ML17279A109
| ML17279A109 | |
| Person / Time | |
|---|---|
| Site: | Kansas State University |
| Issue date: | 09/28/2017 |
| From: | Bahadori A Kansas State University |
| To: | Document Control Desk, Office of Nuclear Reactor Regulation |
| References | |
| Download: ML17279A109 (4) | |
Text
,. KANSAS STATE TRIGA Mk II Nuclear U N I v E R s 1 T y Reactor Laboratory US Nuclear Regulatory Commission Washington, DC 20555-0001 September 28, 2017
Subject:
Reportable Occurrence on September 19, 2017, at the KSU TRIGA Mark II Nuclear Reactor Facility To Whom It May Concern: On September 20, 2017, I notified the NRC of a reportable occurrence that occurred at our facility on September 19, 2017. The reportable occurrence involves a violation of one of our facility limiting conditions for operations (LCOs), as documented in our Technical Specifications (TS). The LCO in question is TS 3.5, Gaseous Effluent Control. TS 3.5.3(1) states: "The reactor bay ventilation exhaust system SHALL maintain in-leakage to the reactor bay". The applicable condition is: "A. The reactor bay ventilation system is not OPERABLE". The required action, which was not taken, is "A.2.b. Secure EXPERIMENT operations for EXPERIMENT with failure modes that could result in the release ofradioactive gases or aerosols." The purpose of this letter is to fully document the reportable occurrence, in accordance with reporting requirements. Background The exhaust plenum monitor (EPM) consists of three detectors used to measure radioactive particulate, iodine, and noble gases. The EPM has a relay which controls the ventilation system in the reactor bay. If radiation levels are below the alarm set-points, the ventilation system continually exhausts reactor bay atmosphere to prevent the build-up ofradioactive gases, which could cause increased radiation dose to reactor staff. When the EPM detects high radiation levels, or is secured, the ventilation system is turned off to prevent exhausting excess levels of radioactive gases. The EPM must be calibrated on a regular basis to ensure it remains operable. On September 1, 2017, the Reactor Supervisor notified the Reactor Manager that the EPM was due to be sent out for calibration. In reviewing the Technical Specifications, the Reactor Manager determined that while operations were permitted according to TS 3.3, Measuring Channels, and TS 3.5, Gaseous Effluent Control, pulse mode operations, irradiated fuel movement, and operations involving experiments with failure modes that could result in the release of radioactive gases or aerosols were prohibited, since the ventilation system was not considered to be operable (defined in the TS as "capable of performing its intended function in a normal manner"). The possibility of operating the ventilation system using the control room breaker box was suggested by the Reactor Supervisor, and rejected by the Reactor Manager as not consistent with the TS definition of"operable". The restrictions on operations were noted in the reactor logbook, and the EPM was removed from service and shipped out for calibration. On September 18, 2017, the EPM was received back at KSU and returned to service. The noble gas monitor was observed to be giving spurious alarms while the reactor was shut down, which were cleared when the alarm was O \bi 7:i acknowledged. Operational restrictions were lifted by the Reactor Supervisor as noted in the logbook at 9:52 AM CDT on September 18, 2017. Reportable Occurrence Description On September 18, 2017, the Reactor Supervisor designated a senior reactor operator (SRO 1) as in charge of reactor operations for the rest of the week. On September 18, the EPM noble gas monitor continued to give spurious alarms. SRO 1 contacted Thermo Scientific for assistance in diagnosing the issue with the system. After informing the Thermo Scientific representative of various detector operation parameters, it was decided that a more in-depth investigation of the detector malfunction was required. The Thermo Scientific representative gave SRO 1 the contact information for a calibration technician. At approximately 3:00 PM CDT, SRO 1 instructed SRO 2 to secure the EPM noble gas monitor and subsequently the exhaust fan. On the morning of September 19, 2017 at approximately 9:00 AM CDT, SRO 3 contacted SRO 1 about the status of the EPM while conducting the daily checkout. SRO 1 informed SRO 3 that the EPM noble gas monitor was offline, and that efforts were underway to rectify the issue. SRO 1 instructed SRO 3 to indicate that the EPM was operational with the assumption that Thermo Scientific would be able to provide a solution later that day. SRO 1 was in contact with Thermo Scientific three separate times on the morning of September 19, 2017. Calls occurred at 9:26 AM, 10:19 AM, and 10:43 AM CDT. During conversations with Thermo Scientific, the representative asked for the calibration certification for the EPM noble gas monitor and alluded to further contact occurring in the early afternoon of that business day. On September 19, 2017, at approximately 1 :40 PM CDT, SRO 1 arrived in the reactor bay to prepare for planned operations as part of a detector demonstration for a client. SRO 1 concluded that with two of the three units operational, and the continuous air monitor (CAM) operational, detection of radioactive gases was still possible .. SRO 2 rewired the EPM relay to take input from the two operating EPM units. Once rewiring was completed, the ventilation system breaker was returned to the on position, but the exhaust fan did not activate as anticipated. Upon inspection of the ventilation system breaker, a faulty relay was discovered. Based on this assessment, and with the knowledge that the relay was not specified in any documents as a part of the ventilation system design, SRO 1 instructed SRO 2 to perform a 10 CFR 50.59 screening, and then rewire the system to bypass the ventilation system breaker relay. It should be noted that this action did not adhere to the management order regarding 10 CFR 50.59 screenings, which require Reactor Safeguards Committee approval prior to implementation. Since the 10 CFR 50.59 screening did not indicate the need for further evaluation, the ventilation system breaker relay was bypassed and ventilation system was operated from the control room via local breaker. Operations on the day in question involve an experiment that has failure modes involving the release of radioactive gases, namely argon-41. The experimental detector array is composed of a 20-foot long tube of stainless steel which is welded at the bottom and sealed with Swagelok fittings at the top. The detector array receives argon from a pressurized canister connected via plastic tubing. The top of the array has an isolation ball valve. This can be manually opened and closed by an individual with access to the 22-foot level of the reactor bay. Another failure mode 2 *'-.1
. ,* I is bending of the tube which could lead to breaking. The tube was secured to the transient rod enclosure by zip ties and the length of the tube descends into the central thimble. During normal operation, it is unlikely the tube would be put in a condition which would cause a breakage to occur. A third mode of failure is the Swagelok components affixed to the top of the array which could be damaged or unlocked. During normal operation, this failure should not occur. However, the TS in question does not address the probability of occurrence, only potential failure modes. Therefore, in order for the experiment to be secured for operations, there must be no gas flowing from the canister to the array. Operations began at 2:49 PM CDT on September 19, 2017. During operations, SRO 1 monitored the CAM, EPM, and radiation monitors on the 22-foot level of the reactor bay, prepared to cut power to the ventilation system. During operations, the experiment was physically secured and not moved. However, there was argon gas flowing from the canister to the array. At approximately 5: 3 0 PM CDT, the Reactor Supervisor became aware of the condition of operations from SRO 2, who had contacted the Reactor Supervisor to ask an unrelated question. The Reactor Supervisor then notified the Reactor Manager by phone message. Operations concluded at 5:43 PM CDT on September 19, 2017. No indication of elevated radiation levels was observed during or after reactor operations. The Reactor Manager received the phone message at 7:44 PM CDT on September 19, 2017, and discussed the issue with the Reactor Supervisor, ultimately concluding that it should be considered a reportable occurrence. Immediate Corrective Action and Efficacy Upon discovery of the reportable occurrence and after consulting with the Reactor Manager, the Reactor Supervisor communicated to SRO 1 that operations were suspended effective immediately approximately 10:30 PM CST on September 19, 2017. This action is required by TS upon discovery of a suspected LCO violation. The reactor keys were removed by SRO 1 from the SRO safe to prevent use by any other SROs. The immediate corrective action was therefore effective at preventing further operations that could violate TS. Summary of Findings This reportable occurrence was caused by several factors. 1. The dependence of "normal operation" of the ventilation system on the EPM relay is not sufficiently documented in the facility Technical Specifications or the Safety Analysis Report. 2. A lack of clear documentation on the prohibition on operating the ventilation system using the control room breaker only. 3. Ignorance on the part of SRO 1 regarding the Reactor Manager's prohibition of operations with the ventilation system not controlled by the EPM. 4. Ignorance on the part of SRO 1 and SRO 2 regarding the management order covering 10 CFR 50.59 screening and evaluation. Additional Corrective Actions 3 l Additional corrective actions in response to the documented findings will be taken to prevent recurrence of this reportable occurrence. 1. All reactor staff will be briefed on the occurrence in a staff meeting. All reactor staff have already been notified by email of the reportable occurrence. Estimated Completion Date 09/28/2017. 2. All reactor staff will be retrained on operation of the EPM, ventilation system, and 10 CFR 50.59 screening and evaluation requirements at this facility. Estimated Completion Date 10/25/2017. 3. All management orders and other changes in procedures/experiments will be explicitly explained to all reactor staff in person at the earliest possible time after implementation. Will be done on an as-needed basis. 4. A management order will be mitten prohibiting operation of the ventilatio.J;l system using the control room breaker, if the EPM is not also properly connected to the ventilation system. Estimated Completion Date 09/29/2017. 5. Information on the operation of the ventilation system using the EPM relay will be added to appropriate facility document(s). Estimated Completion Date 12/31/2017. 6. The EPM will be restored to operable status within 3 0 days of September 19, 2017, or the reactor will be shutdown in accordance with TS 3.3.4(F.2). 7. The reactor bay ventilation system will be restored to operable status within 30 days of September 19, 2017, or the reactor will be shutdown in accordance with TS 3.5.4(A.1). The Kansas State University Reactor Safeguards Committee approved* the resumption of operations on September 25, 2017. Please do not hesitate to contact me if additional information is required. I swear under penalty of perjury that the foregoing is true and correct. Regards, Amir A. Bahadori, PhD, DABHP Interim Manager, KSU TRIGA Mark II Nuclear Reactor Facility Department of Mechanical and Nuclear Engineering Kansas State University Manhattan, KS 66506 Phone: (785)532-7040 (W) (913)306-4249 (C) Fax: (785)532-7057 Email: bahadori@ksu.edu 4 .* ., *