in a professional and competent manne d. Continuation of Heatup Phase Activities Following the shutdown the licensee performed a startup and continued with the heatup phase. As major evolutions of the licensee's test program occurred, the startup inspection team again provided continuous coverage of operation During this phase of testing a reactor scram occurred on August 29, 1986 when the north reactor feedwater pump (which was feeding the reactor at the time) tripped on low suction pressure. The operators were transferring the feedwater pump supply from the east heater feed pump to the center heater feed pump. Subsequent investigation revealed that the manual discharge valve for the center heater feed pump was closed. Thus, when the east heater feed pump was stopped there was no water supply to the north reactor feedwater pump and it tripped on low suction pressure. During transient recovery operations the operators adjusted the reactor flow limiter down to 3% which caused the main steam bypass valves to close and the primary system pressure to increase above the scram point. Also, during recovery from the scram, the cooldown rate exceeded the administrative guidelines of 90 per hour but did not exceed the Technical Specification limit of 100 per hou Had the operators closed the Main Steam Isolation Valves (MSIVs) sooner (this was done about 53 minutes after the scram), the 90 rate probably would not have been exceeded. The startup team has concluded that this scram was avoidable. The utility has been practicing evolutions of this type on the plant specific simulator to avoid recurrenc Furthermore, a " lessons learned" memorandum was issued to all plant operations personnel describing the even Scram recovery was performed and operations continued until a controlled shutdown was begun on September 2, 1986, to repair an inoperable vacuum breaker. Generally, the operational activities continued to be satisfactory during this time frame. However, there was a reactor operator error on September 3,1986, when an out-of-sequence control rod was inserted during the controlled reactor shutdown. The operator recognized the error when he selected the next rod. The operator stopped all rod manipulations and notified the shift supervisor of his mistak ..

o

.

O Operations were not resumed until the reactor engineer was consulted and determined that, in the condition the reactor was in (only about 25 rods remaining to be inserted) it would not be necessary to withdraw the mispositioned rod to return to the original sequence of rod insertion The inspectors inquired how the analysis was performed by the reactor engineer as there was no documentation in the operator's log book about this. They were informed that the on-shift reactor engineer had evaluated the information available, determined that the misplaced rod fell within the safety envelope and no further action was necessary. Therefore, he informed the operations staff that it could continue inserting rods. The inspectors expressed concern that there was no documentation in the operator's log book on the analysis. The licensee agreed that documentation of the reactor engineer's analyses should be improved and committed to make this improvement. This will remain open until such time that the improvements are made and evaluated by the NRC (50-341/86026-08(DRP)).

With regard to the rod insert error, the inspectors believe it is an isolated event. However, they did speak with the licensee about it and noted that one part of the licensee's corrective action was to remove the entire shift crew from its normal shift and " debrief" them to learn more about how the event happened. After the debriefing, a " lessons learned" memorandum was addressed to all operations personnel, e. Test Condition 1 Following repair of the vacuum breaker valve on September 6, 1986, the unit commenced restart. The NRC Regional Administrator gave authorization on September 12, 1986, to allow the plant to be '

operated up to 20% power. Upon completion of maintenance activities on the off gas system the licensee proceeded into Test Condition 1.

, In general, the inspectors believe that operations were satisfactory

'

and cite as an example the response to a transient which occurred on September 18, 1986, when both steam bypass valves opened fully for no apparent reason, causing an increase in indicated reactor water level. Subsequently, reactor power increased from about 15% to about 18%. The NS0 on duty recognized the transient immediately and began to recite plant parameters as he found them. The second NSO and the NASS were quick to support the first NSO in taking action to mitigate the transient. As a further example of the quick response, about two to three seconds after the event started, the NASS directed that the turbine be tripped (it was being brought on-line for turbine testing at the time). Within about ten seconds, nine to ten persons had gathered in front of the control panel for the turbine (which is near the operator's control panel) causing congestion in that area. The NASS ordered all non-essential personnel away from the contiol board. This was an appropriate response and leads the inspectors to believe that control room personnel are aware of their increased responsibilities now that the reactor is in an operational mod !

-. -_- ..,= - _

, _ - _ _ - - -

o

,

o During the latter part of September, the licensee was having difficulty maintaining the proper reactor water conductivity. A 72-hour LC0 was entered on September 18 when the conductivity reached about 1.3 micro sho/cc. The licensee was able to reduce that value to slightly less than 1.0 micro mho/cc (the Technical Specification value for operations) early on September 20 and thereby exited the LCO. Later that same day; however, the value again exceeded 1.0 and the 72-hour LCO was entered. The Technical Specification has two limiting conditions: 1) operation for periods of time not to exceed 72-hours continuously if conductivity is greater than 1 micro mho/cc; and 2) total operational time not to exceed 336 hours0.00389 days <br />0.0933 hours <br />5.555556e-4 weeks <br />1.27848e-4 months <br /> / year at conductivities greater than 1 micro mho/cc. Since the total operational time was approaching 100-hours and Fermi 2 is still early in the testing phase, the yearly limit could have posed an operational problem when the unit went to commercial production (scheduled for summer 1987). Therefore, the licensee decided to shut the reactor down to investigate the cause of the high conductivity. A controlled shutdown was initiated on September 2 Investigation by the licensee determined that much of the problem was caused by condenser tube leak f. Core Thermal Power Evaluation Prior to escalating power from the heatup test phase to Test Condition 1 (that is, from 5% to 20% power) the neutron monitoring instrumentation is calibrated by the Reactor Engineering organization using the best available data. However, since power levels are low and heat balance information may be unreliable, the instrumentation must be recalibrated as power levels increase. This was done by DECO and it was determined after increasing power above 5 percent that when the indicated power was 8.3 percent, actual power was 13.75 percent. The instruments, therefore, were recalibrated before the licensee increased to 20 percent. This recalibration will be checked several times more as power is increase The inspectors inquired how the calibration is performed and were informed that there are standard DECO procedures for doing so. The inspectors then asked the GE startup consultants on site if they had been contacted by DECO in performing these calibrations, and if they were satisfied with the results. The GE consultants said that the DECO procedures were based on GE procedures, that they had reviewed DECO procedures, and that they are satisfied with the results. They said the discrepancy noted in the actual versus indicated power is not atypical when starting up BWR g. Review of the Control Room Audit Program The team reviewed records of the licensee's control room audit program and occasionally observed audits in progress. The Team Director also observed one complete audit. The records and

o i

o observations indicated that audits were being performed on time and were thorough. One finding identified by the Detroit Edison Group Vice President involved the unawareness by a reactor operator of the complete status of soms of the systems on one of the control panels he was monitoring. The team views continuation of the audit program and strengthening of audit criteria as a sound approach to continued good operator performanc No violations or deviations were identifie . Management Meetings (30702)

On September 24, 1986, NRC Region III met with the licensee at the Nuclear Operations Center to discuss potential changes to the power ascension test progra . Report Review (90713)

During the inspection period, the inspector reviewed the licensee's Monthly Operating Reports for May, June, July, and August 1986. The inspector confirmed that the information provided met the requirements of Technical Specification 6.6.A.3 and Regulatory Guide 1.1 No violations or deviations were identified in this are . Unresolved Items Unresolved items are matters about which more information is required in order to ascertain whether they are acceptable items, violations or deviations. Unresolved items disclosed during the inspection are discussed in Paragraphs 6.b and 1 . Open Items Open items are matters which have been discussed with the licensee, which will be reviewed further by the inspector, and which involve some action on the part of the NRC or licensee or both. Open items disclosed during the inspection are discussed in Paragraphs 6.c, 6.e, 13.a, 13.e, and 1 . Exit Interview (30703)

The inspectors met with licensee representatives (denoted in Paragraph 1)

on September 4 and 26, 1986, and informally throughout the inspection period and summarized the scope and findings of the inspection activities. The inspectors also discussed the likely informational content of the inspection report with regard to documents or processes reviewed by the inspectors during the inspection. The licensee did not identify any such documents / processes as proprietary. The licensee acknowledged the findings of the inspectio