Japan Earthquake: Nuclear Plants at Fukushima Daiichi

[SteveElbows]

The 'spanner' (look it up in the thread here) was more in the way of the destruction than the FHM and it's still on the top level... More or less. So I think the FHM should be there too.

Ah yes, also known by names such as the Stud Tensioner Carousel. I don't know why you think that equipment was more in the way of the destruction than the FHM. It was situated on floor that collapsed to a lower level, but I can't really do a meaningful comparison between what it experienced and the forces that the FHM may have been subjected to. I don't know as the FHM may have been launched far into the air, I could imagine it being blown more south than upwards, and it could have tipped over and tumbled out of the side of the building.

 

[LabratSR]

New Video From TEPCO - Radioactive Water Treatment 4: Fukushima Daiichi
http://youtu.be/Bglf9ZlwrpU

 

[NUCENG]

Sure. I am not implying that 90 year old fuel is harmless. Of course not. It is still so radioactive that it can kill nearby human in seconds if unshielded.



Me too.

I see no harm, though, in keeping spent fuel in dry casks for many years before reprocessing. It actually should make reprocessing easier. In 100 years, Kr-85 decreases by the factor of 1000 (meaning that airborne releases are greatly reduced), Cs-134 by 10^15 (practically zero), Cs-137 and Sr-90 by 10. What's not to like?



Sure. Unless someone is willing to make an electric generator powered by heat of canisters filled by Cs-137 and Sr-90. :D

A timely article about reprocessing. If we discuss this I would suggest we start a new thread.

http://green.blogs.nytimes.com/2011/11/16/a-long-long-road-to-recycling-nuclear-fuel/

Edit: And another article about Yucca Mountain. Note there is not one technical argument is this debate only Politics. That is not the right basis for a decision in either direction.

http://www.lvrj.com/news/illinois-senator-drops-amendment-to-revive-yucca-mountain-133978238.html

These two articles are examples of why I believe spent fuel is a significant issue. Failing to move forward on spent fuel increases risk. Failure to make decisions may be the decision that determines the future of nuclear power. It is irresponsible, but typical of other issues we face on energy, the environment, in the economy, in international affairs, in education, and on social services (and that applies to the world in general - not just the US).

 

[zapperzero]

Yes that's the one. Any ideas about it? I wondered if perhaps it had been mishandled and damaged at some time in the past, and so was kept on its own?

It does look a bit crooked, now that you mention. But I also saw a slant in the Unit 4 walls which wasn't there, so...

 

[LabratSR]

Preparations for the winter season in Fukushima Daiichi NPS

http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_111118_02-e.pdf

 

[etudiant]

TEPCO has issued its latest summary report on the efforts to control the Fukushima accident here:
http://www.tepco.co.jp/en/press/corp-com/release/11111701-e.html
The scale of the effort is somewhat reassuring, there does not seem to be any obvious cheeseparing.

What concerns me is that this scale of response is only possible in a rich society.
It is very unlikely that anything comparable could be done in say Pakistan, another country that has been known to experience large earthquakes, yet the growth in nuclear is in these poorer societies. Projecting out over the next 50 years, we can expect several other similar accidents. with gradually increasing global background burdens, unless a dramatically more robust reactor design is found.
Is this even a consideration currently in the international regulatory apparatus?

 

[clancy688]

unless a dramatically more robust reactor design is found.

It is probably already found. With all those (yet unbuild) GenIII reactors. But that won't help us the slightest when there are still literally hundreds of reactors from the sixties and seventies in use. And probably will be for the next 20 years.

 

[Rive]

... Projecting out over the next 50 years, we can expect several other similar accidents.

Are you sure? Do you know about more old designs built on such stupid places, with underestimated environmental risks?

It's still not the reactor what failed there on the first place, but the placement and the risk analysis.

 

[etudiant]

Are you sure? Do you know about more old designs built on such stupid places, with underestimated environmental risks?

It's still not the reactor what failed there on the first place, but the placement and the risk analysis.

Of course I'm not sure, but the experience to date seems to suggest that.
Afaik, we have somewhere around 500 reactors in service for generating commercial power.
Three have experienced destructive failure, through operator error or external events, during at most 50 years of operation, so we have about 1 major failure per 10,000 reactor years of operation.
It would seem important to try to improve that by at least a factor of 10, preferably a factor of 100.
Is that even possible, given that we are apparently not very good at identifying, much less managing low probability events?

 

[rmattila]

If you aim at
-operational occurrence rate of less than once per year
-limiting functions capable of preventing 99 % of occurrences from propagating into accidents
-safety functions capable of preventing core damage in 99 % of accidents
-containment capable of preventing a large release in 99 % of meltdowns

you will not have to deal with smaller probabilities than 1e-2. Instead, you will have to deterministically ensure that each of the levels reaches its goal and is independent of the other levels. If this can be guaranteed, the probability of a large release is around 1e-6 per annum, which is generally deemed acceptable.

Old plants fail miserably with the last bullet, and have to try to compensate the incapacity to deal with consequences of a severe accident with somehow even more improved preventive measures. This will inevitably result into a more challenging safety case to ensure the acceptable safety level.

 

[NUCENG]

If you aim at
-operational occurrence rate of less than once per year
-limiting functions capable of preventing 99 % of occurrences from propagating into accidents
-safety functions capable of preventing core damage in 99 % of accidents
-containment capable of preventing a large release in 99 % of meltdowns

you will not have to deal with smaller probabilities than 1e-2. Instead, you will have to deterministically ensure that each of the levels reaches its goal and is independent of the other levels. If this can be guaranteed, the probability of a large release is around 1e-6 per annum, which is generally deemed acceptable.

Old plants fail miserably with the last bullet, and have to try to compensate the incapacity to deal with consequences of a severe accident with somehow even more improved preventive measures. This will inevitably result into a more challenging safety case to ensure the acceptable safety level.

Your point of generalizing old plants fails by example. TMI2 was a meltdown accident in an older plant with no large release.

Chernobyl was a deliberately initiated event, the safety systems had been deliberately disabled, and there was no containment.

Fukushima was a 1E-3 event (earthquake/tsunami), the safety systems worked initiially, but failed due to lack of protection from flooding. I am not sure that containment failures at Fukushima would have occurred in a plant that didn't have such a glaring design deficiency.

And finally on your risk targets. PRA has exactly those kinds of targets. The problem comes when estimating probabilities for rare events and anticipating all the threats. TEPCO underestimated the threat of tsunami and didn't design turbine buildings to resist flooding. I understand they had completed the IPE level analysis (internal events) and some level of IPEEE analysis (External Events). I am guessing their results showed that they met the acceptable risk goal you describe.

The 1E-3 tsunami greater than 5.5 m was not recognized. failures of individual safety systems and power sources had probably already met or exceeded your 1E-2 failure probabilities. But that had not recognized the common failure mode of flooding. Suddenly all of the safety systems had a failure probability of 1.0. Containment never had a chance.

The biggest lesson to be learned from Fukushima is the importance of safety culture at all levels from National Regulators, to vendors, to the utility management, to operators, to engineers, and technicians, security and general laborers. The single biggest tool for safety is maintaining a questioning attitude. Thus when the geological evidence of large tsunamis was made known the question should have triggered action by regulators, managers, and techical staff. It didn't, and I attribute that to complacency, and a lack of integrity. Anyone with knowledge of the risk who didn't force the issue is at fault.

I don't know how to measure safety culture or include it in a PRA. Rare accidents are obvious triggers of attention to safety, but the real need is to use every issue, equipment failure, and problem, no matter how small, as a similar trigger. If a breaker trips or a fuse blows, don't just reset or replace it. Consider the circumstances when it blew. Was the operation abnormal? Was the circuit overloaded? Is the fuse or breaker the right size or rating? This sort of thinking becomes a habit. If you know people who have worked in the US nuclear industry, you may have observed this kind of thinking. It is why I believe that Fukushima technical solutions will be applied in US plants. It is not enough to simply implement techical solutions from from Japan, the safety culture must also be part of the mix.

 

[Most Curious]

TMI2 was a meltdown accident in an older plant with no large release.

An older design but, in fact, a reactor that had not been in operation for very long.

No appreciable release to the environmnet, of course.

Man made machines can and do fail - in spite of our best efforts. We learn from those incidents and make the next one LESS likely.

 

[NUCENG]

An older design but, in fact, a reactor that had not been in operation for very long.

No appreciable release to the environmnet, of course.

Man made machines can and do fail - in spite of our best efforts. We learn from those incidents and make the next one LESS likely.

I initially posted before I completed my thoughts - fat finger syndrome. I think what I added is consistent with your point.

 

[Most Curious]

Yes, NUCENG, we agree. I just wanted to make sure the "aging plants" folks did not get any more ammunition through misunderstanding what you meant.

Two ways to look at "aging plants": One is that old designs like Fukushima, even if newly constructed, have many of the faults of the original design on which the plant is built. The other is the accumulated operating hours and "wear out" of the components, buildings etc. The shroud failures in old BWR plants, as an example.


Designs evolve, including NPP. Hopefully, if engineers are guided by science, not politics or cost savings at the expense of safety, each newer generation design addresses safety issues of previous designs, particularly those revealed by accidents like Fukushima. Of course, there are thousands of other less dramatic issues that get addressed as well, such as materials, maintenance access to critical componets, etc.

The "build no new NPP" folks effectively doom us to obsolete designs that are less safe than later generations, just the opposite of what they profess to want. In addition, if we wait for the "perfect design" that has zero risk, no progress will EVER be made, no new plants will EVER be built.

 

[tsutsuji]

http://www.fnn-news.com/news/headlines/articles/CONN00211924.html Tepco says that the work to remove outdoor debris is almost finished.

http://www.47news.jp/47topics/e/222695.php 1.6 Sv was found on unit 3's first floor on 20 November. On 16 November 1.3 Sv had been found nearby.

http://ajw.asahi.com/article/0311disaster/fukushima/AJ201111180006 Aerial photograph showing the large areas covered with water storage tanks.

 

[etudiant]

Imho the discussion of safety culture is not enough. For one, there is no evidence a safety culture can be maintained over many decades without getting ritualistic and stale, as it did in Japan. It did not in the US Navy either, afaik, as the Thresher accident demonstrated.
There has to be a serious focus on minimizing the failure at every level, just as rmattila said.
Otherwise it means we have learned nothing from Fukushimas finding out that events can be beyond plan.
Just as an example, the largest earthquakes in the US were about M8, centered on New Madrid, MO, about 200 years ago. Would current nuclear plants in the area shut down gracefully in the event of a recurrence or would the resultant large scale blackout have severe consequences?

 

[NUCENG]

Imho the discussion of safety culture is not enough. For one, there is no evidence a safety culture can be maintained over many decades without getting ritualistic and stale, as it did in Japan. It did not in the US Navy either, afaik, as the Thresher accident demonstrated.
There has to be a serious focus on minimizing the failure at every level, just as rmattila said.
Otherwise it means we have learned nothing from Fukushimas finding out that events can be beyond plan.
Just as an example, the largest earthquakes in the US were about M8, centered on New Madrid, MO, about 200 years ago. Would current nuclear plants in the area shut down gracefully in the event of a recurrence or would the resultant large scale blackout have severe consequences?

If you have a well implemented safety culture, actions to minimize risk happen automatically. I absolutely agree with you that we must continue to reduce risks where possible and especially when something happens to raise new issues.

As a US Navy officer, I was qualified as Engineer Officer for nuclear submarines. As a SUBSAFE coordinator during a submarine refueling overhaul after the loss of the Thresher, I respectfully disagree. The Thresher accident was caused by a non-nuclear seawater pipe rupture. The pipe had been weakened in explosive shock testing before the yard period. There are reports that the sub made a navigation error during testing which put them closer to the detonation than planned. The sinking occurred during sea trials following repairs. The Navy instituted the SUBSAFE program to require rigorous testing and inspections of non-nuclear systems exposed to sea pressure following the accident. Nuclear systems were already tested to those standards and were not included in the program. I learned much of the habit of safety conscious thought wile I was in the Navy. Perhaps your experience was different. But I believe the Thresher is a poor example when indicting the Navy nuclear safety culture.

My understanding and reading are the basis for the following. I am not presenting myself as a civil/structural or mechanical engineer. The New Madrid earthquake was a series of three main shocks over short time in 1811/1812 and the estimates I have seen indicated they were about M7.4. The new Madrid earthquake was closer to the type of earthquake we just saw in Virginia in a region of low seismic activity with a relatively solid (unfractured) crust. Japan and the west coast of the US are in a region of high seismic activity with well fragmented crust. North Anna saw high acceleration at high frequencies and lower total energy than the basis of the plant design. If the energy is at lower frequencies it has a larger effect on the structures. Current USGS estimates are only about a 10% chance of an relatively small earthquake on the New Madrid fault in the next 50 years. I reviewed information for one plant located several hundred miles from the New Madrid fault. Their evaluation of the effects of the 1811/1812 earthquakes would not result in significant damage at the plant site. I also reviewed the hazard maps on the USGS website and confirmed that those predictions are still valid even relatively close to the New Madrid area.

That said the NRC is preparing regulatory guidance (a Generic Letter) that will require plants in the central and eastern states to reevaluate seismic risk and design based on recent findings of higher potential ground acceleration during earthquakes than USGS had previously estimated. The plant I reviewed is on the list of plants affected by the pending guidance. So like TEPCO, they are aware of the risk AND, unlike TEPCO, are taking action to address that risk. NRC has also concluded from both the GI-199 issue and the North Anna Earthquake that there is no immediate safety concern. A large scale blackout could occur if the grid is brought down, but a tsunami is not much concern in that area. On site emergency diesel generators should remain available. So I would expect a better result than Fukushima. (See the point I made earlier - that doesn't mean we can ignore Fukushima!)

 

[etudiant]

The Thresher disaster surely showed that safety is indivisible, that it is not enough for the reactor to be secure if the surrounding systems are not. To me, the analogy to Fukushima is very direct, in both instances the reactors functioned as expected and nevertheless disaster followed, because of unexpected ancillary failures. That is why as rmattila suggested there must be specific provision to allow for a relatively graceful reactor failure, because somehow over many thousands of reactor years of operations it will happen.

 

[NUCENG]

The Thresher disaster surely showed that safety is indivisible, that it is not enough for the reactor to be secure if the surrounding systems are not. To me, the analogy to Fukushima is very direct, in both instances the reactors functioned as expected and nevertheless disaster followed, because of unexpected ancillary failures. That is why as rmattila suggested there must be specific provision to allow for a relatively graceful reactor failure, because somehow over many thousands of reactor years of operations it will happen.

All I have said is that safety culture makes complacency less likely. It is not a substitute for implementing lessons learned but makes those improvements almost in automatic. In addition by taking action on little problems it makes cascading small problems into big problems much less likely.

I really don't think there is an argument here. The link between Fukushima and the Thresher may be direct in your opinion, but it certainly is not current. USS Thresher sank in 1963. The Navy implemented your "INDIVISIBLE" safety concept 48 years ago. You used it as an example of why safety cultures don't endure. I don't think the Thresher era has any current validity. Heck that was even before I entered the Navy, and I haven't seen any SUBSUNK reports lately. Sounds to me like a pretty good safety record for nearly 5 decades.

But I have also agreed with rmattila and you that actions need to be taken based on Fukushima. And finally that action is needed even though there is a high probability a similar initiator (earthquake) in the US would show successful mitigation. And I believe that action is underway to take that action. So what is missing?

 

[Borek]

My understanding is that Tresher reactor doesn't pose any risk to the environment. In a twisted way it proves that correct design makes reactor safe even in the case of a serious disaster.

 

[dezzert]

The "build no new NPP" folks effectively doom us to obsolete designs that are less safe than later generations, just the opposite of what they profess to want. In addition, if we wait for the "perfect design" that has zero risk, no progress will EVER be made, no new plants will EVER be built.

And you have also just given the reason why in the near future all old plants will be shutdown. And I am amazed at anyone at this point in time who still tries to justify the need for the most dangerous resource ever created, nuclear power. Thank god there are intelligent people on this planet that know better.

 

[Shinjukusam]

And you have also just given the reason why in the near future all old plants will be shutdown. And I am amazed at anyone at this point in time who still tries to justify the need for the most dangerous resource ever created, nuclear power. Thank god there are intelligent people on this planet that know better.

You know, there is a thread specifically for more political discussion...

 

[NUCENG]

And you have also just given the reason why in the near future all old plants will be shutdown. And I am amazed at anyone at this point in time who still tries to justify the need for the most dangerous resource ever created, nuclear power. Thank god there are intelligent people on this planet that know better.

Sorry, in a list of the most the most dangerous resource ever created on this planet, nuclear isn't anywhere near coal (including TMI2, Chernobyl and Fukushima). Even if you don't believe in man-made global warming (or climate change), acid rain, mining accidents, transportation accidents, health effects of particulates and other emmissions, and waste issues make your overstatement clear and your prediction dubious.

Returning to my point of a safety program that has workd, when NASA restarted benchmarking efforts to improve their safety culture and performance they went to the Navy Nuclear Propulsion and SUBSAFE Programs.

They found:

"It is important to note that NASA’s benchmarking of the Navy submarine program has
been focused on the SUBSAFE and Naval Nuclear Propulsion Programs with full
understanding that these programs represent only two of the Navy submarine safety
domains ... This deliberate selectivity results from an early consensus of
the NNBE management team that these two high reliability programs would provide the
most meaningful comparison to NASA’s human rated space flight programs."

Specifically, they noted the record of the US Navy Nuclear propulsion safety record:
Through 2003... "Since its inception in 1948, the NR program has developed 27 different plant designs,installed them in 210 nuclear powered ships, taken 500 reactor cores into operation, and accumulated over 5,400 reactor years of operation and 128,000,000 miles safely steamed. Additionally, 98 nuclear submarines and six nuclear cruisers have been recycled."

There has never been a nuclear accident in any US Naval Nuclear Propulsion Plant or prototype.

Reference: http://www.nasa.gov/pdf/45608main_NNBE_Progress_Report2_7-15-03.pdf

 

[rmattila]

Your point of generalizing old plants fails by example. TMI2 was a meltdown accident in an older plant with no large release.

Chernobyl was a deliberately initiated event, the safety systems had been deliberately disabled, and there was no containment.

Fukushima was a 1E-3 event (earthquake/tsunami), the safety systems worked initiially, but failed due to lack of protection from flooding. I am not sure that containment failures at Fukushima would have occurred in a plant that didn't have such a glaring design deficiency.

You have a point there. However, I was not thinking so much of the experiences we have had so far (the number of which is fortunately very limited), but rather the design bases of the containments. If you don't have large enough volume to accommodate all hydrogen produced by cladding oxidation, a full meltdown will probably result into a release. And if you don't have filters in the vent line, you will probably have a rather large release (and even if you have filters, they will not be able to catch noble gases or organic iodine, unless it's a large dry bed instead of the more compact wet scrubber type).

If you don't have a core catcher, and your containment does not allow for flooding of the drywell in case of melt-through (either due to the pools sitting lower than the drywell or due to fear of steam explosions), you have difficulty controlling the core-concrete interaction, which may result into a containment failure.

Etc. My point was simply that if a full-scale meltdown is not included in the original design basis of the containment, it's difficult to prove it can prevent release in 99 % of the cases, which would be a plausible target for new reactor designs.

 

[dezzert]

You know, there is a thread specifically for more political discussion...

I apologize. I was having a bad hair day. No more political stuff from me. Except the below post ;>)

 

[dezzert]

Sorry, in a list of the most the most dangerous resource ever created on this planet, nuclear isn't anywhere near coal (including TMI2, Chernobyl and Fukushima). ]

There is one major difference. If there is a world calamity, EQs like Tohoku, or just a shutdown of the grid, a coal plant no longer in use is just a coal plant no longer in use, where as a nuke plant is a disaster waiting to happen. Its the future that doesn't look bright.

If Fukushima is to teach us anything, its that the future is entirely unpredictable. And its that uncertainty that needs to be addressed, not whether or not nuclear can be made safe under current paradigms. If it can never be made entirely safe, why would we burden future generations with our need for electric shavers and all the rest.

 

[nikkkom]

There is one major difference. If there is a world calamity, EQs like Tohoku, or just a shutdown of the grid, a coal plant no longer in use is just a coal plant no longer in use, where as a nuke plant is a disaster waiting to happen.

Not necessarily. There is no law of physics which says NPP can't be designed to shut down safely without any electric power.

If Fukushima is to teach us anything, its that the future is entirely unpredictable.

Yes. Theoretically, tomorrow we may be invaded by aliens. So what?

If it can never be made entirely safe, why would we burden future generations with our need for electric shavers and all the rest.

Nothing can be made *entirely* (meaning: 100%) safe.

 

[dezzert]

Quick question concerning the SGTS and the R4 blast. I know that the valves fail open, but I don't understand why the hydrogen would find its way up the SGTS and through the filtration system on 4, (4 being an airtight structure, yes) when it had an obvious vent path through the 3/4 stack. Wouldnt there already be a draw occurring up the stack.

The amount of hydrogen needed to do the level of damage to R4, including damage as low as the 1st floor, would have been massive. How does this much hydrogen enter an airtight structure in these amounts, and penetrate down to the 1st floor from the 4th floor, while most is being exited out the stack, without having built to levels in R3 that would have caused it to explode much sooner.

And if this much hydrogen did enter R4, why did it not exit R4 (after the R3 blast) the same way it came in. After R3 blew the SGTS was severed at the R3 side of the stack, an open vent path to the outside. If the pressure that built up in R3 pushed that much hydrogen out the SGTS, why would not the subsequent pressure build up in R4 push it out the same line which is now open to the atmosphere (and without even having to go through a second filtration system).

This one never has worked for me and still doesnt. Any help is appreciated.

 

[dezzert]

Not necessarily. There is no law of physics which says NPP can't be designed to shut down safely without any electric power.

But there is a law of physics that says that any reactor designed to be water cooled cannot be shutdown safely without water. And you get the water into the plant how?

Yes. Theoretically, tomorrow we may be invaded by aliens. So what?

Not sure what aliens has to do with nuke plants, so I think Ill let this one slide.

Nothing can be made *entirely* (meaning: 100%) safe.

Excuse me? Have you ever heard of distributed power systems.
And the 'nothing can be made perfectly safe' argument is almost as ridiculous as referring to aliens. It all depends on the level of threat you are trying to be made safe from.

 

[Rive]

But there is a law of physics that says that any reactor designed to be water cooled cannot be shutdown safely without water.

Really? Can you please give me a link or something on that one?

Because I can think of some ways to achieve this (also not on-topic here)...

 

[nikkkom]

But there is a law of physics that says that any reactor designed to be water cooled cannot be shutdown safely without water.

Wrong. But anyway...

And you get the water into the plant how?

Water can be transferred by gravity alone.

 

[dezzert]

Really? Can you please give me a link or something on that one?

Because I can think of some ways to achieve this (also not on-topic here)...

I only joined the safety discussion because others brought it up. I agree its off topic. But I would like to know in what ways a water cooled reactor in an emergency like Fukushima can be cooled without it, and why weren't these alternatives brought forth earlier, like say March 11th.

 

[dezzert]

Water can be transferred by gravity alone.

Excellent idea. And how high would the source have to be to produce the amount of pressure needed during a meltdown. And how big would the pipe have to be to deliver water to 5 reactors, 6 SFPs, and one huge common pool in the amounts needed. Practically speaking.

But yes its off topic and I swear I won't go off topic again. End of the safety discussion for me. My only real interests at this time have to do with understanding the current situation. Which is why I was asking about the SGTS and R4. To me this is an important issue that needs deep analysis, and why I came on yesterday to post. I am sorry I got diverted into the safety discussion, because to me safe nuclear is an oxymoron, and therefor pointless to discuss IMHO.

 

[gregtomko]

Sorry, in a list of the most the most dangerous resource ever created on this planet, nuclear isn't anywhere near coal (including TMI2, Chernobyl and Fukushima).

There has never been a nuclear accident in any US Naval Nuclear Propulsion Plant or prototype.

Reference: http://www.nasa.gov/pdf/45608main_NNBE_Progress_Report2_7-15-03.pdf

When the total damage is assessed for coal, every byproduct is gone after a few decades. In nuclear much of the hazardous waste isn't gone for tens of thousands of years. If spent fuel is safe today, how do you know it will be secure 100 years from now? Do you feel comfortable handing off responsibility for your waste to future generations?

 

[clancy688]

Water can be transferred by gravity alone.

Which worked pretty well in the current case of Fukushima...

...as long as they were able to reduce pressure by venting the containments. Wait, they couldn't vent in some cases? So they had to wait until Unit 2 for example already melted down and containment failed, so that pressure reduced and water injection was finally possible...?

 

[rmattila]

Optimization is not always simple. For example, at Fukushima Daiichi, they were able to inject firewater into the reactor because the reactor pressure vessel could be vented down to a sufficiently low pressure.

But the ability to lower the reactor pressure low enough for fire engines would not be possible, if the containment vent lines had a passive scrubber (which would be a good thing from point of view of reducing the radioactive releases), and if the RPV blowdown valves would be steam-operated rather than require compressed air (which would be good from reliability point of view).

If you sum the 2-3 bar required by a typical passive wet scrubber, and the 2-5 bar required by self-powered RPV pressure relief lines, plus the 1-3 bar needed to overcome the hydrodynamic pressure caused by the height of the RPV, you end up with a counter-pressure high enough to prevent direct firewater injection by fire engines, which typically pump to 6-10 bar.

 

[Rive]

When the total damage is assessed for coal, every byproduct is gone after a few decades.

[rant on]
That depends on the person who makes the assessment.
For example I tend to consider the CO2 to being around for some million years, and I really wonder why some ash deposites considered for mining Uranium and some heavier elements.

[/rant off]

 

[Rive]

...

One thing always confuses me about the RCIC. Was it ever considered as a 'final' solution for SBO?

While there is enough pressure difference between the wetwell and the RPV it can maintain a continuous water flow to the RPV. As I know it has only two limits: the battery power and the heat capacity of the wetwell.

But I see no reason why its turbines can't be used also for emergency generators to power the RCIC control in exchange for some more heat to the wetwell.
And some low pressure heat exchangers installed in the wetwell could take care on the heat, even with makeup water sources, like diesel powered agrocultural pumps.

 

[gmax137]

One thing always confuses me about the RCIC. Was it ever considered as a 'final' solution for SBO?

Station blackout (loss of all AC power) was not an original design basis for plants of that time. In the US, SBO first appears as a design basis (AFAIK) during the licensing of St. Lucie Unit 2 (in the mid 1980s), where the event is defined as a blackout lasting four hours. Later, the SBO was added to the US regulations with a formula for determining the duration based on site characteristics. But it's always 4, 8, or 12 hours. The 'final solution for SBO' was, the power comes back on at 4 (or 8 or 12) hours. So, the RCIC (or turbine driven auxiliary feedwater) has to operate for a fixed, finite time.

What Fukushima drives home is (1) the absolute importance of the external event design bases (since these events can lead to common cause failures like loss of both trains of service water, or loss of both diesel generators, or loss of all the switchgear) and (2) the need to consider much longer duration SBO. A third point which some people see (me included) is a kind of flaw in the design basis concept - it misses the need to design to fail gradually if the design basis is exceeded (rather than the design basis being a 'cliff edge'). This last point is hard for some to grasp.

 

[tsutsuji]

http://www.tepco.co.jp/nu/fukushima-np/images/handouts_111122_02-j.pdf Assessment of the operation of unit 1's Isolation Condenser. An English translation will probably follow soon. I attach a partial translation of the last figure.

 

[nikkkom]

Excellent idea. And how high would the source have to be to produce the amount of pressure needed during a meltdown.

A few meters above the reactor would work. Look at IC condensers at Unit 1. Their shell side is at *1 atm* at all times. As long as you have water to pour into it, they will keep reactor cooled.

In emergency (such as "IC is damaged"), there should be a procedure to depressurize reactor vessels to ~1 atm and feed the water directly to the reactor vessels. IIRC it is normally not allowed because associated rapid drop in temperature can cause cracks in piping, necessitating costly repairs. But it is infinitely better than meltdown.

And how big would the pipe have to be to deliver water to 5 reactors, 6 SFPs, and one huge common pool in the amounts needed. Practically speaking.

Who says one pipe should deliver water to all reactors? That would be a bad design wrt safety. I would even require at least two separate gravity-fed water sources per each reactor.

 

[nikkkom]

> Water can be transferred by gravity alone.

Which worked pretty well in the current case of Fukushima...

...as long as they were able to reduce pressure by venting the containments. Wait, they couldn't vent in some cases? So they had to wait until Unit 2 for example already melted down and containment failed, so that pressure reduced and water injection was finally possible...?

My points are:

(a) At Fukushima they did not ever expect to be left without electricity. They were not trained for this. The accident manuals did not tell them what to do in this case. Even emergency lights went out - how come, aren't they supposed to be battery-backed?!?

(b) Even if they would be trained for this - they did not have gravity-fed water sources. All they had is water in IC good for about 8 hours of cooling. But they were NOT trained and as such, they failed to use even that!

Both these points need fixing - worldwide.

 

[thebluestligh]

Here are some composite images of the R4 SFP generated from the TEPCO video released on 11/11/2011

http://www.tepco.co.jp/en/news/110311/movie_1111/1111_19-e.html
http://www.tepco.co.jp/en/news/110311/movie_1111/1111_20-e.html
http://www.tepco.co.jp/en/news/110311/movie_1111/1111_21-e.html

 

[dezzert]

Does anyone here buy the SGTS explanation for R4s hydrogen buildup, and if so, why?

 

[NUCENG]

Does anyone here buy the SGTS explanation for R4s hydrogen buildup, and if so, why?

I am open minded on the cause of damage on Unit 4.

Initially I received information that operators believed a second explosion had occurred in Unit 3. Other theories were of steam explosions, which I never found convincing. We basically rehjected that theory because we couldn't find photographic changes in unit 3 indicating a second explosion. Then the possibility of hydrogen from damaged fuel in a drained spent fuel pool was proposed based on NRC statements that the pool was empty. That was proved wrong by photos showing no significant damage. For a while people were looking for welding gases supporting the shroud chane project or lubricating oils as sources for the explosion.

The next theory was radiolysis of water in the spent fuel pool. Calculations were made that showed that was possible if the pool was boiling. There was some good work on this by other posters.

Then TEPCO postulated the Unit 3 source via the stack piping. Apparently that has gained credence by the photos and inspections of the debris on the 3rd and 4th floors. It is relatively easy to see a difference between an explosion inside ducting and ducting that has benn crushed or damaged from an exterior explosion. The isometric sketches of the SBGT ducting seems consistent. We don't really know when the exterior duct failed on the side of unit 3. Recent information confirms the dampers fail open and there are no backdraft damper. The explosion of unit 3 could have followed a period when the hydrogen was present with increasing hydrogen concentrations and pressure forcing hydrogen into unit 4 or the explosion itself may have been the driving force. The explosion may not have been propagated to unit 4 because the hydrogen concentration in the ducting was to rich for deflagration or detonation. The timelag before the Unit 4 explosion could have been awaiting an ignition source or dilution to a combustible or explosive mixture.

In summary I still have unanswered questions. to change the maybes into facts. If I have to answer your question, I will default to "Mythbusters" terminology and say it is somewhere neither BUSTED, nor CONFIRMED, but somewhere between POSSIBLE, and PLAUSIBLE.

 

[NUCENG]

My points are:

(a) At Fukushima they did not ever expect to be left without electricity. They were not trained for this. The accident manuals did not tell them what to do in this case. Even emergency lights went out - how come, aren't they supposed to be battery-backed?!?

(b) Even if they would be trained for this - they did not have gravity-fed water sources. All they had is water in IC good for about 8 hours of cooling. But they were NOT trained and as such, they failed to use even that!

Both these points need fixing - worldwide.

Just a partial answer. On (a) the latest reports revealed that not only EDGs were flooded, but the essential AC and DC electrical panels as well in some of the plants. That may have explained why even some DC systems or lighting failed. The timelines also suggest that some time short elapsed before flooding took out some of the systems. Some time later they were repeatedly frustrated by bringing in external systems only to see them wiped out by explosions. On (b) only Unit 1 had an IC and there is conflicting information about its use. .

 

[NUCENG]

Station blackout (loss of all AC power) was not an original design basis for plants of that time. In the US, SBO first appears as a design basis (AFAIK) during the licensing of St. Lucie Unit 2 (in the mid 1980s), where the event is defined as a blackout lasting four hours. Later, the SBO was added to the US regulations with a formula for determining the duration based on site characteristics. But it's always 4, 8, or 12 hours. The 'final solution for SBO' was, the power comes back on at 4 (or 8 or 12) hours. So, the RCIC (or turbine driven auxiliary feedwater) has to operate for a fixed, finite time.

What Fukushima drives home is (1) the absolute importance of the external event design bases (since these events can lead to common cause failures like loss of both trains of service water, or loss of both diesel generators, or loss of all the switchgear) and (2) the need to consider much longer duration SBO. A third point which some people see (me included) is a kind of flaw in the design basis concept - it misses the need to design to fail gradually if the design basis is exceeded (rather than the design basis being a 'cliff edge'). This last point is hard for some to grasp.

I would ad a bit of a clarification to your cliff edge analogy.The problem with Fukushima, and which may be present elsewhere, is that the unrecognized seismic and tsunami risk was indeed a cliffedge.

From the very beginning of nuclear plant design, there was a recognition that accidents that exceed the design basis are possible. So plant designs had two levels of design and analysis This remains true today.

The design basis for safety systems is to prevent core damage that would release any radiation. This was based by conservative deterministic analysis with margin. In the case of containment safety systems the design basis was assumed to be the safety systems had failed resulting in an "arrested core melt accident" (Arrested meaning the core was damaged, but stopped after a partial melt. Again conservative, deterministic analysis was used to assure that containment would limit radiation dose to workers and the public within limits. This included margin for system leakage. The offsite release models were based on limiting atmospheric models and that the persons exposed were at the site boundary for two hours during the worst radiological dose or continuously in the low population zone for thirt days with no evacualtion.

The second type of analysis used is for severe accidents. Here the use of probabilistic approaches is allowed. This is supported by PRA for events that resu;lt in core damage and for containment failures. The consequences are also treated as probabilities. The WASH-1400 study was the first example of a systematic PRA approach to bring it all together. Other studies have followed, indicating the WASH-1400 study overestimated the consequences. NRC has initiated a recent update in the SOARCA project.

If you think about it, TMI2 was consistent with a beyond design basis reactor accident but within the design basis of the containment. (The lack of containment at Chernobyl is outside the process I am describing.) Fukushima is a severe accident but the consequences to date seem to confirm much of the severe accident analysis.

To relate this back to your post the SBO probabilities were used to justify the coping periods. They were based on grid performance studies that are available in reports from national labs. In addition to PRAs there are Integrated Plant Examinations of External Events (IPEEE) that are PRAs for external events. Clearly the Fukushima lessons learned witll include three basis response areas. The IPEEE for seismic and flooding events will need to be reviewed and updated. The basis of the SBO programs will need to reconsider duration bases on changes to IPEEE and likely new regulations. Design changes will probably need to be installed. PRAs and IPEEEs will ned to be updated reflecting new plant modifications. And during this whole process the results of each step will need to be reviewed to identify other vulnerabilities and guide corrective actions and modifications to areas with the greatest impact.

So my contention is that the current design of plants provides complete protection for design basis events and a reasonable process to install additional protection should accidents get past that point. As a result, absent a glaring design deficiency such as the Fukushima tsunami protection, it may be closer to a hillside than a cliff. I expect that the slope of that hill will be even shallower when the process is complete.

 

[NUCENG]

When the total damage is assessed for coal, every byproduct is gone after a few decades. In nuclear much of the hazardous waste isn't gone for tens of thousands of years. If spent fuel is safe today, how do you know it will be secure 100 years from now? Do you feel comfortable handing off responsibility for your waste to future generations?

It would be unfair of me to interpret that as a defense to continued use of coal over nuclear. I think what you meant is the long term storage of spent nuclear fuel is one issue where coal is "less harmful." Is that fair?

I agree the spent fuel issue remains unsolved, but the volume of nuclear waste is miniscule compared to fly ash, which is a larger uncontrolled source of environmental radiation than spent fuel in geological storage. The chances of controlling access to that spent fuel will be much lower if we can eventually combine it in a single repository.

As to the distant future, I don''t know. That doesn't mean I don't care. But I do recognize that failing to provide for energy sources for the future will not help solve that issue - after society collapses.3

 

[NUCENG]

There is one major difference. If there is a world calamity, EQs like Tohoku, or just a shutdown of the grid, a coal plant no longer in use is just a coal plant no longer in use, where as a nuke plant is a disaster waiting to happen. Its the future that doesn't look bright.

If Fukushima is to teach us anything, its that "the future is entirely unpredictable. And its that uncertainty that needs to be addressed, not whether or not nuclear can be made safe under current paradigms. If it can never be made entirely safe, why would we burden future generations with our need for electric shavers and all the rest.

Respectfully, "the future is entirely unpredictable, and that uncertainty needs to be addressed" assumes that we might somehow be able to make the future predictable? I doubt that was what you meant.

In the same vein you said, "If there is a world calamity ... a nuke plant is a disaster waiting to happen." Is a disaster on top of a calamity worse than the calamity?

Unfortunately, I do not know how to make the future predicatable or to prevent furure calamities. I do know that 7 Billion people cannot be sustained on this Earth without technology and a stable society. Providing safe and reliable energy is essential for preserving that stability. Right now I do not think we have any other alternative than to include nuclear power in the mix of energy sources.

 

[nikkkom]

Just a partial answer. On (a) the latest reports revealed that not only EDGs were flooded, but the essential AC and DC electrical panels as well in some of the plants. That may have explained why even some DC systems or lighting failed.

Interesting.

I though "battery-backed emergency lighting", when used in the context of nuclear power plants, means lighting *integrated with battery*. Maybe even the lamp and battery in an air- and water-tight unit. With photodiode detector which switches it on automatically when it detects darkness. You know, something designed to be fail-safe.

How naive I was...