IR 05000151/1986001
| ML20213G151 | |
| Person / Time | |
|---|---|
| Site: | University of Illinois |
| Issue date: | 11/03/1986 |
| From: | Greger L, Tambling T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20213G134 | List: |
| References | |
| 50-151-86-01, 50-151-86-1, NUDOCS 8611170346 | |
| Download: ML20213G151 (8) | |
Text
.
..
.
'
U.S. NUCLEAR REGULATORY COMMISSION
REGION III
Report No. 50-151/86-01(DRSS)
Docket No. 50-151 License No. R-115 Licensee: University of Illinois 214 Nuclear Engineering Laboratory 103 S. Goodwin Avenue Urbana, Illinois Facility Name: Advanced TRIGA (ATR)
Inspection At: Urbana, Illinois Inspection Conducted: October 8, 1986 t Inspector: T. N. Tambling /o/3 I /%
Date
'
Approved By: L. R. Greger, Chief Facilities Radiation Protection Section Il!k Ob Date Inspection Summary Inspection on October 8,1986 (Report No. 50-151/86-01(DRSS))
Areas Inspected: A special announced inspection to review the circumstances associated with the discovery and correction of a wiring error associated with a safety interlock on the fast transient rod. The inspection also covered followup on previous inspection item Results: One violation was identified (failure to receive the need to prior NRC approval for continued operation with a wiring error in the safety interlock (Paragraph 4d).
,
.
8611170346 861103 PDR ADOCK 05000151 G PDR
_ - . . _ _ . - _ . --. . .. .-
- '
. .
O
'
DETAILS 1. Persons Contacted
- B. G. Jones, Acting Chairperson, Nuclear Engineering Program
- B. Micklick, Chairman, Nuclear Reactor Committee
- J. G. Williams, Director, Nuclear Reactor Laboratory
- C. S. Pohlod, Reactor Supervisor
- Indicates those present at the exit meetin . Licensee Actions on Previous Inspection Findings (Closed) (50-151/85001-01) Failure to maintain documentation and evaluations on two licensed operator requalification examinations in 1984. By a review of the records and discussion with a licensee representative, it was verified that the two licensed operators were reexamined, the results of the written exams were filed, and the oral exams logged in the daily reactor log book. Although the practice of logging oral exams in the log book has been an accepted practice in the past, the inspector noted that a check list or similar document with dates, times and subjects filed with the operator written exams would provide better objective evidence of the oral exa It was noted that an operator requalification exam given in 1986 had been filed and the oral exam logge (Closed) Deviation (50-151/85001-03) Inoperable constant air particulate monitor (CAM) described in Sections VII.B.B of the Safety Analysis Report. By review of the records and field observation it was verified that the licensee had received and installed two operable CAMS (May 30, 1985 and August 20,1985) in place of the original failed CAM. It was noted that action had been initiated in 1983 for the two new CAMS as replacements, but they were not received until 198 . Organization Change The University is in the process of making organizational changes to their Nuclear Program. Representatives of the licensee indicated that they will communicate this change in accordance with regulatory requirements. This item is considered open pending the reporting of the change to the NRC (50-151/86001-01).
4. Standard Control Rod Position Interlock Background On September 9, 1986, the Reactor Supervisor for the University of Illinois TRIGA reactor reported to Region III by telephone (approximately 1030 hours0.0119 days <br />0.286 hours <br />0.0017 weeks <br />3.91915e-4 months <br />) and telegraph that they had discovered an apparent wiring error that could circumvent the operation of the Standard Control Rod Position interlock. The discovery was made on
.
- *
. .
.
September 8, 1986 at approximately 1300 hours0.015 days <br />0.361 hours <br />0.00215 weeks <br />4.9465e-4 months <br /> during the performance of the Control Rod Drop Time Semi-annual Surveillance. The initial report was followed by the required 10-day written report dated September 17, 198 The Standard Control Rod Position interlock is required by Technical Specification 3.5. As described in the Safety Analysis Report. The interlock was incorporated in the reactor control design to prevent an inadvertent pulse operation when the reactor was operational with the mode switch in the Steady-state position. The interlock was to prevent possible operator error. Also it was to prevent the potential loss of operational data since reactor instrumentation would not be properly aligned to insure that pulse parameters were recorde b. Discovery of the Interlock Defect At approximately 1:00 p.m. on September 8,1986, during the performance of the Control Rod Drop Time Semi-annual Surveillance, the reactor operator discovered that the ready fast transient rod light for the (FTR) came "on" when either the safety or shim reached their full up position with the respective "Up" light o This meant that the FTR could be fired even though the reactor mode switch was in the steady-state position and either the safety or shim rod was withdrawn from its full down positio (The assumption was not actually verified by trying to fire the FTR.) The reactor operator brought this to the attention of the Reactor Supervisor. The twa initiated an immediate investigation into the cause and safety consequence of the defect in the interloc While this particular defect in the interlock operation could have been discovered previously during past performances of the semi-annual control rod drop time testing, it apparently had not bee Detection on September 8, 1986, was probably enhanced by the fact that general control room lighting was off at the time. The dim background light accentuated the control panel lights. Since rod drop time testing is independent of the interlock, operator attention normally would be on the individual control rod light The lights are mounted on the vertical pc- ion of the control pane Normally, operator attention would not be on the recessed FTR ready light and pulse button which is mounted on a horizontal portion of the control pane Normal interlock surveillance checks only test the interlock functions by withdrawing the control rod off the fully inserted positio c. Corrective Action The defect was traced to an apparent wiring error in the design totarlock circuit for the safety and shim rods, and as such existed since original assembly of the control system. The "Up" limit switch on these two rods had been wired in series with the interlock circuit using a normally closed contact. When either rod reached its upper
l
!
.
..
.
limit of travel the normally closed contact opened circumventing the interlock. In this one specific case with either rod or both together withdrawn to their full up position and all other rods in their fully inserted position, the interlock would not meet its design intent to prevent the inadvertent firing of the FT The schematic of the control wiring indicated a very simple fix for the apparent wiring error. This fix was verified using a temporary jumper to bypass the specific contacts on the upper limit switches for the safety and shim rods. The temporary jumper was removed pending a thorough design review by G. A. Technologies (designer of the reactor) and the review and approval of an wiring change by the Nuclear Reactor Committe The Reactor Supervisor analyzed the safety consequence if the FTR *
was fired with either or both the safety and shim rods fully withdraw This analysis showed that with both the safety and shim rods fully withdrawn, the reactor would be subcritical by $1.92 reactivity (shutdown by definition is 51.00 negative reactivity). If the FTR was inadvertently fired at this time, it would add approximately
$2.55 reactivity. The net positive reactivity would be approximately 50.63. This would result in a peak power pulse of approximately 275 KW and a resultant steady-state power of approximately 95 K G. A. Technologies was requested by telephone on Septembe.- 8,1986, to review the apparent wiring error and the proposed fi It was reported that GA's review and concurrence on the corrective action was delayed while they located their technical manual on the University of Illinois reacto On September 8,1986, the Reactor Supervisor also wrote a Standing Order regarding the interlock wiring error that administrative 1y restricted control rod operation that could circumvent the interlock function. The Standing Order was discussed with each reactor operato Each operator initialled that they had read the order. The Standing Order remained in effect until G. A. Technologies approved the wiring change and the defect in the interlock was permanently corrected and tested on September 10, 198 The Nuclear Reactor Committee met at 1300 hours0.015 days <br />0.361 hours <br />0.00215 weeks <br />4.9465e-4 months <br /> on September 9, 1986, for approximately one hour and 20 minutes to review the interlock wiring error, its effect on operations, the safety analysis performed by the Reactor Supervisor, the Standing Order issued September 8, 1986, the proposed wiring modification, and NRC reporting require. tents. A l review of the Committee meeting minutes indicated that the Committee approved the proposed wiring modification pending G. A. Technologies review. They reviewed the use of a temporary jumper, but considered it unwise because it could come loose. They concurred in continued operation in view of the lack of any hazard associated with the FTR being fired, the Standing Order, the existence of a corrective remedy, the thorough briefing of the operators, and the fact that this knowledge of the problem actually increased the safety of the operation.
l
-,
____
.
..
l
.
'
On September 10, 1986, after G. A. Technologies verbally concurred in the proposed fix, the interlock control circuit was modified to bypass the leads to the safety and shim rod "Up" limit contact Also, the leads to the contacts were disconnected at the control card. The control rods were put through all their checks to verify that all parts of the interlock functioned properl After being notified by the University of Illinois, G. A. Technologies notified the University of California of the problem on September 10, 1986, and on September 15, 1986, notified Region V of a Part 21 Report on the wiring defect followed with a written report on September 18, 198 d. Continued Reactor Operation After identifying the cause for the interlock defect, its effects on control rod operation, and the safety consequence if the FTR was inadvertently fired with the safety and shim rods fully withdrawn, the Reactor Supervisor authorized continued operation of the reactor on September 8, 1986, under a Standing Order. The Nuclear Reactor Committee approved the continued operations under the Standing Order on September 9, 1986. This order administratively limited the withdrawal of the safety and shim rods in a way that the interlock could not be circumvented. The defect was discovered on September 8, 1986 at approximately 1300 hours0.015 days <br />0.361 hours <br />0.00215 weeks <br />4.9465e-4 months <br /> according to the licensee's repor Under the limitations of the Standing Order, the reactor daily log book indicates that the reactor was operated twice on September 8 and three times on September 9, 1986, at the following times:
Date Start Approach Type of Power Time of to Critical Operation Level Shutdown 9/8/86 1347 hrs Steady-state 600 KW 1441 hrs 9/8/86 1448 hrs Steady-state 600 KW 1520 hrs 9/9/86 1438 hrs Steady-state 250 KW 1458 hrs 9/9/86 1504 hrs Steady-state 250 KW 1516 hrs 9/9/86 1525 hrs Steady-state 600 KW 1609 hrs 9/10/86 No Operations On September 10, 1986, the reactor was not operated and the interlock wiring defect was corrected and tested to bring it into complete conformance with the design inten All operations on September 8 and 9, 1986, involved routine sample irradiation. According to representatives of the licensee there was no urgency or compelling reasons why the samples had to be irradiated those days other than they were scheduled and they saw no safety reasons for not performing these irradiations as planne Four members of the Nuclear Reactor Committee stated that they believed that the operational limitation in the Standing Order was an acceptable resolution to the problem. This decision was based, in part, on information provided them that this was an acceptable practice based upon a previous proble ,
.- 1
,,
.
t 10 CFR 50.59, Changes, Tests and Experiments, requires prior Commission approval of any change in the facility or procedures described in the safety analysis report if it involves a change in the Technical Specifications incorporated in the licens Any change in the Technical Specifications must be submitted as an application for an amendment to the licensee. On September 8, 1986, the Reactor Supervisor and on September 9, 1986, the Nuclear Reactor Committee authorized under a Standing Order operation of the reactor with the Standard Control Rod Position interlock in a condition different from that described in the Safety Analysis Report for the Illinois Advanced TRIGA dated August, 1967 as amended by Change No. 3 to the Technical Specifications dated December 21, 197 Since the change also involved a change to the Standard Control Rod Position interlock required by Technical Specifications 3.5, Reactor Safety System, prior Commission approval was required to operate with the known defect in the interlock. This is considered to be a violation of 10 CFR 50.59 (50-151/86001-02).
5. Housekeeping During previous inspections of the reactor facility, housekeeping was noted to be poor. During a tour of the facility, significant improvements in housekeeping were noted. Reactor process areas were being well maintained; however, some individual experimenter areas still needed attention. The licensee indicated that they are pursuing this with the individual experimenter . Followup to the Operator Licensing Examiner Report (50-151/0L-86-01)
On January 16, 1986, the chief examiner held an exit meeting and made several remarks concerning the reactor facilit During this inspection it was noted that the following measures had been taken in response to the examiner's remarks: The locking mechanisms on two beam ports had been modified to provide adequate lockin The license now has a procedure to cover Tornado Respons The licensee decided not to color code piping because of the simple process desig . Exit Meeting The inspector met with the licensee representatives (listed in Paragraph 1)
on October 8, 1986, and summarized the scope and findings of the inspectio . - -- _ -_ _ .
.
..
.
The licensee acknowledged the inspector's remarks regarding the apparent
'
violation associated with the continued operation of the reactor on September 8 and 9, 1986, with a known defect in the Standard Control Rod Position interlock. The licensee provide the following for justification of their action at the time: The cause for the interlock defect was identified and understoo The proposed wiring change was purposely held up as a precaution to get G. A. Technologies concurrence that nothing had been overlooked that would create another safety concer The safety concerns had been analyzed and an inadvertent firing of the FTR was well within safety and operational limitations, It was felt that the use of a Standing Order was an acceptable method to control operation in their situation until the wiring change was approved by G. A. Technologies. A Standing Order had been used under similar circumstances in the pas It would require a deliberate action by the operator contrary to normal operating requirements to circumvent the interlock, While there was no compelling reason to continue operations on September 8 and 9, 1986, the licensee stated that their determination that no safety concern existed led them to continue reactor operation On October 28, 1986, the inspectors also discussed the likely informational content of the inspection report with regard to documents of processes reviewed by the inspectors during the inspection. The licensee did not identify any documents or processes as proprietar . Enforcement Conference On October 28, 1986, an enforcement conference was held in the Region III Office to discuss the result of the inspection, the apparent violation, and the NRC enforcement option Licensee Attendees:
Dr. B. G. Jones, Arting Chairperson, Nuclear Engineering Program Dr. B. Micklick, Chairman, Nuclear Reactor Committee Dr. John Williams, Director, Nuclear Reactor Laboratory Craig Pohlod, Reactor Supervisor NRC Attendees:
A. B. Davis, Deputy Regional Administrator W. D. Shafer, Chief, Emergency Preparedness and Radiological Protection Branch -
L. R. Greger, Chief, Facilities Radiation Protection Section T. N. Tambling D. E. Miller, Senior Radiation Specialist
,
l
-*
..
l
.
Representatives of the NRC explained that in this particular instance
~
'
the issue was not the safe operation of the reactor, but the failure of the Reactor Supervisor and the Nuclear Reactor Committee to recognize l the legal operating requirements of their license and the regulation It was acknowledged that an appropriate safety analysis is a primary priority. However, it was pointed out that a safety analysis demonstrating the lack of a safety problem does not by itself alleviate a licensee from the terms and conditions of the license regulatory requirement The licensee acknowledged the NRC concern, and stated that it was not the intent of the University staff to ignore regulatory requirement The licensee claimed they failed to fully understand the NRC's interpretation of some of the regulations. When the decision was made to continue operation, it was felt that regulatory requirements were being met. Now that the staff understands NRC's interpretations, there should be no future regulatory compliance problem The licensee provided preliminary information on probable structure and organization changes in Nuclear Engineering programs. Since these changes could involve changes to the organization described in the technical specifications, a appropriate license amendment will be required. (This is carried as an Open Item. See Paragraph 3.)
.
8