Japan Earthquake: Nuclear Plants at Fukushima Daiichi

[steve olsen]

And we already know that the tsunami wrecked a big part of the dosimeters stored at the plant site. So one could assume that the same happened to other devices as well. For example portable devices to correctly measure massive radiation.
​

In a situation like this, 6 hours later, there should have been military helicopters delivering anything and everything that could have possible been needed. No excuses on that. The road should have been cleared within 24 hours, like a military operation. Those Fukushima 50 needed better support, sheesh, they didnt even have proper futons to sleep on even weeks into this. Taking care of basic human needs is square one, for any employer, and especially in an ongoing emergency.

 

[Pakman]

but we're trained to not go where it's unknown
and we're trained that to go into a very high area one takes special equipment and precautions and plans the entry carefully...

"One does not simply walk into reactor building"

That's what residents are waiting from you, NPP operator - being able to save your own life, not theirs and their children's.

By the way, are you trained to operate without both AC and DC power?

 

[clancy688]

In a situation like this, 6 hours later, there should have been military helicopters delivering anything and everything that could have possible been needed.

Not so sure about that either. Nobody in Tokio had a clue what was going on at the plant. That's what tsutsujis translations show us. Then there's a big catastrophe going on, stretching 100 miles to north and south, with every available night flight capable helicopter in whole Japan being busy rescuing people somewhere else.
Not mentioning the actual tsunami devastation at the plant itself, coupled with energy loss at three reactors. I wouldn't expect those guys to take inventory in those conditions. Whatever the tsunami took away from them, they probably didn't realize it before they needed it.
Still, that's all mostly speculation, so we should stop here.

Those Fukushima 50 needed better support, sheesh, they didnt even have proper futons to sleep on even weeks into this. Taking care of basic human needs is square one, for any employer, and especially in an ongoing emergency.

There I agree with you. I remember the case of the first worker who died with an heart attack - with no doctor and no ambulance on station at the plant. Weeks after the tsunami.

That's what residents are waiting from you, NPP operator - being able to save your own life, not theirs and their children's.

Best way for one saving other people is not getting himself killed in the first place IMHO.

 

[steve olsen]

Originally Posted by Pakman View Post

That's what residents are waiting from you, NPP operator - being able to save your own life, not theirs and their children's.

Best way for one saving other people is not getting himself killed in the first place IMHO.

Well said, one of the first requirements and training of emergency's responders is not to get yourself killed or injured.

Back to the "restock the plant" idea...everyone should have known that in the absence or knowledge, assume the worst in a case like this. I think they were embarrassed to call on the US military for assistance, regardless, they didn't. More resources could have and should have been brought to bear immediately.

 

[Most Curious]

Dead rescue folks are useless! In turn, losing them may cause far MORE innocent injuries / deaths.



I am disappointed the Japanese did not learn the serious lesson of Chernobyl concerning radiation instruments! I hope US plants have examined these two accidents and made changes to prevent a third occurance. During a crisis, I would think ONE team member would be equipped with a high range instrument that had a scale covering the maximum field they were willing to enter.


I agree, they should have sought military help sooner - from US or anyone able to provide assistance. SUch a disaster is not the time for secrecy or undue national pride.

Slightly at topic's edge, anyone ever hear if there were repercussions for the captain of the carrier that was contaminated downwind of the NPP?

 

[NUCENG]

Dead rescue folks are useless! In turn, losing them may cause far MORE innocent injuries / deaths.



I am disappointed the Japanese did not learn the serious lesson of Chernobyl concerning radiation instruments! I hope US plants have examined these two accidents and made changes to prevent a third occurance. During a crisis, I would think ONE team member would be equipped with a high range instrument that had a scale covering the maximum field they were willing to enter.


I agree, they should have sought military help sooner - from US or anyone able to provide assistance. SUch a disaster is not the time for secrecy or undue national pride.

Slightly at topic's edge, anyone ever hear if there were repercussions for the captain of the carrier that was contaminated downwind of the NPP?

Captain Thom Walter Burke is still in command of USS Ronald Reagan. He has just over a year on board and typical command tour is about 2.5 years.

 

[Pakman]

Best way for one saving other people is not getting himself killed in the first place IMHO.

Radiation does not kill instantly. There is a lot of time to make a lot of useful things in the reactor building, such as the checking of the local instruments status and the attempt to bring the IC back to operation.

So much time and so many possibilities, which the Chernobyl staff never had, were wasted by means of personel safety.

 

[NUCENG]

Radiation does not kill instantly. There is a lot of time to make a lot of useful things in the reactor building, such as the checking of the local instruments status and the attempt to bring the IC back to operation.

So much time and so many possibilities, which the Chernobyl staff never had, were wasted by means of personel safety.

Hi Pakman, welcome. Since their dosimetry was off-scale high they couldn't know how much time they had. A 1 Sv/hr dose rate for one hour will result in vomiting and rapid onset of radiation sickness. A 5 Sv/hr dose rate will reach that point in 12 minutes and can result in a lethal dose in less than 20 minutes.

What would you have wanted them to do? The Chernobyl operators knew they were receiving large doses and they died without making any difference in the outcome. In any case according to repports I've read they stuck around for only two reasons - to make sure no one was trapped, and to try to see what the could toppass on to the responders. I don't think any mitigation actions were attempted by the operators. That was heroic but futile.

The Fukushima operators followed their training. There was some degree of human error in F1-1 dealing with the IC, But F1-2 and F1-3 were pretty much doomed from the time the tsunami struck. If their training was deficient, and if their procedures were inadequate, it was too late for them, but I am impressed with the ways they tried to work around valve operations and external water makeup. And they stayed there and fought to do what they could. There have been comments that many of the Fukushima 50 expected to die. They were just as heroic as the operators at Chernobyl.

In reply to your question about training. After 9/11 the US nuclear plants added AC and DC independent water supplies and added significant procedures to deal with large aircraft impacts and fires. These equpments are routinely tested and exercised. Shortly after the accident at Fukushima NRC required all planmts to inspect, test and validate their procedures for dealing with that type of event.

 

[nikkkom]

I don't know about that. I'd figure that any device which could measure massive radiation (mSv/h to Sv/h) would be very, very unsusceptible to radiation in the uSv/h range.

This is a very easy problem. Incorporate both sensitive and high-range radiation meters into one device. Add a bit of logic which automatically displays high-range reading when sensitive one goes off-scale high.

 

[nikkkom]

lots of tough calls to make that day... those poor guys are heroes in my book...

old jim

Well, unfortunately, they aren't heroes. They could have saved the plant, but they did not.

Don't take it as I blame them. We must not rely on heroism of operators to save the bad day. The blame lies squarely on TEPCO and on wider nuclear power industry for not spending relatively modest amounts of money and effort in preparations and training for severe accidents.

How much money was saved worldwide on not considering and not training for a prolonged SBO scenario? A billion dollar or so, tops?

 

[Pakman]

There was some degree of human error in F1-1 dealing with the IC

Нi NUCENG. Nice to join your discussion.

You say "some degree"? I'd say it's the highest level of.

Imagine, you are Unit 1 Shift Superviser. You get all indicators extinguished in UCR. You also have the reactor just shut down and isolated. Last time you saw the IC status, it was switched off. Wouldn't you send man to the location to check if the reactor is cooled or not? Wouldn't it be your priority task right now? Or you will be waiting for when the electric staff turn the light on?

When I wrote about the time and possibilities I meant exactly IC in first two hours since the tsunami. This was purely an operational task, not the headquarter's. And shift on duty should had solved it.

Pardon my English, please. I'm not from here.

 

[tsutsuji]

http://www3.nhk.or.jp/news/genpatsu-fukushima/20111231/index.html The ERSS system which collects plant data and monitors radiations in real time from the whole country became partly unavailable on 30 December at around noon. It recovered on 31 December afternoon.

 

[clancy688]

When I wrote about the time and possibilities I meant exactly IC in first two hours since the tsunami. This was purely an operational task, not the headquarter's. And shift on duty should had solved it.

tsutsujis translation made clear that operating the IC right after the tsunami hit would've been impossible anyway.
Because of the fail-safe function the isolation valves inside the containment closed when DC power was lost. And apparently it's not possible to operate them manually, at least not during operation. Even if you have access to the containment.

 

[jim hardy]

"""The blame lies squarely on TEPCO and on wider nuclear power industry for not spending relatively modest amounts of money and effort in preparations and training for severe accidents.

How much money was saved worldwide on not considering and not training for a prolonged SBO scenario? A billion dollar or so, tops?""

above post struck closer to the real trouble, imho.

when those 1995 reports of bigger tidal waves showed up somebody should have set down with TEPCO executives and showed them how their diesels -in-the-basement made them sitting ducks.
if somebody decided to ignore them, i'd hate to be him.

and yes pakman at my plant we drilled on severe accidents including loss of everything.

 

[tsutsuji]

tsutsujis translation made clear that operating the IC right after the tsunami hit would've been impossible anyway.
Because of the fail-safe function the isolation valves inside the containment closed when DC power was lost. And apparently it's not possible to operate them manually, at least not during operation. Even if you have access to the containment.

As long as the status of the PCV inner side valves is unknown, I think it would be appropriate to assume that they are at least open a little bit and try to run the IC as much as the inner side valves status can allow. That means trying to do everything possible to open the outer side valves of both IC (A) and IC (B) :

* open them manually with a handle if they are equipped with a handle
* open the motor housing and try to turn the axis manually using tools or improvise a makeshift handle
* bring in batteries and connect them to the motor

The Quebec Bridge collapsed in 1907 because the top engineer stayed in his office in the United-States instead of going to the construction site in Canada to look at the bridge and talk with the workers.

Should not plant manager Yoshida have gone to the control room, bringing a cup of coffee to the operators, and talk with them to undestand their mood, instead of staying inside the seismic-isolated building ?

 

[Pakman]

tsutsujis translation made clear that operating the IC right after the tsunami hit would've been impossible anyway.
Because of the fail-safe function the isolation valves inside the containment closed when DC power was lost.

Pretty old news. I read this here back in May. There are latest conclusions that the inner valves was not fully closed, if any.

Any way, operators on duty couldn't know about this freaky interlock. They should try to operate with outer valves in any case. Why they had not even tried?

 

[nikkkom]

Pretty old news. I read this here back in May. There are latest conclusions that the inner valves was not fully closed, if any.

Any way, operators on duty couldn't know about this freaky interlock. They should try to operate with outer valves in any case. Why they had not even tried?

We do know why they did not try (at least did not try hard enough). They were not prepared to SBO. Basically, they found themselves in a situation they were never trained for. Therefore they had to improvise - which is much harder to do than to go through a checklist. Switching on ICs wasn't the only obvious thing to do - for one, they might be also worried that piping and other structures might be damaged. They had to make choices under stress, and they did not make the optimal ones.

Almost continuous aftershocks, tsunami debris and water everywhere were not exactly helpful too.

 

[NUCENG]

This is a very easy problem. Incorporate both sensitive and high-range radiation meters into one device. Add a bit of logic which automatically displays high-range reading when sensitive one goes off-scale high.

Caveat - the following is based on my experience in US nuclear power plants.
Part of the reason for making different range dosimeters is to emphasize the importance of an alert on a high range dosimeter. It is simple human factors that carrying that high range dosimeter is a special case requiring careful self ALARA practices. Such things as having a low dose standby area to wait during communications or for other work interruptions.

Health physicists take initial surveys and prepare maps that are used to brief workers prior to entering high radiation areas. In emergencies operators are trained and may do self-surveys. In both cases the initial surveyor will use hand carried dosimetry, not just personal dosimeters.After the initial survey the type of dosimetry to be carried by subsequent workers is specified and alerts and alarms are established based on expected radiation levels and the dose margin for the worker. If that worker gets a dose rate alarm it means something has changed and training is for him to leave the area and contact the HPs. If he gets a dose margin alarm it means he has reached his dose limit and he is trained to do the same - leave and contact HP.

 

[NUCENG]

Нi NUCENG. Nice to join your discussion.

You say "some degree"? I'd say it's the highest level of.

Imagine, you are Unit 1 Shift Superviser. You get all indicators extinguished in UCR. You also have the reactor just shut down and isolated. Last time you saw the IC status, it was switched off. Wouldn't you send man to the location to check if the reactor is cooled or not? Wouldn't it be your priority task right now? Or you will be waiting for when the electric staff turn the light on?

When I wrote about the time and possibilities I meant exactly IC in first two hours since the tsunami. This was purely an operational task, not the headquarter's. And shift on duty should had solved it.

Pardon my English, please. I'm not from here.

I say some degree, because I am still trying to understand everything that tsutsuji has interpreted from the interim investigation report. I believe the operator who iniotially turned the IC off because of the rapid cooldown rate was a mistake. However, if the procedure didn't cover the emergency cooldown case which this accident was, he followed the procedure. In US plants we have cooldown rate limits, but they can be exceeded in an emergency. It just means there will have to be a lot of analyses before a restart.

I believe subsequent IC restart failures could have been caused by loss of natural circulation in the IC system due to non-condensible gases (hydrogen) air-binding the system. If that is the case the valve problems they encountered were of little impact.

As to what I would have done, I would like to think I woud have saved the day, but it is also possible I would have been equally bewildered as the Fukushima operators apparently were. Lack of lighting, failed instrumentation, high radiation, worrying about loved ones that may have been in the tsunami zone, I am thankful that I wasn't in that position. I do know the operators stayed at there posts and did what they could. Even though they failed that took courage.

If you read through this forum you will see my opinion that the managers and regulators who failed to act on knowledge of the tsunami threat ten years ago. If they had acted, this event could have been more similar to North Anna than Chernobyl. I find it hard to blame the reactor operators who were stuck with the result of that failure to act.

 

[NUCENG]

tsutsujis translation made clear that operating the IC right after the tsunami hit would've been impossible anyway.
Because of the fail-safe function the isolation valves inside the containment closed when DC power was lost. And apparently it's not possible to operate them manually, at least not during operation. Even if you have access to the containment.

I agree with your interpretation if tsutsuji and the system diagrams that have been published are correct. However I have looked at the GE patents for IC and it looks to me like the original GE design had MOVs outside containment and a check valve inside containment. The fact that they reported steam from the pig noses indicates to me that something was boiling the water in the ICs. I am looking for information from US plants that have ICs to see what they actually have. I believe tsutsuji's interpretation is accurate (as always) but there may be a problem with what the report says.

In US parlance the IC system allows use of Category C containment isolation since it is a closed loop outside of coontainment that allows the 2nd isolation valve (inside containment) to be a check valve.

 

[rmattila]

In US parlance the IC system allows use of Category C containment isolation since it is a closed loop outside of coontainment that allows the 2nd isolation valve (inside containment) to be a check valve.

That would be the return line valve (MO-4A/B), but for the outgoing line (MO-1A/B), there should still be some kind of actively operated valve.

 

[NUCENG]

That would be the return line valve (MO-4A/B), but for the outgoing line (MO-1A/B), there should still be some kind of actively operated valve.

No argument, But there are options including pneumatics or actually having both isolation valves outside containment. I am just trying to resolve why the Japanese seem to have departed from the original containment isolation design, even if it is only on the cold leg. It just tweaks my experience that you would allow an ECCS system to be isolated and unrecoverable with loss of power to a valve that is inside containment. If I recall, F1-1 was actually built by GE, and I wonder why they would deviate from their original design. But that may be the case, despite my thinking. Right now the preponderance of evidence tells me they did it that way as Tsutsuji and the drawings show. But I'm wondering why.

The whole point of the IC is to allow the system to operate passively.

 

[rmattila]

No argument, But there are options including pneumatics or actually having both isolation valves outside containment. I am just trying to resolve why the Japanese seem to have departed from the original containment isolation design, even if it is only on the cold leg. It just tweaks my experience that you would allow an ECCS system to be isolated and unrecoverable with loss of power to a valve that is inside containment. If I recall, F1-1 was actually built by GE, and I wonder why they would deviate from their original design. But that may be the case, despite my thinking. Right now the preponderance of evidence tells me they did it that way as Tsutsuji and the drawings show. But I'm wondering why.

I've been wondering about the same solution myself. I've not (yet, however - that might change in the near future) been involved with a reactor equipped with an IC, and the complications posed by the outgoing line. In the Finnish system, which as far as BWRs are concerned was based on the ASEA philosophy, the inner valve is usually a check valve, and the outer valve is DC operated or pneumatic. But for the emergency cooling systems, the isolation interlock (the "H" isolation in ASEA terminology) works the other way round, i.e. loss of DC will prevent, rather than cause, closure of the outer valves.

Another issue related to the fail-safe modes in Fukushima that I was not aware of previously is that none of the RPV pressure relief valves failed open as the result of the loss of DC. This is of course quite natural, since both the IC and RCIC rely on the intact primary circuit to ensure sufficient cooling, but it was nevertheless arguably an important factor in the delay to pump firewater into the reactors after the RCICs had failed. In our BWRs, those pressure relief lines that are equipped with DC operated control valves have their quick-opening valves failing open, which is a precaution to prevent high-pressure melt-through of the RPV, which might compromise containment integrity in a severe accident. Furthermore, these valves are equipped with pneumatic connections to enable keeping them open with pressurized nitrogen to make sure the RPV pressure can be lowered equal to the containment even in total blackout.

My personal feeling is that re-evaluation of the different fail-safe modes of valves in the emergency cooling systems should be one item to deserve attention based on the unfortunate experiences from Fukushima Dai-ichi.

 

[jim hardy]

"If I recall, F1-1 was actually built by GE,..."

if indeed there was a design shortcoming -

here's an absolutely unfounded guess based on nothing except years of watching big organizations undertake big projects.

A/E builds containment, to meet NSSS suppliers specs.
GE was NSSS,
Who was A/E? Seems to me Ebasco did unit one, Japanese industry the rest.
what was their standard in 1968-69 timeframe ?

design trouble is almost always at an interface...two groups of bright people and a 'failure to communicate..'

 

[gmax137]

design trouble is almost always at an interface...

that's what I've seen too, but the failure position issue can be very tricky. If the F1 units differ from the standard GE design, maybe there are documents outlining the rationale. Someone at GE could probably dig it up if they had a mind to...

 

[Pakman]

We do know why they did not try (at least did not try hard enough). They were not prepared to SBO. Basically, they found themselves in a situation they were never trained for.

It's quite impolite to think about Fukushima operators as about circus monkeys who do only what they are trained for.

 

[Pakman]

Lack of lighting, failed instrumentation, high radiation, worrying about loved ones that may have been in the tsunami zone, I am thankful that I wasn't in that position.

All of us are. But the story's not abut us.

 

[steve olsen]

It's quite impolite to think about Fukushima operators as about circus monkeys who do only what they are trained for.

True training in disaster preparedness is unlikely to happen. Why spend money on a "black swan" which is so unlikely? As much as I like and respect the Japanese culture, they do far better at operating inside the box, than thinking outside the box.

 

[Pakman]

In our BWRs, those pressure relief lines that are equipped with DC operated control valves have their quick-opening valves failing open, which is a precaution to prevent high-pressure melt-through of the RPV, which might compromise containment integrity in a severe accident. Furthermore, these valves are equipped with pneumatic connections to enable keeping them open with pressurized nitrogen to make sure the RPV pressure can be lowered equal to the containment even in total blackout.

It looks like your NPPs are well prepared for total SBO.

With your SBO fail-safe logic, do you realize that in case of depressurization from nominal pressure to about atmospheric the reactor loses at least 38% of its Inventory immediately due to evaporation? With this in mind, TAF will be reached in a very short time, if any time is needed at all. You believe you can implement the firewater to cool the core before the melting starts?

 

[rmattila]

With your SBO fail-safe logic, do you realize that in case of depressurization from nominal pressure to about atmospheric the reactor loses at least 38% of its Inventory immediately due to evaporation? With this in mind, TAF will be reached in a very short time, if any time is needed at all. You believe you can implement the firewater to cool the core before the melting starts?

Most probably not, and due to the passive filtered venting in the containment, there would probably be too much backpressure anyhow to get firewater in. The purpose of the fail-open of these two small-capacity blowdown lines is rather to prevent high-pressure melt-through, not to enable injection of firewater in case of total loss of DC.

Normally (=if DC is available for the measurement circuits), the depressurization signal is triggered by TAF +0,7 m with some delay, and its purpose is to first of all switch from the high-pressure core injection mode to attempting low pressure core injection, and secondly to reduce the pressure in case the low-pressure core injection would fail and there was a risk of melt-through. This forced blowdown will use a larger number of blowdown lines, not just those two small ones with the fail-open valves.

The weakness of the later ASEA BWRs is the total reliance on AC driven emergency cooldown systems in contrast to the GE line of reactors. But on the other hand, this means that there's no need to keep the reactor pressure high to enable the operation of the cooling systems, which makes it easier to depressurize the reactor in case of emergency.

It's not obvious which approach is the best - it's all a matter of optimization. IC and RCIC are good since they don't need electrical power to operate, but if they fail, it seems that pressure relief was too difficult to manage to be succesful.

 

[Pakman]

I believe subsequent IC restart failures could have been caused by loss of natural circulation in the IC system due to non-condensible gases (hydrogen) air-binding the system. If that is the case the valve problems they encountered were of little impact.

I believe if the IC restart would be taken earlier, before melting, there would be no problem of natural circulation.
The melting started at about 17:40. SBO occurred at abut 15:37.
Two hours seems to be enough to stop fearing, to realize things and to start acting.

I just want to know, what does the phrase in report: "This was because it was wrongly assumed that the IC was operating normally" exactly mean. That Japanese NPP operators are trained to assume instead of to know?

 

[Pakman]

but we're trained to not go where it's unknown
and we're trained that to go into a very high area one takes special equipment and precautions and plans the entry carefully...

Here's what came to mind: to follow these rules in a severe accident is like for pilots of passenger plane to be prepared to bail out with parachute.

 

[jim hardy]

"Here's what came to mind: to follow these rules in a severe accident is like for pilots of passenger plane to be prepared to bail out with parachute.""

do you then suggest they descend into mountain clouds with no altimeter?

 

[Rive]

It's not obvious which approach is the best - it's all a matter of optimization. IC and RCIC are good since they don't need electrical power to operate, but if they fail, it seems that pressure relief was too difficult to manage to be succesful.

- are they really exclude each other?
- the cold leg of RCIC can be kept cold with low pressure input, it's just matter of heat exchangers in the wetwell.

 

[Yamanote]

Let's get back to presence, they are now dealing with a leak in the skimmer surge tank and/or the spent fuel pool of unit 4:

http://www.tepco.co.jp/en/press/corp-com/release/12010201-e.html

I think this could become a quite dangerous incident.