Sunday, May 25, 2014

The Truth About Nuclear Power - Part 16

Subtitle: Near Misses on Meltdowns Occur Every 3 Weeks

This is the second of approximately one dozen articles on nuclear safety, these will (or do) include (1) the relationship between plant operators and the regulatory commission, NRC, and show that safety regulations are routinely relaxed to allow the plants to continue operating without spending the funds to bring them into compliance.  (2) Also, the many, many near-misses each year in nuclear power plants will be discussed.   (3) The safety issues with short term, and long-term, storage of spent fuel will be a topic. (4)  Safety aspects of spent fuel reprocessing will be discussed.  (5) The health effects on people and other living things will be discussed.  The three major nuclear disasters (to date) will be discussed, (6)  Chernobyl, (7) Three Mile Island, and (8)  Fukushima.   (9) The near-disaster at San Onofre will be discussed, and (10) the looming disaster at St. Lucie.  (11)  The inherent unsafe characteristics of nuclear power plants required government shielding from liability, or subsidy, for the costs of a nuclear accident via the Price-Anderson Act.  (12) Finally, the serious public impacts of evacuation and relocation after a major incident, or "extraordinary nuclear occurrence" in the language used by the Price-Anderson Act, will be the topic of an article.   Previous articles showing that nuclear power is not economic are linked at the end of this article. 


[UPDATE - 6/9/2014: The sordid saga of Rancho Seco nuclear plant near Sacramento, California -- see below]

[UPDATE 2 - 5/10/2015:  The results for 2014 are now available; the conclusion remains the same at one near-miss every 3 weeks, on average.  There were nine additional incidents in 2014.]

[UPDATE 3 -3/20/2016:  The results for 2015 are now available; the conclusion remains the same at one near-miss every 3.5 weeks, on average.  There were ten additional incidents in 2015.]

In the four year period 2010-2013, inclusive, the US nuclear reactors had 70 near-misses.  These occurred in 48 of the 103 reactors.  Some, therefore, had multiple near-misses in the same year.  One plant, Columbia, had 3 near-misses in the same year.  Wolf Creek, and Ft. Calhoun each had one near-miss in three of the four years.  On average, that is 17 near-misses per year, or roughly 17 percent of the reactor fleet.  Put another way, every 3 weeks, another near-miss occurs.  The frequency of near-misses is expected to increase over time, as the aging reactors have more equipment degrade and fail, and new systems are installed that are unfamiliar to the operators.

What is common in these incidents are old and degraded equipment that fails due to improper inspection, replacement equipment that either does not work as expected, or operators are improperly trained, and in one notable case, improperly trained workers left critical bolts improperly tightened on the reactor head.   

The most serious incident, in my view, occurred at the Byron Station, Unit 2, in January, 2012, in Illinois.  A complete loss of cooling water at Unit 2 was temporarily replaced with water from Unit 1. Had this been a single-reactor plant, with no operating reactor close at hand, the loss of cooling could have resulted in a partial or full core meltdown, exactly what happened at Fukushima, Japan.  This is completely unacceptable. 

Some, the nuclear proponents, will argue that the safety systems are adequate since no meltdowns occurred.  However, the sheer number of serious incidents shows that eventually, another catastrophe will occur.  The US has been lucky, but that luck is likely running out as the plants grow older and more mishaps occur.  

Information in these incidents are taken from Union of Concerned Scientists' series of annual reports, 2010 - 2013, inclusive.  The commentary is my own.  Links to the four (now five six) reports are:

2010 see link
2011 see link
2012  see link
2013  see link
2014  see link    (link added 5/10/2015)
2015  see link    (link added 3/20/2016)

Incidents in 2013 (Fourteen incidents)

Arkansas Nuclear One, Units 1 and 2, March, 2013

During a routine shutdown for fuel, repairs, and maintenance, a temporary crane moving the generator stator on Unit 1 collapsed. The heavy load fell through the turbine building floor, killing one worker and injuring eight others. The dropped load also caused Unit 1 to be disconnected from the offsite power grid and caused the Unit 2 reactor to automatically shut down from full power.  This incident was categorized as an AIT, or augmented inspection team response.  An AIT is for incidents that have a 100-fold increase in risk of damage to the reactor core.  

Brown’s Ferry Nuclear Plant Units 1, 2, and 3, May, 2013

SIT: Security problems prompted the NRC to conduct a special inspection. Details of the problems, their causes, and their fixes are not publicly available.  The little public information available indicates there may have been tampering on a fuel oil line to an emergency diesel generator.  An SIT is a special inspection team, for incidents that have a 10-fold increase in risk of reactor core damage. 

Columbia Generating Station, First Incident, February, 2013

SIT: Security problems prompted the NRC to conduct a special inspection. Details of the problems, their causes, and their fixes are not publicly available.   There were no violations nor significant findings. 

Columbia Generating Station, Second Incident, September, 2013

SIT: Security problems prompted the NRC to conduct a special inspection. Details of the problems, their causes, and their fixes are not publicly available. There were no violations nor significant findings. 

Columbia Generating Station, Third Incident, May, 2013

SIT: Workers discovered that the air conditioning unit for a room containing essential electrical equipment had become degraded to the point it might not have been able to prevent the equipment from overheating and failing.  Five violations were issued.  This problem had existed for years.

Sowell commentary: failure to identify such an issue over several years is a gross breach of duty.  Nuclear power plants have safety systems and support in place that must be available if and when needed.  Overheated electrical components that fail to work cannot be tolerated in a nuclear power plant.  Routine inspections should have identified the filter roll was installed improperly.  This incident only increases the perception that nuclear safety is not taken seriously by those who operate the plants.  

Fort Calhoun, December, 2012

SIT: The methods used to install anchor bolts for the raw water pumps deviated from the approved design and would not properly support the pumps from forces during an earthquake, or a barge collision on the Missouri River at the raw water intake point.  Seven violations were issued.   This and similar problems had existed for many years. 

Sowell commentary: as with the previous incident, improperly installed equipment was not identified during many inspections over several years.  This is unacceptable in a nuclear power plant.  

LaSalle County Station, Units 1 and 2, April 2013

SIT:  A lightning strike caused an electrical disturbance that resulted in both reactors automatically shutting down.  After the reactors shut down, the operators declared some emergency systems on both units inoperable.  This was a case of designers not foreseeing a potential problem.  It was considered impossible for a lightning strike to result in shutdown of both reactors.  Investigation showed that poor workmanship during the original construction of the 138,000-volt switchyard facilitated degradation of the grounding system protecting a bank of batteries that were designed to operate switches on the high-voltage systems.  However, the faulty grounding system caused several switches to open that should have remained closed.   This triggered both reactor shutdowns and startup of the five emergency diesel-powered generators.   Operators struggled with several safety systems, including the reactor’s residual heat removal pump.   NRC found the operators had inadequate training.  

Sowell commentary:  This incident illustrates that complex nuclear power plants and their numerous safety systems can be analyzed properly for “what if” scenarios, but that analysis yields a false sense of safety when systems are improperly installed.  In this case, the battery grounding system was improperly installed, so it did not function as analysts and operators expected it to function.  In nuclear power plants, this is simply unacceptable.  The risks of core meltdown and radiation release is simply too great.  This was compounded by inadequate operator training. 

Oyster Creek, October, 2012

SIT: Hurricane Sandy caused an abnormally high water level and caused the site to lose its supply of electricity from the offsite power grid.  The water rose to within 28 inches of critical water pumps and their electric motors.  The motors will not work if water enters them.  A violation was issued but only for the plant operators giving the opposite wind direction to emergency authorities.  Had a radiation release occurred, authorities would have been evacuating people in the wrong area.   NRC is requiring plants with similar vulnerabilities to rising water to take appropriate measures, but only by the end of 2016.  

Sowell commentary: One hopes nothing happens in the interim.  For example, nuclear power plants in hurricane-prone areas include South Texas Nuclear Project only seven miles from the Gulf of Mexico at the mouth of the Colorado River, several plants in Louisiana, several more in Florida, and even more in South Carolina, North Carolina, and further north along the East Coast.  Along the Gulf Coast, the land is subsiding each year.  The safety margin in having critical pumps and their motors on elevated foundations is decreasing as the land subsides. 

Shearon Harris, May, 2013

SIT: Workers reviewing results from inspections conducted in spring 2012 of metal tubes passing through the head of the reactor vessel identified degradation, a cracked weld, which should have been fixed but was not.  Almost one year later, workers again identified the cracked tube, then the reactor was shut down for three weeks for safety reasons while the tube was repaired.  The plant had been operating for nearly one year because plant operators mis-identified the crack in the tube weld. 

Susquehanna Unit 2, December, 2012

SIT: A reactor feedwater valve stuck in the closed position.  Operators opened the circuit breaker to the valve’s control motor so they could manually open the valve.  The open circuit breaker caused the feedwater control system, a new digital system that replaced the aging analog system, to close all the reactor feedwater valves.  The reactor quickly began losing water level, potentially causing a loss-of-cooling-accident.  The reactor control system automatically shut down the reactor.  NRC issued three violations, including failure to have a procedure for a stuck valve, and failure to train the operators on using the new digital feedwater control system. 

Sowell commentary: the lack of a proper procedure to deal with a sticking valve is troubling.  Valves are known to stick, sometimes closed, sometimes open, and sometimes in between.  In nuclear power plants, there is no room for “winging it.”  The failure to properly train operators on a new control system is inexcusable.  

Incidents During 2012 (18 incidents)

Brunswick Unit 2, November 2011

SIT: Workers failed to tighten properly several bolts that hold the reactor head in place.  At 100 percent rate, radioactive water poured out of the reactor head joint at more than 10 times the allowable rate.  Workers shut down the reactor.  The bolts were found to be loose enough to be turned by hand.  Investigation showed that workers assigned to tighten the bolts during the previous shutdown had mis-read the torque wrench, applying only one-tenth the specified torque.  The same workers had mis-read the bolt tension-elongation gauge.  Faulty training was the reason.   The plant’s owner had discontinued training on this procedure in 2004.  No citation was issued by NRC. 

Sowell commentary: this is elementary, the tightening of bolts to the proper torque.  That the NRC did not issue a citation is highly questionable.  One wonders how many other bolts in the plant are also not properly tightened.  

Byron Station, Unit 2, January, 2012

SIT: Unit 2 reactor automatically shut down from full power because of an electrical fault in the plant’s switchyard.  An undetected design deficiency prevented the electrical protection system from isolating the fault as intended. Consequently, the fault propagated to cause all the emergency equipment for the unit to be de-energized. The operators took steps to isolate the fault eight minutes later and to restore power to vital equipment from the emergency diesel generators.  No sanctions or violations issued by NRC. 

Sowell commentary:  This points up the problem of undetected design deficiencies.  The Unit 2 reactor was started up in 1987 – this problem had lain dormant for 24 years.  The electrical issues caused Unit 2 to have zero cooling water available – no power to motors for pumps, therefore no flow.  Operators avoided a meltdown only because the second reactor, Unit 1, was operating normally.  Cooling water from that reactor was routed to Unit 2 while the electrical issues were being sorted out.  The report states that full, normal operation was restored 10 hours after the initial problem.  This is completely unacceptable for a nuclear power plant.  Had this been a single-reactor plant, the loss of cooling could have resulted in a partial or full core meltdown, exactly what happened at Fukushima, Japan.  (so much for redundant safety systems). 

A second point: per the UCS report, “Shortly after 10:00 am on January 30, 2012, a portion of the “C” phase power line for the Unit 2 station auxiliary transformer (SAT) in the 345,000-volt switchyard broke and fell to the ground causing an electrical fault.”  In English, this means that a high-voltage power line (the “C” phase, where alternating power is conducted in three phases: A, B, and C), somehow fell apart, broke, and fell to the ground.   The question to be asked is, Why?  Was the line corroded, or did something heavy fall on it, or did some other problem cause the line to break?  Are inspections performed on such lines, and if not, why not?  This line, like the rest of the plant, was presumably only 24 years old.   If the nuclear power plants have issues such as this, with power lines breaking and falling to the ground without warning, this is a serious problem.  

Catawba, Unit 1, April 2012

SIT:  Three unrelated electrical problems caused the Unit 1 reactor to shut down automatically from full power, both reactor units to be disconnected from the offsite power grid, and an emergency diesel generator to fail.   Age-related degradation of the insulation for a power cable to one (of four) reactor coolant pumps on Unit 1 caused an electrical fault that stopped the pump.  That resulted in the automatic shutdown of Unit 1.  Subsequent chain-reaction of electrical issues caused the entire plant to be isolated from the grid, although isolation from the grid was not supposed to happen, if the design was proper.   Fortunately, four emergency diesel-powered generators started as designed to supply emergency power to the plant.   The cause of the isolation from the grid was improperly specified replacement relays for that part of the electrical system.  The original system had proper design and would not have isolated the plant.   NRC issued two violations. 

Sowell commentary: again, we see old and degraded equipment (in this case, wiring insulation) caused the initial problem.  The question is, who does the inspections of the wiring and insulation?  Was the problem inaccessibility?  Or, did nobody think to check the wiring?   Secondly, as NRC noted, it is improper to specify different parts for equipment replacement.   Parts must be like-for-like, else they can (and in this case, did) compromise safety.  

Farley, Units 1 and 2

SIT: a security-related incident for which no public details are available.  

Fort Calhoun, June, 2011 (first incident)

SIT: A replacement electrical breaker overheated, caught fire, and a series of cascading events stopped one of the spent fuel pool cooling pumps.  A second spent fuel cooling pump also lost power as the operators tried to remedy the electrical system.  The spent fuel pool was without circulating cooling water for 90 minutes before a pump could be restarted.  The water in the pool increased 3 degrees F in temperature.  Workers had noticed and reported an acrid smell (electrical smoke) three days earlier, but did not follow up to measure temperatures on the electrical equipment.   From the report:

“The failed electrical breaker was among 12 breakers replaced in November 2009. The replacement breakers were of a different size and material than the original breakers. These design differences created the potential for the breakers to experience higher temperatures during operation due to increased resistance to electrical current flow. The higher temperature exacerbated the situation by increasing the oxidation rate of internal parts, adding even more resistance to current flow. 

During installation, the replacement breakers did not align properly in the breaker compartments so workers made unapproved on-the-spot changes to make them fit. Following installation, workers used a hand-held mirror to visually determine if the pieces seemed to fit together properly. They did not measure the incoming and outgoing electricity to confirm consistency with characteristics of the original breakers.”

The NRC issued three violations. 

Sowell commentary: Again, replacement parts are at the root of an incident at a nuclear power plant.  In this case it affected the spent fuel cooling pool.  The replacement breakers were not like-for-like compared to the approved, original design breakers.   In addition, workers wasted valuable time trying to close a breaker from the control room because they did not know that that breaker required on-site, local manual re-setting.   It is not clear from the report if the breaker that would not close automatically was one that was recently replaced, or was original equipment.  Either way, the operators must be properly trained on all equipment in a nuclear power plant.  As the plants continue to age, this type of problem is more and more likely. 

Fort Calhoun, (second incident)

SIT: a security-related incident for which no public details are available.  

Harris, April, 2012

SIT:  During a shutdown, one of the three main steam isolation valves took 37 minutes to close, and a second took 4 hours and 7 minutes to close during testing. All three valves are supposed to close within 5 seconds during an accident to limit how much radioactivity is released to the atmosphere. Workers disassembled the valves to learn why they did not close properly.  The found that corrosion caused internal components to swell, effectively preventing the spring inside from closing the valve.  The valves had been in service since the plant was built, more than 25 years.  Since 2000, the plant had not tested these valves at all.    The main steam isolation valves serve to block the steam line from the steam generator to the steam turbine in the event of an incident that could release radioactive steam into the atmosphere.  These are critical valves and are designed to close within 5 seconds.  

Sowell commentary: once again, old and aging plant equipment is not performing as designed.  These valves were known to be critical, but were not tested at all for 12 years.  They would not have functioned properly in an emergency.  This is completely unacceptable in nuclear power plants.  

Palisades, April 2012

SIT: Workers shut down the reactor because of a leak of about 18 gallons per hour of cooling water determined to be through the reactor coolant pressure boundary. The plant’s operating license does not permit the reactor to operate for more than six hours with such leakage; however, the reactor operated for nearly a month under those conditions.  The leak was via a crack in one of the control rod drive mechanisms.  This type of leakage requires immediate reactor shut down, however, the reactor had been recently started up and had run for one month with this condition.  

Palo Verde, Units 1, 2, and 3 (no date and no data available)

SIT: a security-related incident for which no public details are available.  

Perry, January 2012

SIT: The plant owner reported failures to prevent unauthorized individuals from entering secure areas. Details are not available.  However, it is significant that this event occurred more than a decade after the 9/11 attacks.  NRC issued one violation. 

Sowell commentary:  how can a nuclear power plant have deficient security measures at this point, fully a decade or more after 9/11?  

River Bend, May, 2012

AIT: A trip of the reactor due to loss of a single feedwater pump—cascaded into loss of the normal heat sink and loss of cooling water to emergency and non-emergency equipment, because of problems in the in-plant electrical distribution system.   A 13.8 kV power cable failed, and started a fire.  The loss of electricity due to the failed power cable stopped two of the four cooling water circulating pumps, leading to reactor shutdown.  The workers put out the fire, then closed a switch to connect the functioning power to all four pumps.  The reactor was started back up, but a second electric power cable failed.  This led to the loss of all pump power, and shutting down the reactor again.  Workers started up emergency cooling systems and eventually cooled the water in the reactor to below 212 degrees, which is considered a safe temperature.   NRC issued 8 violations due to this event.  Essentially, the workers violated several industry practices, they cut corners. 

Sowell commentary: Again, electrical cables failed that caused a reactor shutdown.  Electrical switches also failed and had not been tested per manufacturer’s requirements.   Electrical cables had not been properly tested, either.  

San Onofre Units 2 and 3, January, 2012

AIT: Unexpected degradation was identified for the tubes within the recently replaced steam generators on Units 2 and 3, causing a release of radioactive steam to the atmosphere.   The new steam generators failed much sooner than expected, and inspection showed a new form of tube erosion.  NRC required the plant owner to identify the cause of the eroded tubes before startup would be allowed.  The plant owner instead decided to shut down both reactors permanently. 

Sowell commentary:  the replacement steam generators were not like-for-like compared to the original equipment, and the tubes failed much more quickly.  The records indicate the new steam generators had more tubes but with a smaller inside diameter.   The experts concluded the tubes banged against each other during operation, from excess vibration.  The banging tubes led to erosion and leaks.   As more and more reactors reach the stage where steam generator replacement is required, we can expect more of this.  In the San Onofre case, the owner chose to shut down permanently rather than spend the time to identify the exact cause as NRC required.    A cascade of events followed the decision to shut down.  More gas-fired power plants are now operated, and some customers’ bills have increased.  The local public utility commission is trying to decide how much of the costs related to replacement power to allocate to customers, and how much to the utility.   Also, new power plants are now approved for construction to replace the generating capacity, with at least 75 MW of that being grid-scale storage.  

Wolf Creek, September 2011 (first incident)

SIT:  One emergency diesel generator experienced load swings of up to 500 kilowatts when loaded to 5,800 kilowatts during a test run on September 1, 2011. The cause was determined to be an improper adjustment to the control system in May 2011 during an attempt to resolve another problem.  NRC issued two violations for improper testing and improper procedures. 

Wolf Creek, January, 2012 (second incident)

AIT:  Two separate, unrelated electrical faults resulted in loss of the plant’s normal sources of electricity. While both emergency diesel generators automatically started and supplied power to essential equipment, other equipment problems complicated the operators’ response to the event.  NRC issued no violations from this event.

Incidents in 2011 and 2010 had similar issues. 

[UPDATE-6/9/2014: The short-lived Rancho Seco nuclear plant near Sacramento, California.  The plant was shut down permanently after only 18 years of operation (1971 - 1989) due to an incredible number of leaks, radiation emissions, fires, mechanical breakdowns, the list is very long.  see link    -  end update]

Previous articles in the Truth About Nuclear Power series are found at the following links.  Additional articles will be linked as they are published. 

Roger E. Sowell, Esq. 
Marina del Rey, California


Christina Macpherson said...

Another very valuable and informative article. THANK YOU!

Elstel said...

I enjoyed reading this blog post. It was inspiring and informative. nuclear disaster