ML091170048

Reportable Event 44991; Violation of Limiting Conditions of Operation in Technical Specification 3.5B

ML091170048
Person / Time
Site:	North Carolina State University
Issue date:	04/23/2009
From:	Hawari A North Carolina State University
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
44991
Download: ML091170048 (11)

Text

North Carolina State University is a land-grant university and a constituent institution of The University of North Carolina Departnent of Nuclear Engineering NCISTATE-M, IVý RSIf, College of Engineering Campus Box 7909 Raleigh, NC 27695-7909 919.515.2301 (TEL) 919.515.5115 (FAX) 23 April 2009 US Nuclear Regulatory Commission Document Control Desk Washington, DC Re:

Reportable Event 44991; Violation of Limiting Conditions of Operation in Technical Specification 3.5B License No. R-120 Docket No. 50-297 Attached please find a written report with a detailed explanation of Reportable Event 44991 "Violation of Limiting Conditions of Operation in Technical Specification 3.5B" and actions taken and planned to prevent recurrence.

If you have any questions regarding this report or require additional information, please contact Gerald Wicks at 919-515-4601 or wicksgncsu.edu.

I declare under penalty of perjury that the forgoing is true and correct. Executed on 23 April 2009.

Sincerely, Ayman I. Hawari, Ph.D.

Director, Nuclear Reactor Program North Carolina State University

Enclosures:

Reportable Event 44991; Violation of Limiting Conditions of Operation in Technical Specification 3.5B cc:

Duane Hardesty, US NRC zr~P~L

Event Discussion:

The stack particulate radiation monitor reading was noticed as being abnormally low at approximately 9 AM on 13 Apr 2009. Upon checking this abnormality, the stack sample pump was noticed as being off.

The radiation recorder data was reviewed and it was discovered that the stack sample pump lost power at approximately 9:30 AM on 10 Apr 2009. This observation is based on the sudden decrease in count rate for the stack particulate monitor (refer to attached recorder data). The operations log was reviewed and indicated that the reactor was operated on 10 Apr 2009 from 11:09 AM to 4:05 PM. The reactor was shut down over the weekend on 11 Apr and 12 Apr 2009. The stack sample pump was started at approximately 9:30 AM on 13 Apr 2009 by pulling the power switch to the on position. The circuit breaker for the stack sample pump was already on.

Reactor Operations was informed about these observations at approximately 10 AM on 13 Apr 2009.

This situation was determined to be a reportable event at approximately 10:30 AM on 13 Apr 2009.

Information about this event was gathered on 13 Apr and 14 Apr 2009.

Prior to each reactor startup, a checklist is performed to verify that all requirements for reactor operation are met. On 10 Apr 2009, the checklist was performed. The stack sample pump, stack gas monitor, and stack particulate monitor were verified to be operable prior to 9:30 AM on 10 Apr 2009.

After the stack sample pump and monitors were verified to be operable, checks on the Control Room Console began. While performing the Control Room Console annunciator panel test at approximately 9:30 AM, the Source Range Inhibit bulb did not come on. Upon troubleshooting, the card for the Source Range Inhibit board shorted out. The Source Range Inhibit card was swapped out with another board and the annunciator panel was successfully tested at approximately 10:15 AM. Upon completion of the startup checklist, reactor operation commenced at 11:09 AM and continued until 4:05 PM on 10 Apr 2009.

Radiation monitoring system logs were initiated at 11:08 AM on 10 Apr 2009 and were taken every 30 minutes as required by operating procedures. Initially a reading of 1.3 K cpm (1300 cpm) was recorded at 11:08 AM, and the readings decreased through the day to a few hundred cpm. Normally, this channel has steady reading of several thousand cpm (- 3 K cpm to 5 K cpm).

At 9:30 AM on 10 Apr 2009, the stack sample pump apparently lost power rendering the stack gas and stack particulate radiation monitor channels inoperable. This time coincides with the failed annunciator panel test. The stack sample pump and control room console receive power from the Control Room Distribution Panel using separate circuits.

Power loss to the stack sample pump was not noticed by the operators performing the checklist or by any of the reactor staff during operation of the reactor on 10 Apr 2009. The stack sample pump power light is located in the Control Room above the stack gas and stack particulate meters. The light is green in color. There is also a red indicator light that turns on if the flow rate drops below 2 cfm if the stack sample pump is on. If the stack sample pump is off, the red low flow light will also be off.

Configuration of the stack sample pump power and flow rate alarms are summarized below:

1.

If both the green power light and red flow light are off, the stack sample pump has failed. This condition would render both the stack gas and stack particulate channels inoperable.

2.

If both the green power light and red flow light are on, then the sample pump is on and the flow rate is below 2 cfm. This condition would render the stack particulate channel inoperable.

3.

If the green power light is on and red flow light is off, then the sample pump is operable and has a flow rate in excess of 2 cfm. This is the normal operational condition.

Data for the radiation monitoring system is given on the following pages. The time indicated is EST rather that EDT. The recorder data indicates that stack particulate filter activity began to decay away starting at approximately 9:30 AM (8:30 AM EST) on 10 Apr 2009. This response would occur due to low flow or sample pump failure. The low flow light was found off on 13 Apr 2009. The stack particulate radiation detector has a failure alarm if there are no counts recorded for a period of approximately one minute. No detector failure alarm occurred.

The Constant Air Monitor (CAM) and stack particulate monitor typically have similar count rates. As indicated in the recorder data, this was the case until 9:30 AM (8:30 AM EST) on 10 Apr 2009. The CAM continued to indicate normal count rates of a few K cpm throughout the day while the stack particulate monitor count rate dropped from the normal count rate of a few K cpm to a couple of hundred cpm.

Technical Specification (TS) 3.5b requires the stack particulate and stack gas radiation monitoring channels to be operable during reactor operation. With the stack sample pump off, neither of these two channels were operable. During maintenance, the Auxiliary monitor may serve as a substitute for one of the two required channels. Neither the stack gas or stack particulate channels were declared to be in maintenance. As a result, the reactor was operated in violation of TS 3.5b.

This event is a reportable event as defined in TS (1.2.24d, 6.6.2, and 6.7.1) since both channels were not operable. As a result, the US Nuclear Regulatory Commission (NRC) was informed by telephone (refer to following pages) on 14 Apr 2009 and will be sent a written report by 24 Apr 2009.

Other radiation monitors were in service to monitor airborne radioactive effluent (Auxiliary monitor) and reactor bay airborne activity (Filter monitor and CAM). All of those monitors indicated typical levels during reactor operation as shown on the following pages. A description of the air monitors is provided on the following pages.

The Auxiliary monitor, Filter monitor, and Stack Gas monitor are similar. The Auxiliary monitor is located in the reactor building exhaust ductwork and therefore serves as an effluent monitor. The Filter monitor is located in the recirculation air duct and serves to monitor the reactor building bay. The alarm set points for the Auxiliary and Filter monitors are conservatively set when compared to the Stack Gas monitor. Initiation of the reactor building confinement system and evacuation horns would occur if the Auxiliary monitor set point was exceeded.

Given the alarm set points, the CAM visible and audible alarm would initiate before the stack particulate channel alarm. The CAM and stack particulate would have similar airborne activity readings with the reactor in the normal ventilation mode. If the reactor is in the confinement ventilation mode, the CAM

reading would be higher than the stack particulate reading. However, the CAM does not respond to airborne releases occurring in reactor beam tubes or the pneumatic transfer system. The reactor was in the normal ventilation mode on 10 Apr 2009.

Based on the readings for the Auxiliary monitor, Filter monitor, and CAM on 10 Apr 2009, it is concluded that the airborne activity levels were at normal levels and inconsequential.

This event occurred due to inattention to detail, human factor errors, design errors, and an unusual circumstance regarding completion of the startup checklist. There is no apparent reasonable cause for the stack sample pump power failure. On 10 Apr 2009, the Reactor Operators did not notice that the green power light was off. A better design would have the stack pump initiate an obvious visible signal and audible alarm upon failure alerting the Reactor Operator to a change in operability. Additionally, the Reactor Operators misinterpreted the low stack particulate channel count rate as being caused by a filter change and local ventilation of an experiment.

Following discovery of this event on 13 Apr 2009, Reactor Operators were verbally informed

1.

About this event, and

2.

To check the status of the stack sample pump while operating by observing the green power light and to shut down the reactor if the green power light is off On 14 Apr 2009, the details of this event were compiled and reported to the US NRC at 4 PM by telephone. The reactor project manager at the US NRC was given a telephone message about this event on 14 Apr 2009 and received a briefing by telephone on 15 Apr 2009.

On 15 Apr 2009, a minor change to operating procedures was made to have reactor operators log the status of the stack sample pump every 30 minutes during reactor operation. This minor change remained in place until installation of a flow switch that provides a visual and audible alarm upon loss of power to the stack sample pump and loss of sample flow was complete on 21 Apr 2009.

On 21 Apr 2009, a flow switch was installed at the inlet to the stack sample pump. This flow switch initiates an annunciator on the Control Room console if power is lost or if the flow rate drops below 0.5 cfm. The flow switch was successfully tested separately for loss of power and low flow after being installed.

On 23 Apr 2009, equipment to display the stack sample flow rate on the radiation monitoring system recorder in the Control Room was installed and tested.

The following actions have been taken to prevent recurrence:

1.

Equipment has been installed to provide a visual and audible alarm upon loss of power to the stack sample pump and loss of sample flow. Procedures have been revised as necessary for operation, maintenance and testing, and response to alarms for the newly installed equipment.

2.

Equipment has been installed to provide a display of the stack sample flow rate on the Radiation Monitor System Recorder in the Control Room. Procedures have been revised as necessary for operation and maintenance of the newly installed equipment.

3.

There is no apparent reasonable cause for the stack sample pump power failure. With completion of item 1 above, a power failure to this equipment is now readily identified.

The following actions are being taken to prevent recurrence:

4.

Review if other required equipment has a similar failure problem and initiate necessary changes.

The review will be completed and documented by 29 May 2009.

5.

Review this event with reactor operators and discuss normal readings, abnormal readings, and equipment failure for the radiation monitoring system in a reactor operator re-qualification training session. This training item will be completed and documented by 13 May 2009.

6.

Consideration may be given to revising TS 3.5 to allow temporary use of other radiation monitors, if operable with appropriate alarms and set points, in place of the required channels.

This item may be initiated at a later date.

Documentation of this event and associated design and procedure changes, equipment review for similar failure, and training will be maintained as required by Technical Specification 6.8.

Radiation Monitor System Recorder Data for 4/10/09 is shown below and on the following pages. A sudden decrease in the stack particulate monitor count rate occurred at 9:30 AM (8:30 AM EST).

6000 - Alh 5500 +

5000 +

4500 +

4000 t 3500 4 3u00 2500 4 2000 +

1500 +

V, 1000 +

500 i

i

- - +

I I

I I

07:00:59 04/10/09 08:00 08:30 09:00 (hh:mm) 09:30 10:00 10:59:59 04/10/09 Maxmum Minimum Averagel

-6 PART 6027 508 3423 6 PART 6 PART The Stack Particulate, CAM, Auxiliary, and Filter monitor readings are shown on the next page. The CAM, Auxiliary, and Filter monitor readings were normal.

The Stack Particulate monitor readout for 4/8/09 and 4/1/09 through 4/13/09 are shown on the following pages. The readings on 4/1/09 and 4/8/09 show the normal sudden drop in the count rate followed by a sudden increase in the count rate following a filter change. Also shown is the abnormally low count rate starting on 4/10/09 through 9:20 AM (8:20 AM EST) on 4/13/09.

6000.00 5500.00 5000.00 4500.00 4000.00 3500.00 3000.00 2500.00 2000.00 1500.00 1000.00 500.00 0.00

. A I

I I

07:00:59 04110/09 08:00 08:30 09:00 (hh:mm) 09:30 10:00 10:59:59 04/10/09 Maximum Minimum Average 6 PART 6027 508 3423 6 PART 6 PART 7 AUXGMM 228.00 101.90 133.10 7 AUX GM 7 AUX GM

-8 FILT GM 315.10 190.10 224.35 SCFILT GM 12 CAM 5938.0 5141.0 5564.5 6000.00 5500.00 5000.00 4500.00 4000.00 E3500.00 3000.00 2500.00 2000.00 1500.00 N

1000.00 500.00 0,00 04/10/09 00:00:59 10:04 10:06 10:08 10:10 10:12 10:14 (dd:hh) 10:16 10:18 10:20 04/10/09 23: 59:59 Maxamum Minimum Average 6 PART 6196 154 2233 6 PART 6 PART 7 AUX GM 358.20 98.14 166.07 7 AUX GM 7 AUXGIM 8 FILIT GM 394 30 187.90 262.47

- 8 F 5IT GM 12 CAM 5938.0 3265.0 4784.2

4000.0 35000

~

~i ~h2 3000 0 2500.0 E

.~2000.0 1500.0 -I 1000.0 +

500.0 +

0.0 -

04/08/09 00.00259 08204 08206 08208 08:10 08212 08214 08:16 08.18 08:20 (dd-hh) 6 PART Ma:amum Minimum Average 6 PART 6PART 4265.0 39.5 2903.7 04/08/09 23:59:59 6000.00 5500.00 500000 4500.00 4000.00 3500 00 3000.00 2500.00 2000.00 1500.00 1000.00 500.00 0.00 fl t

e In (4

ii~ 6

/!1 04/01/09 00:00:59 04104/09 04/05109 04/06/09 04/07/09 04108/09 04/09/09 04110/09 04/11/09 (mmldd/yy)

Maamum Minimum Average

-6 PART 6196 154 2233 6 PART 188.30 38.88 87.24 6 PART 5098.0 38.2 3234.3 04/13/09 08:19:59

The air monitors at the reactor facility are listed and described below:

Channel Type Name Units Radiation Detector Detector 5

Process Stack Gas cpm Beta, Gamma GM 6

Process Stack Particulate cpm Beta, Gamma Plastic Scintillator 7

Process Auxiliary cpm Beta, Gamma GM 8

Process Filter cpm Beta, Gamma GM 12 Process Constant Air Monitor cpm Beta, Gamma GM Set points are based on TS limits for Channels 5, 6, 7, and 8 and on 10 CFR 20 limits for occupational workers for Channel 12. Air monitor set points in gross cpm are listed below:

Monitor Warn Alarm Units Bases Stack Gas 550 2500 cpm TS Stack Particulate 20,000 70,000 cpm TS for alert, Sr-90 EAL for alarm Auxiliary Monitor 4000 15,000 cpm TS for alert, Xe-133 EAL for alarm Filter Monitor 4000 15,000 cpm TS for alert, Xe-133 EAL for alarm CAM N/A 12,000 cpm

< 0.3 DAC for Co-60 or 1-131 DAC is Derived Air Concentration as defined in 10 CFR 20 EAL is Emergency Action Level as defined in the facility Emergency Plan GM is Geiger-Mueller tube

Ventilation System:

rron outside to MER Pump CAM in Reactor Bay 200 cf" 190 cfm if in use 1-40D cfm Recu Bo n MCR Auxiliary Monitor Ex*u$st ron 270 cfm Ventilation Room Return Air The reactor building is exhausted to the inner stack. The outer stack ventilates another part of the building complex and may be used as a source of clean, dilution air. The stack gas and stack particulate monitors receive a sample collected from the reactor inner stack from isokinetic nozzles, associated piping, and the stack sample pump. The sample flows through the fixed particulate sampling unit (FPU) monitored by the stack particulate radiation detector and then flows into a gas chamber monitored by the stack gas radiation detector. The sample is then returned to the reactor building exhaust duct.

The Auxiliary monitor is located in the reactor building exhaust duct. The reactor building air is monitored by the Filter monitor and Constant Air Monitor (CAM). The Filter monitor is located in the reactor building recirculation duct. The CAM is located in the reactor building bay.

The radiation detectors for Channels 5, 6, and 7 provide annunciation in the Control Room and initiate the Confinement Ventilation System and Evacuation Horns if the set point is exceeded or upon loss of power to the radiation detector. The radiation detector for Channel 8 provides annunciation in the Control Room if the setpoint is exceeded or upon loss of power to the radiation detector. The CAM (Channel 12) provides a local visible (beacon) and audible (bell) alarm if the setpoint is exceeded.

The stack sample pump power is currently indicated by a lighted switch (green light) located in the Control Room during normal operation. The light goes off is there is no power. If powered, the stack sample pump initiates a flow fault (red) light if the sample flow rate drops below 2 cfm. 2 cfm is used in the basis for the stack particulate monitor set point. The operability of the flow fault light may be tested in the Control Room by pressing and holding it.

Stack Monitoring and Sample Pump with New Flow Switch:

Lines Sample Line Switch Key:

CMP-DiG-4.0 is the stack sample pump IV is Isolation Valve FCV is Flow Control Valve FPU is Fixed Particulate Unit GSU is Grab Sampling Unit F] is Flow Indicator (Rotometer or Digital Display e.g.)

FS 1 is Flow Switch I - initiates annunciator at 0.5 cfin or loss of power FS2 is Flow Switch 2 -initiates red low flow light at 2 cfm

Report made to US NRC on 14 Apr 2009:

Research Reactor Event Number: 44991 Facility: NORTH CAROLINA STATE UNIVERSITY Notification Date: 04/14/2009 RX Type: 1000 KW PULSTAR POOL TYPE Notification Time: 16:00 [ET]

Comments:

Event Date: 04/10/2009 Region: 2 Event Time: 09:30 [EDT]

City: RALEIGH State: NC Last Update Date: 04/14/2009 County: WAKE License #: R-120 Agreement: Y Docket: 05000297 NRC Notified By: GERALD WICKS HQ OPS Officer: JOHN KNOKE Emergency Class: NON EMERGENCY Person (Organization):

10 CFR Section:

MALCOLM WIDMANN (R2)

NON-POWER REACTOR EVENT DUANE HARDESTY (NRR)

Event Text VIOLATION OF LIMITING CONDITIONS IN TECHNICAL SPECIFICATION 3.5B "On Monday, 13 Apr 2009 at approximately 9 AM, the stack sample pump was observed to be off. On Friday, 10 Apr 2009, the reactor operated from approximately 11 AM to 4 PM.

"Upon investigating this situation it was learned that at approximately 9:30 AM on 10 Apr 2009 the stack sample pump apparently lost power. The stack pump remained off until being re-started on 13 Apr 2009.

"Technical Specification (TS) 3.5b requires the stack particulate and stack gas radiation monitoring channels to be operable during reactor operation. With the stack sample pump off, neither of these two channels were operable during reactor operation on 10 Apr 2009.

The Auxiliary GM may serve as a substitute for one of the two required channels. As a result, the reactor was operated on 10 Apr 2009 in violation of TS 3.5b.

"This event is a reportable event as defined in TS (1.2.24d and 6.6.2) under TS 6.7.1 since both channels were not operable as required by the Limiting Conditions for Operation TS 3.5b. The US Nuclear Regulatory Commission (NRC) was notified by telephone on 14 Apr 2009 at approximately 4 PM. The event will be investigated further with a detailed explanation of the event and actions taken and planned to prevent recurrence. The NRC will be given a written report on or before 24 Apr 2009.

"Consequences of this event were not significant. Other radiation monitors were in service to monitor airborne radioactive effluent and reactor bay airborne activity. All of those monitors indicated typical radiation levels during reactor operation."