IR 05000243/1998201
| ML20249B808 | |
| Person / Time | |
|---|---|
| Site: | Oregon State University |
| Issue date: | 06/19/1998 |
| From: | NRC (Affiliation Not Assigned) |
| To: | |
| Shared Package | |
| ML20249B805 | List: |
| References | |
| 50-243-98-201, NUDOCS 9806240246 | |
| Download: ML20249B808 (15) | |
Text
.. .. . - - _ _ . _ _ _ _ - _ - _ _ _ _ - - _ - - _ _ _ _ - - _ _ _ - _ - - _ _ _ _ _ _ _ _ _ _ - _ - - _ -
v
...
...
~ U.S. NUCLEAR REGULATORY COMMISSION l OFFICE OF NUCLEAR REACTOR REGULATION Docket No.: -50-243
.
.
. License No.: R-106
^
Report No.: 50-243/98-201
~ Licensee: Oregon State University
- Facility: TRIGA Mark-Il Reactor Facility Location: Raolation Center, Oregon State University Corvallis, OR 97331-5904 Inspection Conducted: February 18-20 and May 6 and 11-13,1998 Inspector: C. Bassett, Senior Non-Power Reactor inspector Approved by: Seymour H. Weiss, Director .
Non-Power Reactors and Decommissioning
' Project Directorate Division of Reactor Program Management
- Office of Nuclear Reactor Regulation
- - ..
6 402M N M P.DR ADOCK O y g _
_ = _ _ _
.
C
.
.
EXECUTIVE SUMMARY Oregon Ltate University TRIGA Mark-Il Reactor Facility NRC Inspection Report No. 50-243/98-201 The primary focus of both portions of this reactive inspection was the on-site review of the circumstances surrounding an event on February 17,1998, involving operation of the TRIGA Mark-Il Reactor without the required scrams being operabl Licensee Conclusions e The root cause of the problem was that the reactor console key switch became stuck in the " reset" position and no scrams were operable due to a change in the wiring of the console completed at some time in the pas e The reactor could still have been scrammed by other means including turning off the power to the console, e The problem, although serious, was of very low safety significanc Licensee Corrective Actions
- Examine, clean, and verify proper operation of the reactor console key switc e Modify the reactor console circuitry to make it consistent with that shown and evaluated in the original design drawings for the TRIGA reacto e Conduct a point to-point and e!ectronic check of the scram loop circuitry to provide assurance that the as built condition matches the circuitry shown in the facility documentatio e Complete a semiannual surveillance of the reactor consolo to verify proper functioning and voltage of the console, the power range monitor, and the left-hand drawe e Modify the reactor start-up procedure to add a scram test that would confirm that the control rod magnetic power is de-energized when the console key switch is in the
" reset" positio NRC Review e The licensee reported the event of February 17,1998, as required by Technical Specification. A 14-Day Report was submitted on March 2,199 e The licensee completed all the corrective actions as stipulated in the letter to the NRC l dated March 2,199 g l
.
-- _ _ _ _ - . _ _ _ _ _ _ _ _ _ . - _ _ - _ _ - _ _ - _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ - _ - _ _
..
.
,
e it could not be dew r9d who may have made the change nor when the actual change to the console ;;ottty tas complete e The surveillance eef 3intenance of equipment and training of operators were being conducted as 'Nwre * Two apparent violations were identified: 1) the reactor was operated for approximately 14 minutes without the required scrams being functional; and,2) the licensee apparently had not rnaintained updated facility drawings that reflected the change that had been made.
L
.
_ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _
.
.
.
REPORT DETAILS Summary of Plant Status The licensee's one and one-tenth megawatt (1.1 Mw) research reactor with pulsing capability was shut down on February 17,1998, following operation of the reactor without the scram functions required for shutdown being available. Durin0 the first part of the inspection, February 18-20, the licensee reviewed the problern, evaluated the root cause(s),
and was preparing to take corrective actions based on the results of the review and evaluation. During the second part of the inspection, May 6 and 11-13,it was verified that the licensee had completed all the corrective actions planned and had begun operating the reactor agai . Background Information Before startup of the reactor at Oregon State University (OSU), a comprehensive list of checks is completed. The startup c. heck sheets are reviewed by the Reactor Supervisor who then approves reactor startup if everything is determined to be satisfactory. Each day the core excess reactivity is measured, with the reactor at a power level of 15 watts, and the shutdown margin calculate The OSU TRIGA Mark-Il reactor has a console key switch used to supply control rod magnet current and to reset scram relays. To operate the reactor, a key is inserted into the switch, pushed down, and turned to the right to the " operate" position. The key can be turned further to the right, to the " reset" position, against a spring to reset the reactor scrams. When released, the spring is designed to force the key back to the
" operate" position for normal operatio . Summary of the Event On February 17,1998, at approximately 8:15 a.m. licensee personnel began routine start-up checks of the Oregon State University TRIGA Reactor (OSTR). These included individually checking each scram. The transient rod was used to test the operability of the manual scram. All the scrams checked out and functioned properl At about 8:41 a.m., a reactor operator (RO) trainee, under the direction of a licensed RO, began start-up and completed the daily core excess measurement. This was done with the reactor power level at 15 watts. At approximately 8:55 a.m., after completing the core excess operation, the RO instructed the trainee to manually scram the reactor.
,
When the trainee attempted to comply by pushing the manual scram button, the I
reactor failed to scram. The RO immediately attempted to scram the reactor by pushing the manual scram button. When the scram button failed again, the RO checked the reactor key and felt it move or " click" from a position a little to the right towards the " operate" position (it seemed to have been between the " operate" and the
" reset" position). The RO then pushed the manual scram button again and a proper l reactor scram resulted. The RO secured the reactor key and notified the Reactor Supervisor of the event. The RO estimated that, from the first attempt to scram the
'
reactor to the actual scram, the time was approximately two to three second *
.
.
The Reactor Supervisor subsequently notified the Reactor Administrator who then notified the Radiation Center Director. The Reactor Supervisor and Reactor Administrator declared that the reactor was not to be operated until further notice. The Reactor Administrator and the Director met to discuss the situation and determined that the event was reportable under requirements of the Technical Specifications (TS).
Subsequently the Chairman of the Reactor Operations Committee (ROC) was notified of the event. The entire ROC met during the afternoon and the members were briefed l concerning the event.
! Notification Process On February 17,1998, at approximately 10:40 a.m., the Director of the Radiation Center notified the Senior Project Manager at the NRC of the event by telephon (Later that day at 4:35 p.m., when the licensee had obtained more information concerning the event, the NRC Operations Center was contacted.) The Director stated that the OSTR had been operated for a brief period that morning without all the scram functions required for shutdown of the OSTR being available. That report was promptly supplemented by a letter issued the same day. The letter was written to confirm that the licensee was taking or planning to take various actions. Those actions included:
o Completing a fullinvestigation of the event to find the most basic root cause of the failure of the OSTR to scram on deman * Developing corrective actions as warrante * Maintaining the reactor in a shutdown condition and not restarting until the problem was correcte * Conferring with the NRC before restar . Licensee Review of the Event and Root Cause Determination immediately following the event, the licensee initiated an investigation of the problem and a review of various documents to determine the cause. The documents reviewed included:
e Reactor installation log, e Reactor console log books covering the period from initial start-up of the OSTR in l 1967 through the present, o Reactor Supervisor's log books from start-up through the present, e 10 CFR 50.59 Safety Evaluations completed for the OSTR since 1968, e Electrical wiring diagrams for the reactor console, e Active and inactive experiment log books, and
- Meeting minutes of the RO The licensee also contacted the Reactor Supervisor, the RO, and the Health Physicist who were working at the facility when the OSTR first went criticalin March 196 !
They were asked whether they recalled anything about changes to the reactor console wiring. None could remember any pertinent detail . _ _ _ - _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ - _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _
_
a
.
.
3-Based on the results of this investigation, the licensee determined that there were two aspects of the root cause of the proble First, the reactor console key switch had apparently become stuck in the " reset" position. This was evidently due to a long-term buildup of dirt in the switch. On the morning of February 17, the buildup prevented the switch from returning to the
" operate" position after it had been switched to the " reset" position according to procedur Second, the licensee determined that the console was not configured as shown in the General Atomics (GA) Instrument Maintenance Manual wiring diagrams nor in any drawings showing subsequent changes to the wiring diagrams. One Jumper as shown on the original wiring diagrams (see Figure 1) was missing and a second jumper had been installed (see Figure 2). The consequence was that an alternate path was created to supply power to the control rod magnets. With this wiring configuration in place, when the key switch was in the " reset" position, the scram bus was disabled. As a result, instead of the magnet power being disrupted, the alternate power source kept the magnets energized resulting in the failure to scra The licensee concluded that, during the event, the key switch became stuck in the
" reset" position and no required scrams were operable due to the change in the wiring of the console. The licensee was not able to determine who had made the change to the wiring nor when the actual change was mad . Corrective Actions (69001)
The initial corrective actions taken by the licensee included:
- Verifying that the scrams functioned properly following the event, e Examining and cleaning the reactor key switch, e Conducting a physical " point to point" check of the reactor safety system circuitry to determine the "as-built" status, and e Electronically checking the scram loop circuitry of the reactor consol During the event review and evaluation, the licensee determined that other actions would be needed to prevent recurrence of the problem. However, these actions would involve a change to the reactor console wiring and one OSTR Operation Procedure (OSTROP). Consequently, a 10 CFR 50.59 Safety Evaluation was complete On February 20,1998, a meeting of the ROC was convened. The licensee presented the results of their investigation to the ROC and the proposed changes to the console and procedure. This was done according to their procedure governing 10 CFR 50.59 safety reviews. The licensee proposed the following supplemental actions: Modify the reactor console circuitry to make it consistent with that shown and evaluated in the original design drawings for the OSTR.
b e ._ _
.
.
.
4 Electronically check the remaining portions of the safety system circuitry to provide assurance that the as-built condition matches the circuitry shown in the facility documentatio Complete OSTROP 15.11, Semi-Annual Surveillance and Maintenance Procedures, Revision (Rev.) 12/94, Section 15.11, Console Checklist to verify proper functioning and voltage of the console, the power range monitor, and the left-hand drawe Modify the start-up procedure, OSTROP 2, Reactor Startup Checklist Procedures, Rev. 2/95, to add a scram test that would confirm that the control rod magnetic power is de-energized when the console key switch is in the " reset" positio The ROC considered the recommended changes proposed by the licensee staff and voted to approve the change On February 20,1998, the licensee also submitted a letter to the NRC describing the event and the corrective actions to be taken. On February 24,1998, a telephone conference was held with the licensee, NRC management, and the inspector participating. The event was discussed further and the completion of the corrective actions was verbally verified by the licensee. (These actions and the subsequent operation of the reactor were reviewed during the second part of the NRC inspection on May 11-13,1998.) { Safety Significance The licensee suggested that the event was not very significant because, despite the wiring problem, the reactor could still have been scrammed by other means if the switch had remained stuck in the "seset" position. This could have been accomplished by:
I pushing the " CONT /ON" buttons for each of the control rod electromagnets thus '
interrupting control rod magnet power, pushing the " AIR" button for the transient rod, turning the key switch to the "off" position, or turning the console power "off."
The licensee also suggested that the event was not very significant because the reactor was only operating at a power level of 15 watts and had only operated for a few minutes. The licensee asserted that, even if the reactor had been operating at full operational power of 1 MW, any problem with reactor operation would have been observed quickly by the RO. The RO would have acted in the same menner as he did during the event and the reactor would have been shut down within two or three second I
'
.
.
I
-
l The licensee also stated that there is an inherent basic safety feature of all TRIGA reactors. This is the large prompt negative temperature coefficient of reactivity that becomes increasingly negative as the fuel temperature increases. Any uncontrolled power rise is automatically limited by this temperature feedback effec i NRC Review (69001)
' Insoection Scoos i The initial part of th!s special inspection was conducted to assure that the events and consequences were as reported by the licensee and to assure that the licensee ;
understood the root cause of the problem. The subsequent part of the inspection !
was conducted to review the corrective actions taken by the licensee and the )
resumption of operations of the reacto j j
The inspector reviewed the following to ensure that the requirements of 10 CFR 50.59, TS Section 3.5.3, TS Section 6.6.k, and TS Sections 6.7.a.4 and 6.7. were met: I
- Selected Surveillance records for the reactor consolo and associated systems, e Selected operator training and requalification records, e Selected records documenting the meetings held by the ROC and the issues reviewod, i e The Reactor Construction Log Book dated August 15,1966, through March 6, i 1967, o Selected records of the contractor, General Atomics, detailing the reactor construction and initial testing dated from November 1966 through March 1967 (these records were reviewed during an inspection of the General Atomics facility in California during the week of May 4,1998), )
- Selected portions of the Reactor Console Log Books, l e Selected portions of the Reactor Supervisor's Log Books, l e Selected 50.59 Safety Evaluations completed for the OSTR since the initial ,
evaluation in 1968, l e 10 CFR 50.59 Safety Evaluation, Number 98-2, dated February 20,1998, ;
- Electrical wiring diagrams for the reactor console and various updates of the I diagrams, j e Oregon State University's (OSU's) letter to the NRC dated February 17,1998, i e OSU's letter to the NRC dated ebruary 20,1998,and
- OSU's 14-day report to the NRC dated March 2,199 The inspector also interviewed various licensee personnel, including the RO on duty during the event, verified that the corrective actions outlined by the licensee had been completed, and observed various actions taken by the licensee that included:
e Examining the reactor key switch, i e Electronically checking the circuitry of the reactor console,
I I l
.
'
.
i I' 6
- Physically checking the circuitry of the reactor console, and l
- Observing the review of the event and the proposed 10 CFR 50.59 Safety l Evaluation, Number 98-2, by the RO Observations and Findinos 1) Reporting of the Event After the Director was briefed on the event on February 17,1998, he made a preliminary notification to the Chairman of the ROC and notified the NRC Operations Center. He also notified the Senior Project Manager at the NRC of the event. That afternoon, the Director wrote a letter to the Ni;C agreeing to investigate the event and determine the root cause. This notification met the requirements in TS 6.7.a.4 and 6.7. During the February inspection, the licensee indicated that a written report would be submitted to the NRC within 14 days of the event to meet the TS Section 6.7.b requirement. A 14-day report was subsequently submitted to the NRC on March 2,1998. A review of the 14-day report during the May inspection was completed. The report reiterated the event and the notifications that were made, provided the licensee's determination of the root causes of the event, and outlined the corrective actions taken to prevent recurrence. The report also specified the licensee's evaluation of the safety significance of the event. The report was acceptable to fulfill the requirements specified in the licensee's T ) Operation of the Reactor Without Scrams Available On the morning of February 17,1998, the reactor checkout was completed by procedure before start-up which confirmed the proper operation of the safety system, including the automatic scrams and the manual scram. The daily core excess measurement was initiated at 8:41 a.m. and was done at a power level of 15 watts as prescribed by OSTROP 4, Reactor Operation Procedures, Rev. O, dated March 1996. At 8:55 a.m., the problem with the nonfunctional scrams was noted. Following the guidance in OSTROP 1, Emergency Operating Procedures, Rev. O, dated March 1996, the RO scrammed the reactor a few seconds later. Therefore, the reactor was operated without scram protection for about 14 minute ) Review of the Actions Taken by the Licensee As noted above, the inspector reviewed selected OSTR procedures, ROC meeting minutes, reactor console log book entries, and electrical wiring diagrams for the reactor console. The inspector also observed as the licensee electronically checked the scram loop circuitry. On February 20,1998, the inspector attended a meeting cf the ROC Uuring which the ROC reviewed the 10 CFR 50.59 Safety Evaluation describing proposed corrective actions. No problems were noted with the procedures, minutes, or log books reviewe Also, the 10 CFR 50.59 Safety Evaluation was acceptable according to the licensee's procedure governing such evaluation ._-___ ______ _ -
~
.
.
During the May 11-13 period, the inspector verified that the licensee had completed all the corrective actions stipulated in the OSU letter to the NRC dated March 2,1998. These actions included:
a) Cleaning and relubricating the console key switch, b) Restoring the console wiring to its as-designed condition, c) Physically checking the scram circuitry and remaining portions of the safety system circuitry and demonstrating that the circuitry is as originally designed and approved, d) Adding an item to the reactor startup check procedure, OSTROP 2, which requires withdrawing one of the control rods a short distance and then turning the console key switch to the " reset" position and observing the rod drop due to the magnet current circuit being opened, and e) Completing the semiannual console checklist according to tho appropriate procedure, OSTROP 1 ) Review of Early 10 CFR 50.59 Safety Evaluations and Logs A review of the console wiring showed that the "as found" OSTR circuit differed from the original "as designed" circuit shown in the GA drawings and manual. This suggested that, at some point in time, a change had been made to the circuitry and the licensee's facility wiring diagrams had not been updated to reflect what actually existed. From a review of the lie:ensee's logs and GA's logs deailing the construction and initial testing of the reactor it could not be determined specifically who may have made the change nor when the actual change to the circuitry was complete A review of the licensee's previous 10 CFR 50.59 Safety Evaluations (from 1968 through the present) showed that no safety evaluation had been completed which would have authorized the change to reactor console wirin ) Review of Surveillance, Maintenance, and Training During the review of surveillance related to the console and reactor interlocks, it was noted that the surveillance were completed as require One minor problem was noted with respect to timeliness when the semiannual console check was completed on July 24,1996, but not completed again until May 29,1997. However, the surveillance was completed twice each year as required.
l Maintenance records were initially kept in the Supervisor's Log Book. A listing at the end of the log indicated the type of problem and the page within the log that rnore fully described the situation and the resolution. It was noted that (
I _ _ _ _ . _ _ _ _ _ _
.
.
this type of record bad been superseded and maintenance items were being tracked in the Reactor Event and Non-Routine Maintenance Log. This latter log had been initiated in March of 1997 and was being maintained as require A review of this log did not suggest any prevalent or recurring problems with the console, the wiring, or the scrams or scram circuitr With respect to training, the inspector noted that training and requalification training had been completed and was on schedule as required by the licensee's training and requalification progra ) Apparent Violations identified a) TS 3.5.3 requires that the reactor not be operated unless the safety channels described in Table I are operable. The safety channels in Table I include the automatic and manual scram A review of the event indicated that, as a result of a change to the reactor console wiring at some point in the past, the reactor was operated for approximately 14 minutes without any of the required automatic or manual scrams being available on the mornirig of February 17,199 The licensee was informed that operating the reactor without any of the scrams being available was an apparent violation of TS 3. (VIO 50-243/98-201-01).
b) TS 6.6.k requires that the licensee prepare and retain indefinitely records of updated, corrected, and as-built drawings of the facilit A review of the facility drawings indicated that a change was made to the r; actor console wiring and circuitry and the drawings were not updated 1.nd corrected or retained at the facility reflecting the modificatio :
1 % licensee was informed that failure to maintain updated drawings of the j facility was an apparent violation of TS 6.6.k (VIO 50-243/98-201-02). l
!
c. Conclusions l l
The licensee reported the event of February 17,1998, as required by TS. A 14-Day Report was subsequently submitted on March 2,1998. The reactor was operated for approximately 14 minutes without the required scrams being functional. The licensee completed all the corrective actions as stipulated in the letter to the NRC dated March 2,1998. It could not be deterrnined who may have made the change nor when the actual change to the console circuitry was completed. The surveillance and maintenance of equipment and training of l
operators were being conducted as required. Two apparent violations were noted:
1) the reactor was operated for approximately 14 minutes without the required scrams being functional; and,2) the licensee had not maintained updated f acility
- drawings that reflected the change mad l l
i
- - _ - ---
_ -- _ _ ____ ___ _ __ _ _ _ _ _ _ _ _ _ _ _ _ __ _____--.-__.. _---_---- -----_ _
,
.
.
"
9 Exit interview The inspection scope and results were summarized on February 20 and May~ 13,1998, with licensee personnel. The inspector described the areas inspected and discussed in -
detail the inspection finding No dissenting comments were received from the licensee. The licensee did not identify as proprietary any of the material provided to or reviewed by the inspecto _ - _ _ - _ _ _ _ _ - _ _ - _ _ - - _ - - - - - _ _ - _ _ _ - _ _ _ _ - _ - _ _ _ _ _ _ _ _ - _ _ _ - _ - . . _ _ _ _ _ _ _ _ - _ . _ _ _ _ - _ _ _
.
..
.
PARTIAL LIST OF PERSONS CONTACTED Licensee Emolovees S. Binney, Chairman of Reactor Operations Committee B. Dodd, Director, Radiation Center A. Hall, Reactor Supervisor J. Higginbotham, Reactor Administrator S. Crail, Reactor Operator S. Smith, Scientific Instrument Technician G. Wachs, Senior Reactor Operator INSPECTION PROCEDURE USED IP 69001 Class ll Non-Power Reactors ITEMS OPENED. CLOSED. AND DISCUSSED Ooened 50-243/98 201-01 V!O Operating the reactor without any of the required scrams being availabl /98-201-02 VIO Failure to maintain updated drawings of the facilit Closed None ;
I LIST OF ACRONYMS USED CFR Code of Federal Regulations GA General Atomics Mw Megawatt NPR Non-Power Reactor NRC Nuclear Regulatory Commission NRR Nuclear Reactor Regulation OSTR Oregon State University TRIGA Reactor OSTROP OSTR Operating Procedere OSU Oregon State Un8versity PDR Public Document Room RO Reactor Operator i ROC' Reactor Operations Committee i TS Technical Specification VIO Violation i I
l l'
.a .
f ._ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _
j l Il lll! j
! !
!ij
.
,
.
-
-
.
-
_
)
H (
_
C A
d
) e N 7h- c (
6i t C 1 w A KS
-
- 4
-
- ^ _ K
-
-
' '
d
+O- cM0 'y C CV c '- .
l e
n g
) 3 5 H i ( s C 4 0 e A T 2 d
- K -
d1 eB 1 2 ' ' '_ s hT c A r
e 0 it w p 2 cdOhOCM- ,
S m3D Ju P i
t u
e 9 c v 1 1 r T B3 '^
is , - K i s ' . C im1(
r e 0 0 P e 2 B3 c<0h0C t a
B B3 T
t e
a T r e
T r
e 2 p 0 1 1 1 p 3I- I O 2(
-
2c
-
O P )
N r 2(
P
P (
o 4 h C t 5 6 3 4 c A c 3 3 D g5 - E t a
3 3 S P T i0 w e
P
P P A ET S R
8 R E S y7 m P E e MOR eee e K a r .
2 2 3 l o
'
c S 3%33 33T 1 B 2" B A#AC[ ^ 's .i s
n d a lN O s^ % ~% ^
2 e r
[ o m
K C e u M 1 A
ij
{
o f
o t
r t
x E i F
g a
P
S N {c/c- l
k- (
)
(
N C
B 1 A B
C
,
,
,
-
> ll ' l
ii i{ ! i l i l 1i 1j1 lj'1l\\1 l l 1 l1
.
.
+
..
.
)
H (
C A
d e
) h c
'N it
_
C w A S
2 K
_
.
_
. '.
-
d
_
' n
_
_
h- w CVC GOC u o
_
_
F
. ) 5 H -
(
4 0 ?
C 2
.A T K
-
s
-
s- -
d1]
eBg 2 t
hT t
c 0
CCWCCW i u
iw 0 c S 3 r P 9 i.
_ 1 r K
0 N
) 1
-
'
-
'- e
- 1 B
( B4 T t a
.
.
.
C r T
cc44_Cy
_
.
A 2 e 0 0 B3 p
B 1E B
T T
O T 2 r 0 1 1 3a -
)
o
- t
- P N 2C
2E
( c a
P P C e 4 4 h c A R 5 6 3 - E 3 h D5P T t
i w0 3%
-
6 3 S
" A ET s1 B P P R E s yT
.
e m 2 P E K a r 3 e O e '
c s o k9 9% e r
2 lo g l* ~. ~. u
.
_
1 S
B8 8kdA 2[T$ i s
n o
a m i g
- C e F F^
- f t o x E
O t r
{
a P
'
R -
E
S W O
P l
C gl I
(
B
B C
l'