Operational Reactor Safety
22.091 /22.903
Professor Andrew C. Kadak Professor of the Practice
Lecture 21
Davis Besse - N ear Miss 2002
Prof. Andrew C. Kadak, 2008 Page 1
Department of Nuclear Science & Engineering
Topics to Be Covered
• History of Davis Besse
• Review of Alloy 600 cracking
• Review of Davis Besse Vessel Head Leakage
• Contributing Factors
• Failures of Operator, NRC, INPO, Oversight
• Lessons Learned
Prof. Andrew C. Kadak, 2008 Page 2
Department of Nuclear Science & Engineering
Davis Besse - 8 73 Mwe Babcock and Wilcox Design
Department of Nuclear Science & Engineering
Prof. Andrew C. Kadak, 2008 Page 3
Source unknown. All rights reserved. This content is excluded from our Creative Commons license. For more information, see http://ocw.mit.edu/fairuse .
History of Davis Besse
• 1995 World Record of a 99.2% capacity factor
• 2001 - 99.7% capacity factor
• 2001 – 500 day run completed in October 2001
• 5.5 million hours worked without lost time accident in 2001
• Considered a good performing plant by NRC and INPO
Department of Nuclear Science & Engineering
Prof. Andrew C. Kadak, 2008 Page 4
Primary Water Stress Corrosion Crackin g o f Vesse l Hea d Penetratio ns
• First observed in France – B ugey 3 Reactor in 1991
• Associated with PWSCC of Alloy 600 (inconel)
• PWSCC function of temperature, pressure and time
• NRC sent out information notices – r equired inspections
• Industry did assessment of susceptibility of reactors (BW/CE)
– Established a scale based on full power hours of operation
– B ased on head temperature
– I ndustry did not consider this a significant issue since US reactor head were built differently than French reactors.
• Inspections difficult due to access and dose
Prof. Andrew C. Kadak, 2008 Page 5
Department of Nuclear Science & Engineering
Source unknown. All rights reserved. This content is excluded from our Creative Commons license. For more information, see http://ocw.mit.edu/fairuse .
• Perception was that if cracks occurred they would be axial not circumferential and detectable
• Carbon steel vessel degradat ion was considered but not judged to be significant due to flashing of steam and leaving boron crystals (>500F) – n ot as a liquid – 4 inches/yr if water
• Inspection of Oconee Nuclear Station 1 (Nov. 2000), Arkansas Unit 1 (Feb. 2001), Oconee Unit 3 (Feb. 2001) and Oconee Unit 3 (April 2001) showed both axial and circumferential cracks in Control Rod Drive Mechanisms.
• NRC Issues Bulletin 2001-01 ordering inspections of highly susceptible plants by December 31, 2001.
• NRC prepares a shutdown order for Davis Besse
Prof. Andrew C. Kadak, 2008 Page 6
Department of Nuclear Science & Engineering
Source unknown. All rights reserved. This content is excluded from our Creative Commons license. For more information, see http://ocw.mit.edu/fairuse .
Prof. Andrew C. Kadak, 2008
Department of Nuclear Science & Engineering Page 7
Results
• Davis Besse requests an extension to next spring outage.
• NRC grants extension February to 16, 2002.
March 2002
Department of Nuclear Science & Engineering
Prof. Andrew C. Kadak, 2008 Page 8
Typical PWR Reactor Vessel
Prof. Andrew C. Kadak, 20
ent of Nuclear Science & Engineering Page 9
08
Departm
STP Penetration # 46
Prof. Andrew C. Kadak, 2008 Page 11
Department of Nuclear Science & Engineering
Source unknown. All rights reserved. This content is excluded from our Creative Commons license. For more information, see http://ocw.mit.edu/fairuse .
Nnnle 3 with insulatinn reninved and shielding installed 03- ] 8-IJ2
Source unknown. All rights reserved. This content is excluded from our Creative Commons license. For more information, see http://ocw.mit.edu/fairuse .
Vessel Degradation
Prof. Andrew C. Kadak, 2008
Department of Nuclear Science & Engineering
Page 14
Boric Acid Deposits
Prof. Andrew C.
Department of Nuclear Science & Engineering
Page 15
Kadak, 2008
Davis-Besse Reactor Vessel
Department of Nuclear Science & En
Prof. Andrew C. Kadak, 2008 Page 16
gineering
Control Rod Drive Mechanisms
Department of Nuclear Science & Engi
Prof. Andrew C. Kadak, 2008
neering
Page 17
Vessel Head Penetration Nozzle
Department of Nuclear Science & Engineering
Prof. Andrew C. Kadak, 2008 Page 18
Davis Besse Experience with Primary Coolant Leaks
• All BW plants reported boric acid leakage problems including vessel head penetrations
• RPV head vent to steam generator (1992)
• RCS thermowells
• CRDM flange leaks
• Pressurizer spray valve
• Letdown isolation cooler isolation valve
• Pressurizer safety relief valves.
Prof. Andrew C. Kadak, 2008 Page 19
Department of Nuclear Science & Engineering
Davis Besse Indicators
• Containment Air Cooler Clogging with Boron Crystals – C leaning monthly instead of yearly
• Containment radiation monitor filters (1998 -2002)
– Ultimately required replacement every 2-3 days
– F ound brown stains with boron crystals.
• Some bolts on pressurizer spray valves corroded off due to spray valve leakage.
• Leakage increased by a factor of 10 but still within technical specification limits.
Prof. Andrew C. Kadak, 2008 Page 20
Department of Nuclear Science & Engineering
Missed Opportunities
Photo Circa 2000
Prof. Andrew C. Kadak, 2008
Department of Nuclear Science & Engineering Page 21
Source unknown. All rights reserved. This content is excluded from our Creative Commons license. For more information, see http://ocw.mit.edu/fairuse .
Breakdowns
• Utility
• Industry – NEI and EPRI
• NRC
• INPO
• Oversight Boards
Prof. Andrew C. Kadak, 2008 Page 22
Department of Nuclear Science & Engineering
Lessons Learned
• Could have set nuclear industry back (again) – m ajor non-isolable leak – break – i n reactor pressure vessel
– W e are judged by our poorest performer
• Complacency based on good record
• Poor management oversight and awareness
• You can go to jail (several charged with criminal violations – f alsification of records)
• Conservative decision making is important
• Not allowing unacceptable conditions to exist.
• Strong questioning attitude needed
Prof. Andrew C. Kadak, 2008 Page 23
Department of Nuclear Science & Engineering
More lessons
• Focus should be on causes not symptoms
• Engineering organization needs to be engaged in problem resolution not just enabling management decisions.
• Mind set of it can never happen needs to be challenged.
• Oversight organizations need to be aggressive.
– I NPO should have identified the problem
– Outside Nuclear Safety Review Boards should not only listen to management presentations
• NRC resident inspectors did not do their job
• Group think should be avoided
Prof. Andrew C. Kadak, 2008 Page 24
Department of Nuclear Science & Engineering
Even more lessons
• Failure to use experience reports and believe them
• Power production is important but if safety compromised the plant and the industry will suffer.
• Safety culture differentiates excellent performers from bad.
Prof. Andrew C. Kadak, 2008 Page 25
Department of Nuclear Science & Engineering
Consequences
• Davis Besse Replaced reactor vessel head.
• Repairs cost $ 600 million – l oss of revenue
• Plant shutdown for 2 years
– R estart issue was not of adequacy of repairs
– Restart was predicated on whether or not the safety culture of the plant was acceptable for operation!
• Fortunately this event was considered as an isolated event by the public but a failure of the regulatory and oversight process.
Prof. Andrew C. Kadak, 2008 Page 26
Department of Nuclear Science & Engineering
Homework
• Review the FENOC (Davis Besse) request for continued operation sent in late 2001 to justify operation until the spring out.
• Based on the information provided and the experience with Alloy 600, provide a technically based answer to the request – y ou may want to review the NRC letter granting approval to see if you agree – why and why not.
Prof. Andrew C. Kadak, 2008 Page 27
Department of Nuclear Science & Engineering
MIT OpenCourseWare http://ocw.mit.edu
22.091 Nuclear Reactor Safety
Spring 200 8
For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms .