IR 05000456/1998003

Insp Repts 50-456/98-03 & 50-457/98-03 on 980128-30.No Violations Noted.Major Areas Inspected:Operations & Maint

ML20216H264
Person / Time
Site:	Braidwood
Issue date:	03/16/1998
From:	Gardner R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
Shared Package
ML20216H252	List: IR 05000456/1998003 ML20216H247
References
50-456-98-03, 50-456-98-3, 50-457-98-03, 50-457-98-3, NUDOCS 9803200231
Download: ML20216H264 (12)

Text

.

,

i U.S. NUCLEAR REGULATORY COMMISSION REGIONlli Docket Nos: 50-456;50-457 License Nos: NPF-72; NPF-77 Report Nos: 50-456/98003; 50-457/98003

Licensee: Commonwealth Edison Company  !

I Facility: Braidwood Nuclear Power Station, Units 1 and 2 Location: RR #1, Box 84 Braceville,IL 60407 Inspection Dates: January 28-30,1998 Inspectors: J. Neisler, Team Leader J. Adams, Resident inspector, Braidwood N. Hilton, Resident inspector, Byron Approved by: R. Gardner, Chief. Engineering Specialists Branch 2 Division of Reactor Safety

)

-

9903200231 990316 PDR ADOCK 05000456 G PDR l

'

I .

l

-

l'

l k

EXECUTIVE SUMMARY l

Braidwood Nuclear Power Units 1 and 2 NRC Inspection Reports 50-456/98003; 50-457/98003 l

This inspection reviewed and evaluated activities associated with the January 26,1998 reactor trip. A summary of inspection results and conclusions is listed belo .

Control room operators properly responded to the reactor trip by entering and performing the steps contained in procedure 2BwEP-0. Analysis of selected plant parameters indicated that the plant was quickly stabilized in a hot shutdown condition and responded as designed (Section 01.1).

.

The shift manager failed to assign the correct priority to the emergent work on the digital electro-hydraulic control (DEHC) system. The " urgent" priority initiated actions that bypassed the normal screening process which included a review by a minimum of five persons with the required knowledge base (Section 01.1).

. Work control and planning was inadequate. The use of a blanket work request l

bypassed the licensee's normal work control process (Section M1.2).  !

. Operational analysis department (OAD) technicians used uncontrolled documentation in performance of the maintenance activity. The shift manager and unit supervisor failed to question the use of the uncontrolled documentation and also failed to question whether troubleshooting following the replacement of the nuclear programmable logic card was within the scope of the blanket work request, WR No. 980001222, Task 04 .

(Sections 01.1, M1.3).

. No procedure was used by OAD personnel during bench testing the replacement card and proper test equipment were not on site to perform the appropriate test. Although proper test equipment was at Byron, the card was not taken to Byron until after the Braidwood reactor trip (Section M1.3).

l

2 )

l

.

-.

Report Details On January 26,1998, Braidwood Unit 2 experienced a reactor trip during troubleshooting activities. When s digital electro-hydraulic control (DEHC) circuit card was replaced the

- turbine's .103% overspeed protection actuated causing the closure of the turbine governor valves and intercept valve . Operations 01 Conduct of Operations 0 Control Room Ooerator Resoonse to a Unit Two Reactor Trio insoection Scoos (71707)

The inspectors reviewed the conditions causing a Unit 2 reactor trip hitiated during the performance of a maintenance activity on the DEHC system. The inspectors conducted interviews with nuclear station operators, unit supervisor, shift manager, and operational analysis department engineers, on-shift at the time that the reactor trip occurred. The inspectors reviewed the sequence-of-event recorder output; computer point history for selected Unit 2 parameters; and NSWP-WM-08," Action Request Screening Process,"

Revision Observations and Findings On January 26, at 9:53 p.m., Unit 2 experienced a reactor trip during the performance of a maintenance activity on the DEHC system. The inspectors arrived on site approximately one and one half hours after the reactor trip occurred and noted that the plant had been stabilized and was being maintained in a hot shutdown condition. The inspectors conducted interviews with the nuclear station operators, the unit supervisor, and the shift manager discussing the cause of the reactor trip, the immediate actions taken by the operators, and the plant's response. The operators told the inspectors that during routine DEHC computer checks, operational analysis department (OAD)

personnelidentified indications of a circuit card failure. Following further investigation, OAD engineers isolated the failure to a power supply on nuclear programmable logic card No. H54. The nuclear programmable logic card performed the following functions:

. Indicated failures of the overspeed protection circuitry (OPC),

. Illuminated the "OPC Monitor,""OPC Pressure Transducer," and the Megawatts Transducer Monitor" lamps on the "A" DEHC system panel, and

. setting the load drop anticipator circuit for the closure of the turbine govemor and intercept valves when the generator output breakers ope l

.

.

OAD determined from technical manuals that nuclear programmable logic card No. H54 l

was serviceable with the unit on-line, and with the DEHC in either manual or automatic

[ operating modes. Work task 04 was prepared under blanket work request WR N , and a replacement card was requisitioned. The inspectors reviewed the blanket work request and noted that it contained four steps that provided little direction for card removal and replacemen The shift manager assigned an " urgent" priority (priority code "B1") to the work reques The shift manager told the inspectors that the "B1" code means that work should be scheduled and started within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. The inspectors reviewed NSWP-WM-08, " Action Request Screening Process," Revision 1, and determined that since the work was emergent work with "B1" priority, the work request did not require the normal level

. screening and was processed by the shift manager. A priority classification of "B1" was not in accordance with procedure NSWP-WM-08, Exhibit E, " Priority Screening Flow,"

which if followed would have assigned an " emergent" priority (priority code of "B2")

allowing 14 days to schedule and start the work. Assignment of the proper priority code would have allowed the work request to proceed through the normal screening process which includes a technical review by a minimum of five personnel with the required knowledge bas The unit supervisor told the inspectors that a heightened level of awareness briefing was conducted with the nuclear station operators, the unit supervisor, the shift manager, and two operational analysis department engineers prior to the commencement of the maintenance activity. OAD engineers used training manual schematic drawings of a nuclear programmable logic card to explain the card's functions to the operators and advised them that the 103% overspeed protection was not available as a result of the card failure. The inspectors were toid that the training manual drawing was not a controlled document and was used to perform portions of the maintenance activity. The OAD engineers told the operators that this card was replaced in September of 1997, following a lightning storm, with the unit on-line, and with DEHC in the automatic mode with no adverse effect to the unit. The unit supervisor told inspectors that the plans for the replacement of the nuclear programmable logic card included stationing the administrative nuclear station operator at the DEHC to monitor turbine govemor valve position and establishing communications between the administrative nuclear station operator and the OAD engineers at panel 2PA22J. The replacement plans also included contingency actions to be taken in the event that the DEHC did not respond as expected. The unit supervisor instructed the administrative nuclear station operator to

'

shift the digital electro-hydraulic controls to the manual mode if indications of unexpected digital electro-hydraulic control response were observed during the nuclear programmable logic card replacemen The unit supervisor gave permission for switching the defeat switch to the " disabled" position and the removal of nuclear programmable logic card No. H54. These actions were performed without incident. OAD engineers determined that the card required replacement and proceeded to bench test the card received from C-Team. The l

Inspectors were told by OAD personnel that no procedure was used for the bench

! testing of the new card. Following testing, a second briefing was conducted with OAD L

E )

r engineers and control room operators specifically to discuss the installation of the new ;

! '

card. The unit supervisor gave permission for the insertion of the new card and the card l .was successfully nser i e .t d The unit supervisor then gave permission to switch the

[ defeat switch to the " enabled" position. Following the repositioning of the defeat switch to the " enable" position, operators observed the "OPC Pressure Transducer" and the ,

" Megawatts Transducer Monitor" lamps on the DEHC system "A" panel illuminate, but (

no changes in turbine govemor valve position or load were observed. Following l . additional discussions with control room operators, OAD engineers performed several L voltage measurements on the nuclear programmable logic card. Based on the voltage measurements, it was determined that card removal was required. The shift manager, unit supervisor, and OAD engineers to use the same removal plan that was used for the initial card removal. Immediately following the movement of the defeat switch to the

" disabled" position, the unit tripped. The inspectors asked the shift manager and the unit supervisor if they considered the additional troubleshooting steps and the ')

7 repositioning of the defeat switch after receiving unexpected indications from the DEHC to be within the scope of the blanket work package. The shift manager and the unit supervisor stated that they did not consider the scope of the work package in their-discussions with OAD engineer The unit supervisor told the inspectors that the nuclear station operators immediately -

entered 2BwEP-0, " Reactor Trip or Safety injection," and stabilized the plant in a hot shutdown condition. The unit supervisor told inspectors that the plant responded to the trip as it was designed to respond and no problems were noted with the exception of the DEHC problem that initiated the reactor trip. The inspectors reviewed the sequence-of-event recorder output and the computer point history for power range nuclear instruments, control rod position, average reactor coolant temperature, pressurizer pressure and level, steam line header pressure, feedwater pump discharge pressure, generator megawatts, and steam generator levels. The inspectors noted that sequence-of-event recorder output and computer point history for selected Unit 2 parameters supported the unit supervisor's assessment that the plant responded to the reactor trip as it was designed to respon .

The inspectors observed on the sequence-of-event recorder output that the unit tripped on a " Low Low" level in the 2A steam generator. The steam generator " Low Low" level would indicate that the turbine governor valve or intercept valves must have rapidly closed causing the steam generator levels to " shrink" to the reactor trip setpoint. The sequence-of-event recorder output also supports the rapid closure of the govemor or'

intercept valves, in that the turbine trip alarm occurred 0.077 seconds after the reactor trip occurred. The inspectors asked the administrative nuclear station operator stationed at the digital electro-hydraulic controls if he saw any indication of the governor or intercept valves going closed before the reactor tripped. The administrative nuclear station operator told inspectors that he observed the illumination of "OPC Pressure Transducer " and the " Megawatts Transducer Monitor" lamps on the DEHC system "A" panel but did not see any change in the "GV (Govemor Valve) Additive Position" mete The licensee later verified that the configuration of nuclear programmable logic card l; No. H54 caused the rapid closure of the turbine govemor and intercept valves. This was s

! l l

L

!

l

'

g .

-

. .,

l

'

~ confirmed by installing card No. H54 in the Byron Station DEHC simulator which repeated the rapid closure of the turbine govemor and intercept valve Conclusions The inspectors concluded that the control room operators properly responded to the reactor trip by entering and performing the steps contained in 2BwEP-0. Analysis of the ,

computer point histories of selected plant parameters indicated the plant was quickly j stabilized in a hot shutdown condition and responded as designed to the reactor tri The inspectors concluded that the shift manager failed to assign the correct priority to the emergent work DEHC system in accordance with NSWP-WM-08. The " urgent" priority initiated actions that bypassed the normal screening process which included review by a minimum of five persons with required knowledge base. The shift manager and unit supervisor failed to question the use of uncontrolled documentation in the performance of the maintenance activity; and failed to question whether troubleshooting following the replacement nuclear programmable logic card within the scope of blanket work request WR No. 980001222, Task .1 Post-Trio Review of Condensate and Feedwater System Problems Insoection Scooe (71707)

The inspectors reviewed the circumstances surrounding the lifting and failure of 11 feedwater heater relief valves (feedwater side) in the open position, leakage on the 2C feedwater pump discharge check valve, and the lifting of the suction relief valve on the 2B feedwater pump. The inspectors reviewed the computer point history for the condensate pump discharge header pressure, condensate booster pump discharge header pressure, feedwater pump suction header pressure, and the feed water pump discharge header pressure. The inspectors conducted interviews with the nuclear station operator, unit supervisor, and shift manager, on-shift at the time that the event occurred; the secondary systems engineering group leader; and a member of the maintenance staff about the problems observed on the feedwater and condensate systems during the even Observations and Findinas On January 26, at 9:53 p.m., Unit 2 experienced a reactor trip following a maintenance induced closure of the turbine govemor and intercept valves. Following the reactor trip, operators identified 11 feedwater heater relief valves that had lifted and failed in the c open position, leakage on the 2C feedwater pump discharge check valve, and the lifting of the suction relief valve on the 2B feedwater pum The failure of the 11 feedwater heater relief valves required operators to open feedwater heater bypass valves and isolate the affected feed water heaters. The inspectors discussed the failure of the feedwater heater relief valves with the nuclear station operator, unit supervisor, and shift manager, on-shift at the time that the event occurred and were told that feedwater heater relief valve failures were not unexpected for this

$

.

type of a transient. The inspectors were told that feedwater isolation events have

always resulted in relief valve failures on both of the Braidwood and Byron units. The unit supervisor told inspectors that the failure of the valves does not cause any major problems but does divert the operator's attention from the initiating event while they bypass and isolate the feedwater heaters to stop the leak. The inspectors verified that

,

!

the licensee had previously identified and was tracking the condition as an operator work aroun )

l The inspectors discussed the feedwater heater relief valve problem with the secondary systems engineering group leader and a member of the maintenance staff. The L inspectors were told that several modifications that would prevent relief valve failures were currently under evaluation by system engineering but most did not appear to be j

'

economically justifiable since the new relief valves are inexpensive, the relief valves are easily replaced, the valvec fail in the open position providing the necessary overpressure protection for the associated feedwater heater, and precursor events for

]

relief valve failure occur infrequentl The inspectors were concemed that the condensate and feedwater system may have been over pressurized due to a known degraded condition that could cause two of the feedwater check valves,2FWO79C and 2FWO79D, to stick open. The inspector's  ;

'

concern was further supported by the number of relief valves that lifted, the numbe'r of feedwater heater relief valves that failed, and the observed leakage from the 2C feedwater pump discharge check valve. The inspectors reviewed computer point history trends for the condensate pump discharge header pressure, condensate booster pump discharge header pressure, feedwater pump suction header pressure, and the feed water pump discharge header pressure during the event. The pressures recorded on the point histories were normal for the configuration of the condensate and feedwater systems and the peak pressures observed corresponded to the time of the closure of feedwater isolation valves (2FW009A-D). The inspectors did not note any condensate

!

of feedwater pressure increase prior to the closure of the feedwater isolation valves and did not observe any indications that would lead one to believe that check valves 2FWO79C and 2FWO79D failed to close, Conclusions The inspectors concluded that the feedwater and condensate systems responded to the rapid closure of the turbine govemor and intercept valves and the reactor trip as

, - designed with the exception of the 11 feedwater heater isolation valves that failed ope Although contrary to design, the operators expected feedwater heater isolation valve failures to occur during a transient involving a feedwater isolation. The inspectors j

~ concluded that the licensee was aware of the design deficiency with the feedwater heater relief valve and was attempting to identify an economically viable modification to correct the condition. The licensee had identified and was tracking the feedwater heather relief valve problem as a operator work aroun The inspectors concluded that peak pressures for the condensate and feedwater systems corresponded to the time of the closure of feedwater isolation valves 7  ;

}

L L

.

.

L (FWOO9A-D) and the pressure trends recorded on the computer point histories were

! normal for the isolated configuration of the condensate and feedwater systems. The l known degraded condition of two of the feedwater check valves,2FWO79C and 2FWO790, had no observable effect on the pressure transient experienced by the condensate or feedwater systems.

l 11. Maintenance M1 Conduct of Maintenance

, M1.1 General Comments Insoection Scooe The inspectors reviewed the licensee's activities involving the identification, repair and replacement of the DEHC No. 54H circuit card.- Also, the inspectors interviewed licensee technicians, engineers and managers regarding the repair and replacement activities including troubleshooting and testing of the car Observations and Findings During the daily routine computer check of the Unit 2 DEHC memory on January 26,'

1998, the licensee's operations analysis department (OAD) identified a failure of the power supply on NPL card H54. This card has three responsibilities: Indicates failures within the overspeed protection circuitry (OPC). Lighting the "OPC Monitor," OPC Pressure Transducer," and the " Megawatts Transducer Monitor" indicating light . Setting the load drop anticipation (LDA) circuit for closing the turbine govemor valves and intercept valves when the generator breaker opens. Anticipatory overspeed protection for the turbine at 103 percent of rated RP . See specific discussions of licensee activities in M M1.2 Work Control Insoection Scone The inspectors reviewed work activities and work controls involving troubleshooting and replacement of the DEHC circuit car Observations and Findings

'

The licensee's OAD performed the troubleshooting and replacement of the card. During

-

the troubleshooting, OAD found a failed fuse in the power supply. The fuse was

replaced and the card tested. After the new fuse failed during the test, OAD decided to replace the card with a new card from store The Westinghouse 7300 systems use generic cards that must be configured to the application. OAD removed the programmable read only memories (PROM) from the old card and installed the PROMS on the new card, the card was then bench tested and replaced in the DEHC cabinet. When the card was re-inserted and enabled, two of the three annunciator lights immediately illuminated but no operational effects on the turbine or plant were identified. After performing signal checks and discussions with operators, OAD decided to remove the car The enable / disable switch on the card was switched to the disable position. This action resulted in a false signal that closed the turbine control valves and intercept valves. The valve closure was immediately followed by a reactor trip on steam generator low-low leve Work was performed under a blanket work request titled " Unit 2 Computer Maintenance." A work request task, No. 980001222-04, was written by OAD to replace the failed card. The work request task was not processed through the work control j center. No technical reviews of the task were performed, no instructions other than

" replace NPL card, document maintenance alteration," were provided to the technicians l

and no information was provided on the special configuration required for the generic Westinghouse 7300 series card in the Braidwood DEHC syste Subsequent to the trip, licensee management issued instructions forbidding the use of blanket work request for troubleshooting and repairs on sensitive component Conclusion Work control and planning was inadequate prior to commencing work on the DEH !

The use of a blanket work request by OAD effectively bypassed the licensee's work ,

control proces I M1.3. Failure to Procerly Configure DEHC Card Insoection Scooe The inspectors reviewed drawings, vendor information, reports and examined DEHC circuit boards. In addition, discussions were held with various licensee employee Observations and Findings The circuit card used in the Braidwood DEHC unit was a generic card used in Westinghouse 7300 systems. The card must be configured on site for the specific application. The card was not correctly configured prior to installation at Braidwood.

I

l

.

.

The generic card was configured for eight PROM module installation. At Braidwood, only six of the PROM modules were installed leaving two open positions. Each PROM module output was connected to a plug-in inverter with an open collector output. The

!

output of this inverter was tied to the output of three other inverters each driven by the output of a PROM module connected to one of the three other output groups. Vendor instructions state that if any PROM driving a particular bus is left off the card, the inverters associated with that PROM are also left off to ensure proper operation of the bus with the remaining inverters and that the PROM module and the two inverters associated with it should always be either plugged-in or left off the card as a grou The six PROMS were installed, leaving two positions without PROMS, by OAD but the inverters associated with the unused PROM positions were not removed from the car OAD bench tested the card prior to installing in the DEHC rack and determined that the card was functioning properly. However, Braidwood did not have adequate test equipment to properly test the card. The proper test equipment was available at Byron, however, the card was not sent to Byron for testing prior to installation. Therefore, the improperly configured card was installed in the DEH The inspectors noted that OAD used a drawing from a training manual instead of a controlled drawing for troubleshooting and replacement of the DEHC card, in addition, they did not compare the new card to the old properly configured card, and did not recognize the visible difference between the card Conclusion The inspectors concluded that inadequate configuration controls were used for the DEHC card replacement. In addition, the inspectors considered drawing control to be inadequate, card post-modification testing to be inadequate and OAD failed to adhere to vendor instructions for configuring the car V. Manaaement Meetings X1 Exit Meeting Summary The inspectors presented the inspection results to members of licensee management at the conclusion of the inspection on January 30,1998. The licensee acknowledged the findings presente The team asked whether any materials examined during the inspection should be considered proprietary. No proprietary information was identifie *

.

PARTIAL LIST OF PERSONS CONTACTED  !

Licensee

~ M. Cassidy, NRC Coordinator, Regulatory Assurance C. Dunn, System Engineering Supervisor J. Meister, Engineering Manager D. Sager, Vice President, Generation Support G. Schwartz, Site Manager T. Simkins, Regulatory Assurance Supervisor T. Tulon, Vice President NHC C. Phillips, Senior Resident inspector INSPECTION PROCEDURES USED IP 71500: BOP inspection IP 71707: Operations inspection IP 97702 Prompt Onsite Response to Events at Operating Reactors IP 93804 Maintenance Program

- _ _ _ _ _ .

7- .

~

l

,

!- PARTIAL LIST OF PERSONS CONTACTED-Licensee M. Cassidy, NRC Coordinator, Regulatory Assurance

C. Dunn, System Engineering Supervisor

! J. Meister, Engineering Manager D. Sager, Vice President, Generation Support G. Schwartz, Site Manager T. Simkins, Regulatory Assurance Supervisor T. Tulon, Vice President NBC l C. Phillips, Senior Resident inspector INSPECTION PROCEDURES USED IP 71500: BOP inspection 1 IP 71707: Operations inspection  !

IP 97702 Prompt Onsite Response to Events at Operating Reactors IP 93804 Maintenance Program i

l

,

,

LIST OF ACRONYMS USED DEHC Digital Electro-hydraulic Control NSWP Nuclear Station Work Procedure j OAD Operations Analysis Department j PROM Programable Read Only Memory l

NPL Nuclear Programmable Logic

!-

f-I

'

i l 12