ML051880028
| ML051880028 | |
| Person / Time | |
|---|---|
| Site: | Oregon State University |
| Issue date: | 06/24/2005 |
| From: | Reese S Oregon State University |
| To: | Document Control Desk, Office of Nuclear Reactor Regulation |
| References | |
| Download: ML051880028 (5) | |
Text
aso.
Oregon State UNIVERSITY June 24, 2005 Radiation Center Oregon State University, 100 Radiation, Corvallis, Oregon 97331-5903 T 541-737-2341 l F 541-737-0480 1http://ne.oregonstate.edufacilities/radiationcenter U. S. Nuclear Regulatory Commission Document Control Desk Washington, DC 20555
Reference:
Oregon State University TRIGA Reactor (OSTR)
Docket No. 50-243, License No. R-106
Subject:
Self-reporting of a Technical Specification Violation Involving an Inoperable Measuring Channel Gentlemen:
We would like to respectfully report to the Commission that on June 10, 2005, the failure of a mechanical stop on a rotary switch permitted an operator to move the switch into an undefined switch position. This resulted in one of our measuring channels being inoperable while the reactor was operating. This was a violation of OSTR Technical Specification 3.5.2, Limiting Conditions ofOperation, Reactor Control Systems.
Reason for the Violation The OSTR has four reactor power measuring channels. All are required to be operable when the reactor is operating in steady-state or square-wave modes. One of these channels, the Safety Power Level Channel (hereafter referred to as the Safety Channel) obtains information from an uncompensated ion chamber. The information for this channel is displayed on an analog meter located in a console instrument panel known as the Left-Hand Drawer (LHD),
shown in Figure 1. The LHD was installed in 1978.
Fuel Temperature Safety - Calibration Channel _ Switch Calibration Switch Period/Log
- Calibration Switch Figure 1 - Left-Hand Drawer on the OSTR Console Pv__ D
OSU Notice of Violation Page 2 of 5 As part of the startup checklist prior to going to power, the operator is required to perform a channel test of the Safety Channel using the Safety Channel Calibration Switch shown in Figure 2. The Safety Channel Calibration Switch has three positions indicated on the front panel. These positions are ZERO, OPERATE, and CALIBRATE. A scram signal is generated when the switch is in the ZERO or CALIBRATE positions because the Safety Channel is not operable in either of these positions. To perform the test, the operator rotates the switch clockwise from OPERATE to CALIBRATE and verifies that 100% power is displayed on the meter. Once this is done, the operator rotates the switch back counter-clockwise from CALIBRATE to OPERATE, clears the annunciated scram that resulted from the switch movement to CALIBRATE, and proceeds with the checklist.
-. *. z Figure 2 - Safety Channel Calibration Switch During the checkout on the morning of June 10, the operator performed the Safety Channel test as required but unknowingly rotated the switch clockwise past CALIBRATE to a undefined position instead of counter-clockwise back to OPERATE after completing the test.
The operator then cleared the scram that ensued from the test and continued with the checklist. At 8:31 AM, the reactor was brought to a power level of 15 W for a core excess measurement. After completion of the core excess measurement at 8:43 AM, the operator began raising power to the desired power level of 1 MW. At approximately 40% of full power (I MW), the operator noticed that the Safety Channel was not responding correctly and manually scrammed the reactor. The time of the manual scram was 8:46AM. It was quickly realized that the Safety Channel Calibration Switch was out of position and reactor operations were immediately suspended. The reactor had operated for a total period of 15 minutes with the Safety Channel in an inoperable condition. This is a violation of a Limiting Conditions of Operation (LCO) identified in OSTR Technical Specification 3.5.2.
OSU Notice of Violation Page 3 of 5 Because one of the four measuring channels is a wide-range channel and the other three display 0-100% power, the operator must wait until reactor power reaches -1% (10 KW) on the way up to 100% power (1 MW) before they can verify that all four power indication channels are reading the same power level. However, with the switch in the undefined position the ion chamber signal is disconnected from the channel. In this configuration, the meter for the channel reads 5% of full power, which is noise picked up by the channel amplifier. The operator could not recognize the inoperable channel until power level had risen higher than the indicated noise.
The operator was able to clear the annunciated scram during the startup checklist because an automatic scram signal is generated (i.e., a 15 V signal on top of the ion chamber signal causing a relay to trip) only when the switch is in positions ZERO or CALIBRATE. When the switch is in the undefined position (i.e., the next position clockwise past the clockwise stop), an automatic scram signal is not generated and the channel, including the associated scram relays, behaves as if it were receiving the ion chamber signal when in fact it is reading noise.
Shown in Figure 3, the Safety Channel Calibration Switch is a rotating 12 position switch.
The switch is designed with one stop you cannot move. This is accomplished by a pressed in stop tab which is part of the rotary body of the switch and serves as the counter-clockwise stop. For the clockwise stop, a washer with a protruding tab is placed on the shaft of the switch with the tab in the appropriate hole, in this case the fifth hole, leaving three available positions.
Stop Washer Tab Example Position Holes Figure 3 - Safety Channel Calibration Switch internals
OSU Notice of Violation Page 4 of 5 We originally thought that the switch was missing a clockwise stop washer. However, upon further investigation, we discovered that the switch was not missing a stop washer but that the tab on the stop washer had been bent to such a point that it rendered the stop washer nonfunctional. There are two other switches on the Left-Hand Drawer that are mechanically identical but serve other functions. These switches are the Fuel Temperature Calibration Switch (this switch is no longer used) and the Period/Log Calibration Switch (produces a Rod Withdrawal Prohibit signal when out of position). Both of these switches also had clockwise stop washers that were bent and nonfunctional.
An initial phone call was made at 1:00 PM on June 10 to Mr. Alexander Adams informing him that we were reporting a LCO violation, our initial observations, solutions, and desire to keep the reactor shutdown until the OSTR Reactor Operations Committee could meet and review the occurrence. The OSTR Reactor Operations Committee met on June 13 to review this occurrence and agreed with the corrective actions. A follow-up phone conversation was also made with Mr. Marvin Mendonca on June 21 informing him of our determination of the cause (i.e., bent tab on the stop washer) and the corrective actions.
We have no clear idea as to when the tabs were bent on the stop washers. Presumably it occurred from the cumulative effect of rotating the switches against their clockwise stops over time. At the same time, we were not looking for this failure as we typically rotate the switch to an operable position, not an undefined position. The movement of the switch by the operator one position clockwise instead of counter-clockwise was certainly a contributing factor leading to the violation. However, we assume that the stop failure did not happen on June 10 because all three switches were in the same condition, including one that has not been functionally used in several years. This indicates that movement of the switch to the undefined location was the rare exception, not the rule.
Corrective Actions Upon discovery, reactor operations were suspended until such time as an adequate solution could be found and the OSTR Reactor Operations Committee could review the incident. The tabs on the stop washers for all three switches were realigned and verified to be functional.
This change was performed and reviewed through the process defined in 10 CFR 50.59. All of the licensed operators have been thoroughly involved with the discovery and rectification of the issue and are aware of its importance.
Corrective Actions to Prevent Reoccurrence After careful review and discussions between the OSTR Staff and the Chairman of the OSTR Reactor Operations Committee, it was decided that a functional test of the switches to verify that they may not be moved to an undefined position shall be performed as part of the semi-annual console checklist found in OSTR Operating Procedure 15, Semi-Annual Surveillance andMaintenanceProcedures. We also considered visual inspection of the tabs in conjunction with a function test, but we feel the benefit of the inspection is outweighed by the increased opportunity for mistakes in reassembly of the switches on a routine basis.
OSU Notice of Violation Page 5 of 5 Conclusion In summary, the above described switch stop failure and movement of the switch into an undefined position resulted in a violation of a LCO channel operability requirement. The importance of the violation is understood and appreciated.
We would like to reiterate our desire to maintain an open and cooperative nature of our relationship with the Commission. Should there be questions regarding the information in this report or should you require more information, please do not hesitate to contact me.
I declare under penalty of perjury that the foregoing is true and correct.
Executed on: Wcz A'c'-
Sincerely, S.X Director cc: Al Adams, USNRC Craig Bassett, USNRC Ken Niles, ODOE John Cassady, OSU Rich Holdren, OSU John Ringle, OSU Todd Keller, OSU Gary Wachs, OSU