Area 1

WHAT IF?
I never quite understood the value or purpose of this statement when describing an Industrial Accident (or maybe I do).


March 28, 1979 (Day 1) is what it is.

What's to "what if"; maybe the guy who shut the block valve croaked on his way to work? Someone else would have found it; they had lots of help and were working hard.


March 29, 1979 (Day 2) is what it is.

What's to "what if", it is what it is. Or maybe "what if the dog hadn't stopped to $hit, he could have caught the rabbit?"


March 30, 1979 (Day 3) is what it is.

"What if" that bogus Cronkite news story had never hit the airwaves? Would nuclear power expansion still be alive and well in the US today? IMO yes.
The three NRC HQ blunders that caused it:

I didn't invent or make any of this up. It is documented in the Rogovin Report. If anyone's understanding of the total TMI2 Industrial Accident is different, they have been misinformed. The real question to ask is "How and why did the mis-information happen?"


Do you think it possibly could be that the “powers” said we can’t say out loud the NRC is a totally dysfunctional organization? What might the people demand we do in that case? The best way out of this mess is to not say it out loud, focus on something else. And let the NRC corrections proceed over time.

Another term I don't understand that is used to describe the TMI2 Industrial Accident is "near miss." Is that something like an instant death from a head on collision in a car at 130MPH relative velocity on a 2-lane highway? Gee, I "near miss" several thousand times a month. So what’s the real point?

 

TMI2 Day Three (March 30, 1979) Plant Status Summary
The plant had regained RCS forced flow the day before (an RCP was running) and in the process of cramming as much water back into the system (re-covering the core) as possible they realized they had a non-condensable gas bubble in the system. They understood the gas bubble would have to be removed before they could get the plant to “cold shutdown” conditions. They were in communication with outside technical help including B&W, and had a process in place that was de-gassing the system. Basically they were using the “Letdown” system to divert the reactor coolant which was under higher pressure and had the entrained gas, to the lower pressure Make-up Tank where the gas would come out of solution. The Make-up Pumps were pumping that water back in to the RCS, and the gas released into the Make-up Tank would cause the Make-up Tank pressure to increase. They were monitoring the Make-up Tank pressure increase and were venting the Make-Up Tank to the Waste Gas Decay Tanks through a vent header.

During Day 2 they discovered the vent header was leaking radioactive gas to the Auxiliary Building.  The Auxiliary Building atmosphere was then being exhausted continuously to the Station Vent Stack by the Auxiliary Building Ventilation System. The method was working, the RCS was de-gassing, and TMI2 was keeping the proper PA state authorities informed of the releases. By this time TMI2 had a fairly decent understanding of the plant conditions, and understood this process would get rid of the gas bubble and allow plant cool down and depressurization to cold shutdown conditions.

Meanwhile,  

From Rogovin Report Part 1, Starting at PDF Page 75
At 9 a.m. (Friday, March 30), Reilly tells Lamison (PA Emergency Management Association) that the releases have been contained. The same cannot be said for the reports of the releases, which have developed a life of their own. And to these, another report has been added. Back in Bethesda at the Incident Response Center, Lake Barrett is fretting over an 8:45 a.m. telex from regional headquarters reporting a conversation with Inspector Jim Higgins up at the site. The
telex reads:

"The seal return to the makeup tanks was causing excessive gas pressures in the makeup tank which was directed to the waste gas decay tanks which were full. The waste gas tanks were being released to the stack. Pennsylvania Civil Defense was being notified by Licensee."

Barrett has been deeply concerned about those waste gas decay tanks: once they are full of concentrated radioactive gases any additional gas will be released directly to the vent stack. The factor to keep in mind here is that the report is wrong. The waste gas tanks are not being vented to the stack; this is not a valid concern at this time. Barrett, of course, does not know this. He does some quick calculations, based on data he received Thursday evening, and projects an alarming hypothetical release rate from the waste gas tank venting.

John Davis, a member of the NRC's Executive Management Team (EMT), is standing nearby as Barrett arrives at these projections, and he urges Barrett to take them next door to brief the EMT. The duty officers on the EMT at around 9 a.m. are Lee Gossick, Executive Director for Operations; Harold Denton, Director of Nuclear Reactor Regulation; and John Davis, Director of Inspection and Enforcement. Also in the room are Joe Fouchard, Director of Public Affairs; Harold (Doc) Collins, Assistant Director for Emergency Preparedness, and Edson Case, Denton's deputy.

Barrett is asked to estimate what the offsite dose rate might be. Based on his rough projections, Barrett estimates that a member of the public at the site boundary might be exposed to 1200 mr/hr, a figure so high that exposure for only an hour could exceed EPA evacuation guidelines for sensitive individuals.

At this precise moment, as though scripted (Maybe by Saturday Night Live?), there is a call into the EMT from Karl Abraham, calling on the speaker phone from the Governor's offices: there is a report of a 1200-mr/hr reading from "one of the cooling towers"; is it true? The EMT apparently panics; they forget that Barrett's 1200 mr/hr is an offsite ground level number and Abraham's is apparently on site, directly over the plant. They forget that Abraham is asking, not telling. The EMT scrambles, amid talk of "moving people," "biting the bullet," and "better safe than sorry." Consensus is quickly reached: evacuation.

Denton asks Barrett for more advice: How far out for evacuation? Barrett has not given the matter any thought, but suggests 10 miles. A hurried discussion results in shading the recommendation to 5 miles for now: 10 miles would include parts of Harrisburg, and all manner of extra complications.

Sensing the urgency of the situation, Collins asks the administrators what they wish him to do, and is told by Denton to "call the State of Pennsylvania and recommend the implementation of this precautionary evacuation."

It is 9:15 when Doc Collins reaches Henderson at PEMA with the EMT recommendation. He is reassured when Henderson tells him PEMA has already heard of the 1200 mr/hr reading. (It is logical that their reports are self-corroborative: the information has traveled full circle from Floyd (TMI2 Operations Supervisor) to PEMA to Critchlow to Abraham to Collins and back to PEMA's Henderson.) The accelerating factor, unknown to Henderson, is the erroneous Higgins report regarding the full waste gas tanks.

Henderson tells Collins PEMA is awaiting word from the plant before making any moves on evacuation advisories, and Collins responds: "We're recommending here that you go ahead and evacuate people out to 10 miles in the direction of the plume." Henderson responds: "We'll start with 5 miles."

The situation bears striking similarities to Wednesday morning's abortive evacuation alert. The difference is that the NRC's emergency managers, unlike the BRP earlier, have executed an astonishing list of thou – shalt - nots; chief among these are the failure to check precisely where the 1200 mr/hr reading was taken (not offsite, but directly over the plant stack), and what the actual offsite readings were. And once the hip-shot response is formulated, it is taken not to the Governor's office but to a State civil defense apparatus that is eager to be tested.

((The above paragraph represents an extreme example of Rogovin (intentional?) oversight. Another “thou shalt not.” NRC Executives have abandoned a completely functional Emergency Plan. The TMI2 emergency response control center, set up in U1 control room is in continuous contact with the PA BPA, the organization responsible for concurring with any TMI2 recommended PAGs. Did the NRC even have an E-Plan, or were of bunch of misinformed (by their own people) bozos, just winging it?))

From Rogovin Report Part 1, PDF Page 81
Shortly after talking with Hendrie, the President calls Governor Thornburgh, tells him Harold Denton is on his way, and concurs that there is no reason for implementing emergency measures. This is a remarkable recovery for Denton, who only this morning had given Doc Collins the order that set off the evacuation scramble. (This wording is a little too subtle for me.  Exactly what is meant by “remarkable recovery?” Can someone please explain it?)

From Rogovin Report Part 1, PDF Page 87
Despite the distractions Friday evening resulting from the meltdown story, Hendrie calls several staff members, urging them to study radiolysis. Roger Mattson, who says he never heard of the problem, suggests he will get Denwood Ross and Merrill Taylor, two NRC staffers now at the site, to work on it. Hendrie urges Mattson to reach out for a "different set of guys," because the men at the site are tired, and he does not want any slip-ups.

From Rogovin Report Part 1, starting at PDF Page 99
Back in Washington as the Presidential visit to TMI ends, the NRC commissioners, minus Chairman Hendrie who has gone to the site, are still looking at the bubble situation in light of the story in the Sunday papers and what they have learned of Chairman Hendrie's worries. They have been told that, using a worst-case analysis, the bubble could be near a flammable stage, and could burn if ignited. They ask what this might do to the reactor vessel. On that subject, Robert Budnitz, NRC Deputy Director of Research, explains that the burn process might produce a pressure pulse that could damage the vessel:

". . . we might lose that vessel, which we can't afford. Although, by the way, losing it at the top is going to be like a LOCA; it's not like losing it at the bottom, but it's still bad. There is going to be a propagated pulse everywhere in the system. We're going to lose pumps. We just can't stand that."

Budnitz is one of several NRC experts who have been approached by Roger Mattson on Saturday on the chairman's behalf. He had been assigned the task of producing, by Sunday noon, a good answer to what might happen to the reactor vessel in the event of a hydrogen bubble ignition. His first calculations indicated that the vessel might survive with little damage, but "by Sunday noon," recalls Budnitz, "what it added up to was that we'd probably lose it."

The commissioners are troubled by this. On the previous night, the NRC staff prepared for them an emergency decision document that shows that serious offsite consequences can ensue only hours after a serious turn for the worse in the plant. And now, adding to the tensions, Harold Collins tells them that the State authorities are down to "White Alert"; they have told Collins they feel it is important to have at least 4 hours advance notice for a major evacuation. The State police and National Guard are on a 4-hour alert status. PEMA would like to shorten that state of readiness, but the Governor refuses, not wanting to encourage panic.

Thornburgh's resistance to pressure for evacuation is consistent throughout the course of the accident. His recognition of the hazards imposed by stress alone seems to run counter to much of the advice he is getting - by no means all of this from the NRC commissioners. The Governor appears to be listening most closely to his BRP staff, and they seem to be giving him solid evidence that no evacuation is yet warranted, no matter what anyone else says.

What he most wishes to avoid, Thornburgh will say later, is "a show of helmets"- the appearance in the streets and highways of uniformed National Guardsmen whose very presence would bespeak emergency and arouse unwarranted fear, whatever might be accomplished in getting an early start on an evacuation. It turns out to be a winning gamble, and a fairly audacious one for a public official in office only a short time.

After hearing staff reports that an explosion might result in a rupture of the reactor vessel, the commissioners in Washington decide to advise Hendrie at the site that unless he has something different, the commission should recommend a precautionary evacuation to Governor Thornburgh. But the chairman has something different in mind.

At the site Stello has done some more checking with representatives of Westinghouse - Bettis and General Electric - San Jose; both respond that there will be no oxygen added to the bubble, and Stello passes this along to Chairman Hendrie. Suddenly, Joe Hendrie is convinced. His doubts are ended: there would be no hydrogen explosion.

Commissioner Kennedy relays to Chairman Hendrie the word that based upon Budnitz's report about losing the vessel if the hydrogen exploded, three commissioners (Kennedy, Bradford, Ahearne) recommend that the Governor advise a precautionary evacuation out to a 2-mile radius. But Hendrie has made his turn and is headed back the other way. He informs Kennedy that oxygen is not really a problem, as the hydrogen in the water would capture it.

At 7:00 p.m., Hendrie and Mattson meet with the Industry Advisory Group in Middletown, and EPRI representative Ed Zebroski takes the NRC to task for not quickly understanding that oxygen could not be evolved by radiolysis in a hydrogen-rich environment. At this point, the bubble is down to 350 cubic feet and fading fast, due to the efforts of Met Ed and its outside advisors, particularly B&W, to degas the system. At 8:45 p.m., Hendrie and Denton meet once more with the Governor to brief him on the question that has made the past 24 hours a (NRC generated) nightmare.

From Sunday night on, the word that goes out is reassuring: Stop worrying about the bubble exploding. The bubble is disappearing ... and the bubble problem right along with it.

Two days later, on Tuesday, April 3, Denton is able to announce to the world at a press conference, "The bubble has been eliminated, for all practical purposes."

Asked why the bubble had gone away, Denton replied: "I think it was a little bit because of our actions and maybe a little bit of serendipity." Although probably not intending to do so, Denton seemed to have given credit to the NRC for removing the bubble. To the contrary, as NRC Inspector Charles Gallina, who had been at the site from the beginning of the accident, observed, "The hydrogen bubble did not miraculously disappear, it was systematically and professionally eliminated by Met Ed operators." In fact, studies performed for the Special Inquiry Group show that the bubble was probably all gone some 2 days before Denton made it official.

SUMMARY (Starts on PDF Page 102)
The one theme that runs through the conclusions we have reached is that the principal deficiencies in commercial reactor safety today are not hardware problems, they are management problems. These problems cannot be solved by the addition of a few pipes and valves-or, for that matter, by a resident Federal inspector at every reactor. Undoubtedly improvements in the design, instrumentation, and control logic of nuclear plants can be made to reduce the probability of a serious accident, and to better protect the public should such an accident occur. Some detailed suggestions for such improvements are included here and in the in-depth studies to this report. But the most serious problems will be solved only by fundamental changes in the industry and the NRC.

What we have found is a regulatory system consisting primarily of an elaborate apparatus for reviewing the safety of nuclear reactor designs which has served the public well in the past and produced a good safety record to date, but in the process has failed to take timely account of the actual operation of existing plants. We have found that the Nuclear Regulatory Commission itself is not focused, organized, or managed to meet today's needs. In our opinion the Commission is incapable, in its present configuration, of managing a comprehensive national safety program for existing nuclear power plants and those scheduled to come online in the next few years adequate to ensure the public health and safety.

We have found, based upon our study of TMI and our interviews with knowledgeable people in the industry, that many nuclear plants are probably operated by management that has failed to make certain that enough properly trained operators and qualified engineers are available on site in responsible positions to diagnose and cope with a potentially serious accident. The NRC, for its part, has virtually ignored the critical areas of operator training, human factors engineering, utility management, and technical qualifications.

We have found an industry in which the expertise and responsibility for safety is fragmented among many parties - the utility company that operates the plant, the plant designer, the manufacturer of the reactor system, the contractor, and the suppliers of critical components, in addition to the NRC. Coordination among these parties and between them and the NRC, as well as within the NRC, is inadequate. As a result, there are many institutional disincentives to safety, and safety issues that are identified at some point in the system often fall through the cracks. Prior to Three Mile Island, the industry as a whole had made only feeble attempts to mount any industry wide affirmative safety program, and many utilities apparently regarded bare compliance with NRC minimum regulations as more than adequate for safety.

On top of all this, we found that before March 28, 1979, an attitude of complacency pervaded both the industry and the NRC, an attitude that the engineered design safeguards built into today's plants were more than adequate, that an accident like that at Three Mile Island would not occur-in the peculiar jargon of the industry, that such an accident was not a "credible event."

------------------------------------------------------------------------------------------------------------

Observations and lessons learned, by mjd.

The TMI2 Media Event far surpassed the magnitude of the Technical Event (zero radiological affects for anyone). The Media Event was created by NRC errors. But we utilities were made to pay the financial price for these NRC errors.

The Media Event focus was

NOTE: I find it ironic that the NRC perceived solution to unqualified boneheads meddling in a process was to force the plants to spend money to give NRC HQ faster direct access to plant data. It is cheaper and more logical to just get rid of the boneheads and stop meddling.

The subsequent NRC over reaction to the Media Event created by NRC errors focused on regulatory changes, including hardware changes and several capital intensive changes, priced nuke power out of business in the US.

These changes manifested themselves not only as extreme capital expenses utilities had to justify to individual state rate control regulatory bodies for the operating plants, but also for plants under construction, where cost and schedule over runs averaged 250 – 300% in the ‘80s.

And let’s never forget, the NRC failure to address the four TMI2 precursor warnings, which was the NRC responsibility, was the Root Cause of the TMI2 Accident in the first place.

Millions of dollars were spent by each nuke utility upgrading Emergency Response Facilities, E-Plans, staffing, data links, etc., etc. Was the cost proportional to the problem? Or did it even address the problem more so than just reacting to the Media Event? In the case of loss of confidence in Met Ed, the guy lost his job. Did the NRC guy lose his job? Or did the NRC even ever admit they blew it on an explosive hydrogen bubble, under the same magnitude of press coverage the mistake received?

The amount of money spent upgrading Emergency Response Facilities, E-Plans, staffing, data links, etc. although significant, pales in comparison to the money spent inside the plant buildings modifying hardware and adding systems to the plant to insure no plant transient could ever cause system pressure to increase and lift a PORV again. Why? Just close the block valve and run the plant that way. Don't want to do that? From the Rogovin Report discussion of the 1974 Beznau Incident here's something people in police work would call a clue:

"The operators were able to identify that the PORV was open in approximately 2 to 3 minutes and shut the isolation valve (there is no indication of what caused the operators to realize in such a short period of time that the PORV was open)."

NOTE: Other places in the Rogovin Report, in the exact same paragraph as this quote, mention this plant PORVs had a mechanical direct indicating position indication, not an implied indication like DBNPP and TMI. Well…duh! Install a PORV with an actual position indicator. END NOTE

The above quote is from the Rogovin Report discussion of the Beznau Incident (See the Precursor Events here http://www.nukeknews.com/Precursor Events.html)