Japan Earthquake: nuclear plants Fukushima part 2

[zapperzero]

If you think during a severe accident that there's going to be some easy way to do ANYTHING you're going to have a bad time.

I don't, so I think it is all the more important to ensure that there will be as few things to do as possible.

You would need a deep knowledge of the plant, or existing pre-staged procedures and equipment (like the ones the US has had since 9/11) to know which lines likely did not isolate, or know which lines only have isolation valves outside of containment. Example, the third LPCI (Low pressure coolant injection) system at my plant only has an outboard containment isolation valve, (the inboard valve is a check valves), and this is readily accessible and could be a good place to hook a fire truck up to.

This is exactly the kind of thing I am talking about. This need for in-depth knowledge is a vulnerability in and of itself.

The point I'm trying to make, is in all causes you will need to take manual actions. That is the definition of how a severe accident works. If you didn't need manual actions, then you wouldn't be in the severe accident in the first place.

So we could include future Fukushima-type scenarios in the set of non-severe accidents, if only we had the means to obviate the need for manual actions when they happen. Cool!

There is nothing passive that's going to help you. "Passive" filter? Only if you can get the first outboard valve open (and approval to have a pipe penetrating containment without double isolation). Or are we saying that these valves are going to be pre-aligned to start venting automatically (which means during DBAs like a LB-LOCA where my ECCS is working, I'm going to allow unacceptable and unnecessary radioactive releases because my passive filter is going to take care of it?, when my safety systems on site could readily handle it)

I was thinking more along the lines of a vent line with a rupture disk set at some level where you can be reasonably sure that some break will soon develop somewhere else anyway and another line which is controlled with valves in the usual manner. Of course, since this kind of stuff is already in production (rmatilla has posted lots of details a while ago iirc?), I don't need to think much :).
Also, who said anything about unacceptable releases? If your filter is big enough, and it gets used, you won't ever release anything unacceptable, no?

 

[Rive]

Such high doses that there are measured
(and in the torus is 2.2 Sv) can only nuclear fuel.

I would not jump to that conclusion so fast. There was that vent tower bottom, with around 10Sv/h as I recall, far away from the containment...

 

[nikkkom]

Such high doses that there are measured
(and in the torus is 2.2 Sv) can only nuclear fuel.

2.2 Sv/h is not really high enough for nuclear fuel. IIRC the bottom of Fukushima vent stack is 10 Sv/h - and that clearly can't be fuel, it's too far from the reactor building, so it must be fission products.

In Chernobyl, corium typical levels (e.g. "elephant foot") were in 10-100 Sv/h range.

 

[SteveElbows]

This news article seems to indicate uncertainty about what the TEPCO pdf reports call a vent pipe leak.

http://www.japantimes.co.jp/news/20...er-reactor-1-containment-vessel/#.UofgPJHXVhw

Another leak was confirmed just above the suppression chamber, which is a huge donut-shaped chamber connected to the containment vessel, and one of eight vent pipes.

The suppression chamber contains water and is used to reduce pressure inside the containment vessel through vent pipes.

Akira Ono, chief of the Fukushima No. 1 plant, said of the second leak that there is another pipe above the suppression chamber and the vent pipe, and it appears that the water is leaking from around that pipe.

But Ono said it is still unknown where exactly the leak is located, and that it is conceivable the water is coming from the containment vessel.

Still, “these are significant findings to help” find the precise locations of the leaks, he said.

 

[a.ua.]

I would not jump to that conclusion so fast. There was that vent tower bottom, with around 10Sv/h as I recall, far away from the containment...

the most realistic option

It looks like near the edge

http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_120627_02-e.pdf
-------------------------------------------------------------------

the change in the exposure rate indicates the source of a torus.
At 3 meters (10 feet) from the edge of the bubbler. (Tor)

http://www.tepco.co.jp/nu/fukushima-np/handouts/2013/images/handouts_130220_03-j.pdf

https://www.physicsforums.com/showpost.php?p=3978192&postcount=13510



http://ru.fotoalbum.eu/images1/200905/95064/273731/00000058.JPG
http://www-pub.iaea.org/iaeameetings/IEM4/30Jan/Suzuki_d.pdf

 

[a.ua.]

2.2 Sv/h is not really high enough for nuclear fuel. IIRC the bottom of Fukushima vent stack is 10 Sv/h - and that clearly can't be fuel, it's too far from the reactor building, so it must be fission products.

In Chernobyl, corium typical levels (e.g. "elephant foot") were in 10-100 Sv/h range.

It all depends on the time and distances,
at the moment: 2.6 years decay,
2-3 kg of nuclear fuel will give 2 Sv, at the distance of 1 meter, without shielding metal or concrete.

 

[etudiant]

the most realistic option
url]http://www-pub.iaea.org/iaeameetings/IEM4/30Jan/Suzuki_d.pdf[/url]

Thank you, a.ua., for an excellent set of very informative links.
I am honestly impressed by the Japanese effort, which is a far cry from the public media presentation of bumbling incompetence.

Two elements that jump out from the presentation:
1 The plan is to shorten the cooling loop, by sealing the reactor buildings and recycling their cooling water internally. Needs compact decontamination/heat exchange system to be developed and retrofitted, a tall order.
2 The SFP 3 is the next target, after the SFP 4 is cleared. That will require work in a much more messy environment. It seems a very bold step to me.

I can't help but think that the SFP4 cleanup and the work on the ground water are the low hanging fruit here. Anything beyond that will take real developments and come after the Tokyo Olympics, not before.

 

[Hiddencamper]

Agree entirely, severe accidents are like wars, even the simplest thing becomes very difficult.
The concern is that the regulators are missing the good in their effort to achieve perfection.

That said, it is just incomprehensible to me that a nuclear vent stack should be unfiltered. It may never be needed, hopefully, but it sure is much more useable with a filter than without.
Venting relatively safely should be another option for the operator, not a desperation necessity.

The stacks have filters on them though. They just aren't very effective without electricity...which gets us back to the original problem lol

 

[Rive]

the most realistic option

Still don't think so. Mind the assumed position of the core debris, and the strong shielding effect of the water during the measurements you linked!

I think it's simply from the contaminated water from the torus. That water is still from the first days of the accident, with Cs levels at the 10^6 range (or even higher).

To clarify this: I don't saying that there are *no* core debris in the torus. I'm just saying that the radiation levels are not sufficient to imply that there are.

1 The plan is to shorten the cooling loop, by sealing the reactor buildings and recycling their cooling water internally. Needs compact decontamination/heat exchange system to be developed and retrofitted, a tall order.
2 The SFP 3 is the next target, after the SFP 4 is cleared. That will require work in a much more messy environment. It seems a very bold step to me.

Neither of those goals are really difficult. They will get some experience soon with freezing, as they trying to seal the trenches, and actually they are working on decontaminating the top of U3. They should be able to 'cut down' the turbine buildings and set up an acceptable working conditions on top of U3 on planned order.

 

[etudiant]

The stacks have filters on them though. They just aren't very effective without electricity...which gets us back to the original problem lol

Thank you for this input!
Learn something surprising every day.
I clearly don't understand how these filters work, that they need to be powered.
Do they require blowers or electrostatic precipitators to perform properly?
Do you know if this requirement is also true for the Nordic installations that were discussed earlier?

 

[SteveElbows]

I think it's simply from the contaminated water from the torus. That water is still from the first days of the accident, with Cs levels at the 10^6 range (or even higher).

To clarify this: I don't saying that there are *no* core debris in the torus. I'm just saying that the radiation levels are not sufficient to imply that there are.

I believe we need a lot more detailed mappings of radiation levels at a wide range of locations in the reactor 1 torus room. Especially since there is quite a large disparity between the first set of 'probe dangled on a wire' radiation readings at different heights within the reactor 1 torus room, and the second one which was used to make the graphic posted earlier. The large difference in radiation levels in that torus room compared to the others is of interest, but I agree that we should not jump to conclusions. The lack of reactor data during key stages of reactor 1 meltdown does not help. Nor does the failure to locate water leakage points at the other reactors.

Certainly before getting too carried away it is important to compare the several Sv/hr readings from the torus room with the multiple tens of Sv/hr we've seen from, for example, the last survey of the area approaching reactor 2 pedestal. Personally I lack the knowledge to appreciate the full potential of water shielding in the torus and the torus room, that may be an important factor when trying to reach any tentative conclusions.

I think that public awareness and discussion of this stuff has, like so many other aspects of the disaster, not been helped by the failure of various official narratives to really join dots, even tentatively, between possible events that happened and some of the specific data we get. For example the high radiation level at certain locations within the shared reactor 1/2 stack and associated pipework was not met, as far as I know, with a concise narrative about the various possible explanations for this. Throw in a potential lack of public awareness between corium/fuel and various other forms of radioactive elements that found their way into various parts of the reactors, and the crude state of narrative from certain anti-nuclear agenda driven sources (e.g. reactor 3 plutonium fuel fixation), and I am rather underwhelmed by the level of clarity offered to those looking for easily consumable explanations. We know its a long, slow journey to get enough solid data about all manner of things, but in the meantime far more could have been done to understand what the various realistic possibilities are, and to point out when something is discovered that tends to rule stuff in or, more often so far it seems, out. There have been all manner of occasions where accumulated knowledge shared on this forum has had the potential to offer narratives and tentative conclusions that far exceed that offered pretty much anywhere else in public. A summary of where we are at so far in relation to many things could be constructed from it and may be useful, but the level of collaboration required may be tricky, or considered too tedious given that so many question marks remain and that a prize of stumbling on some important revelation does not seem to be on offer at this stage anymore than it was during early photo-gazing.

Neither of those goals are really difficult. They will get some experience soon with freezing, as they trying to seal the trenches, and actually they are working on decontaminating the top of U3. They should be able to 'cut down' the turbine buildings and set up an acceptable working conditions on top of U3 on planned order.

I am far less optimistic about that, it is far from trivial to get the radiation levels down to acceptable levels anywhere near reactor 3 building. I think there is plenty still to be revealed about specific sources of radiation in and around that building. Decontaminating the upper levels is clearly important, but the wider area seems to still have some notable sources of radiation that make dose rates for workers in the entire region of the reactor 3 building rather impractical. How much they can do via remote control is likely to remain important - So far they've done quite an impressive job of removing debris from the upper floors of reactor 3, but the full challenges of dealing with the reactor 3 pool are yet to receive enough detailed public discussion. Likewise the survival of reactor 2 building presents some challenges with gaining access to that pool, and reactor 1 schedule has been lengthened by the need to undo the initial work they did constructing an outer shell for that building. If there were not so much fuel in reactor 4 building, and there had not been such incense concerns about that fuel pool at the height of the disaster, I suspect there would have been more public focus and concern on dealing with these other pools.

 

[zapperzero]

You can all pontificate all you want about how you think it should work, but you need to understand the design of these plants, along with the regulatory design requirements, to understand where the challenges are in just saying that some passive thing can be installed that will magically solve all your problems post accident.

Holy run-on sentences, Batman! Of course you can't pre-solve every possible issue. This doesn't mean passive safety features aren't better than active ones.

No matter what, it will take significant efforts by those at the plant to cope with a beyond design basis accident, with or without a filter.

"No matter what"? That's a sweeping generalization if I ever saw one. With a passive, filtered vent, it will be easier, because it removes the dilemma in which the operators of Fukushima 1 found themselves on the fateful night... to vent and blanket the environs with radio-iodine, or to NOT vent and risk even worse contamination? In the event, the choice was was made for them...

 

[a.ua.]

a radiation unit 3
After removal of the crane girders
http://ru.fotoalbum.eu/images1/200905/95064/273731/00000055.JPG

http://www.meti.go.jp/earthquake/nuclear/pdf/131031/131031_01g.pdf

For[/PLAIN] the "fuel debris retrieval
Fuel debris situation analysis "
j

 

[Hiddencamper]

Holy run-on sentences, Batman! Of course you can't pre-solve every possible issue. This doesn't mean passive safety features aren't better than active ones.



"No matter what"? That's a sweeping generalization if I ever saw one. With a passive, filtered vent, it will be easier, because it removes the dilemma in which the operators of Fukushima 1 found themselves on the fateful night... to vent and blanket the environs with radio-iodine, or to NOT vent and risk even worse contamination? In the event, the choice was was made for them...

The "no matter what", is because, by DEFINITION, a beyond design basis accident is one where all permanently installed onsite plant equipment fails to perform safety functions to prevent a core damaging event.

Under that definition, of a beyond design basis accident, that means the ONLY actions that will be effective are manual actions.

Even with a passive filtered vent, Fukushima operators found themselves unable to open their vents due to rupture disks that failed. They also didnt have portable equipment, plans, procedures, or leadership to ensure scrubbing was performed. Even if they had a passive filter, the fact that the rupture disks at Fukushima failed at multiple units means that passive filters wouldn't have helped at all if they were there.

 

[etudiant]

The "no matter what", is because, by DEFINITION, a beyond design basis accident is one where all permanently installed onsite plant equipment fails to perform safety functions to prevent a core damaging event.

Under that definition, of a beyond design basis accident, that means the ONLY actions that will be effective are manual actions.

Even with a passive filtered vent, Fukushima operators found themselves unable to open their vents due to rupture disks that failed. They also didnt have portable equipment, plans, procedures, or leadership to ensure scrubbing was performed. Even if they had a passive filter, the fact that the rupture disks at Fukushima failed at multiple units means that passive filters wouldn't have helped at all if they were there.

Is this correct?
I was not aware that the Fukushima site had filtered vents, or am I misreading the post?
I had thought that they delayed venting because they were concerned about unfiltered emissions from the failing reactors and that by the time they wanted to vent, they no longer could because there was no power.
The failure of the rupture discs is unsurprising to anyone who has worked in the electronics industry, it is always the high priced chip that gets fried, not the sacrificial diode or such that was supposed to protect it. The rationales posted for why these discs failed in this instance seem a little tortured, but I've not seen the official explanation or analysis, if it has been released. One would think that the Nordic system operators would be quite concerned about this aspect.

Your central point that managing a 'beyond design basis' accident really requires a trained operator staff who have a framework and appropriate tools to keep the beast in check is critical.
Fukushima shows what happens when these are not adequately provided.

 

[Hiddencamper]

Is this correct?
I was not aware that the Fukushima site had filtered vents, or am I misreading the post?
I had thought that they delayed venting because they were concerned about unfiltered emissions from the failing reactors and that by the time they wanted to vent, they no longer could because there was no power.
The failure of the rupture discs is unsurprising to anyone who has worked in the electronics industry, it is always the high priced chip that gets fried, not the sacrificial diode or such that was supposed to protect it. The rationales posted for why these discs failed in this instance seem a little tortured, but I've not seen the official explanation or analysis, if it has been released. One would think that the Nordic system operators would be quite concerned about this aspect.

Your central point that managing a 'beyond design basis' accident really requires a trained operator staff who have a framework and appropriate tools to keep the beast in check is critical.
Fukushima shows what happens when these are not adequately provided.

I'm really referring to the standby gas treatment system, which is a combination of HEPA filters and charcoal beds. SBGT is supposed to maintain a vacuum in the secondary containment to filter any leaks through the primary, and also has a backup function to vent the primary containment. As SBGT is a charcoal based filtration system, its effectiveness relies upon the ability to remove moisture from effluents (as well as active power to open the valves, dampers, run blowers, and run heaters/dehumidification modules). Like I said, it is not a passive filter, and is not very effective compared to wet scrubbing or a large dry filter like those in Europe.

The safety logic in BWRs, when it sees an increase in effluents from the exhaust stack (depending on plant design, typically >10 mRem/hr from the secondary containment or >100 mRem/hr from primary containment exhaust), will automatically shut down all normal exhausting/ventillation systems and activate SBGT. All effluents are then routed through SBGT prior to exhaust to remove radioactive material. This is a standard feature in BWRs. This transfer to SBGT also automatically occurs if a LOCA signal is detected ( below the level 1 low water alarm setpoint or high drywell pressure). Again, all active logic, AC/DC power is required.

 

[Rive]

I am far less optimistic about that, it is far from trivial to get the radiation levels down to acceptable levels anywhere near reactor 3 building.

Definitely not trivial, but even if it'll take some years, it's not 'difficult'. They have everything they need for this task, so they *should* be able to start the manned work there within a year or two.

Should, would, will - that's a different matter, of course. We'll see.

IMHO the difficulties will start with removing the (fuel and other) debris from U3 pool. That'll be something different. The possibility that some fuel debris stuck to the FHM or construction material debris and gets to the surface with them...
Maybe they will cut it all to pieces underwater and cask it all.

If there were not so much fuel in reactor 4 building, and there had not been such incense concerns about that fuel pool at the height of the disaster, I suspect there would have been more public focus and concern on dealing with these other pools.

I'm actually using this as a kind of reliability check. If a source is more concerned about U4 pool/building than U3 pool/building, then it's most likely missed some important points in this story.

All the stuff you wrote about the public awareness and discussion is correct. Well said.

Ps.: some sources are really trying to keep the story going the same speed as in the start, whatever it costs. But it's different now, with much less drama, so these kind of efforts requires much and much 'inventions'.

 

[SteveElbows]

All this talk of rupture disc problems runs the risk of downplaying the other issues that delayed venting, and also runs the risk of making it sound like venting failed at all the reactors, as opposed to the apparent reality that it was mostly reactor 2 where the failure to vent story was allowed to play out in full. Thats certainly the only reactor where they seem rather unsure as to whether the rupture disc ever ruptured, and where no torus-scrubbed venting is thought to have taken place.

Having said that, the delays to venting at reactors 1 and 3 obviously caused additional delays in pumping water into those reactors.

Certainly to my mind the problem with the rupture disc-based system seems to be due to the fact that the initial design considerations for this system were heavily focussed on the idea of containment failures due to over pressurisation. Only in later theoretical loss of coolant accident analysis papers did other modes of containment failure, such as temperature-related failure of seals and penetrations, get due attention. Really the entire system seems inappropriate not just for venting under situations where other containment failures kept containment pressure below the level necessary to rupture the discs, but also for situations where there is a desperate need to reduce containment pressure much earlier on in order to ensure that the RPV itself can be depressurised via SRV's in order to allow pumping in of water.

 

[rmattila]

In the Fukushima design, the rupture disk is in series with the two closed valves that must first be opened in order for the pressure to work on the disk. If the valves are not opened in time, the containment will develop a leak and the pressure might never reach the disk burst pressure again.

This is not the way to design it. The valves in the rupture disk lines should be kept open, and there should be a manual by-pass to the rupture disks.

Back in the 80's, it took about 3 years after the Chernobyl accident to have the filtered vents designed and installed in European BWRs. Of course, a prerequisite for that was that everybody agreed on their necessity and no time was wasted on arguing whether or not they should be built.

 

[nikkkom]

It all depends on the time and distances,
at the moment: 2.6 years decay,
2-3 kg of nuclear fuel will give 2 Sv, at the distance of 1 meter, without shielding metal or concrete.

I have doubts about your numbers.
Almost all radiation from spent fuel comes from fission products and minor actinides. Let's check how much those emit.
IIRC French reprocess the fuel after about 5 years of cooldown.
This document:

http://www.wmsym.org/archives/2003/pdfs/194.pdf

says that at French reprocessing plant (best in the world) after reprocessing, vitrification of the fission products and minor actinides, and pouring of the resulting glass into 0.5 cm thick walled stainless canister, dose rate on contact with canister surface is 14000 Gy/h.

Granted, it is 500 kg of material, not 3. OTOH, with canister diameter of 43 cm there is substantial self-shielding, and canister's wall shields all betas and most of low-energy secondary gammas, while in your situation ("without shielding metal or concrete") there is no such effect.

And it is a contact reading, not 1 meter reading.

Still, 14000 Gy/h is vastly higher than measly 2 Sv/h (~=2 Gy/h) you provided. I think you are wrong by at least an order of magnitude.

 

[zapperzero]

Having said that, the delays to venting at reactors 1 and 3 obviously caused additional delays in pumping water into those reactors.

Iirc reactor 3 depressurized all by itself? The plots are somewhere in the mega-thread.

Certainly to my mind the problem with the rupture disc-based system seems to be due to the fact that the initial design considerations for this system were heavily focussed on the idea of containment failures due to over pressurisation. Only in later theoretical loss of coolant accident analysis papers did other modes of containment failure, such as temperature-related failure of seals and penetrations, get due attention. Really the entire system seems inappropriate not just for venting under situations where other containment failures kept containment pressure below the level necessary to rupture the discs, but also for situations where there is a desperate need to reduce containment pressure much earlier on in order to ensure that the RPV itself can be depressurised via SRV's in order to allow pumping in of water.

Obviously such situations can be handled by other means. For example you could have a vent path that is under operator control and feeds into the same filter. I don't see how you can argue against the necessity of a passive venting system by pointing out that there are failure modes it doesn't address. Shall we not install curtain (lateral) airbags in cars, because there are also lots of head-on collisions?

 

[a.ua.]

nikkkom

I have doubts about your numbers.

Should be considered not in the mass, but the number of terabecquerels.
For exact calculation is also required to know the power of fuel burn per day, amount of fuel enrichment.
Just keep in mind that the radiation power as a function of the distance does not vary linearly.
is the square of the distance.

The figures I quoted were made by an experienced expert in dosimetry.
not by me

 

[SteveElbows]

Iirc reactor 3 depressurized all by itself? The plots are somewhere in the mega-thread.

I'd need to go back and check. All the same I think they think reactor 3 venting did eventually happen via the stack. Obviously another problem they had was a lack of stack instrument functionality due to power failure, so they were using crude methods such as checking the webcam for evidence of emissions from the stack.

Obviously such situations can be handled by other means. For example you could have a vent path that is under operator control and feeds into the same filter. I don't see how you can argue against the necessity of a passive venting system by pointing out that there are failure modes it doesn't address. Shall we not install curtain (lateral) airbags in cars, because there are also lots of head-on collisions?

I was not arguing against various types of venting, just pointing out some of the flaws. In an ideal world the best solution would really be to close down all old reactors that, at a minimum, have the first type of containment design which has long been recognised as being inadequate, but obviously that isn't happening.

 

[zapperzero]

I'd need to go back and check. All the same I think they think reactor 3 venting did eventually happen via the stack. Obviously another problem they had was a lack of stack instrument functionality due to power failure, so they were using crude methods such as checking the webcam for evidence of emissions from the stack.

You are right:
http://www.tepco.co.jp/en/press/corp-com/release/11031310-e.html

Unit 3:[...]
In order to fully secure safety, we operated the vent valve to reduce the
pressure of the reactor containment vessels (partial release of air
containing radioactive materials) and completed the procedure at 8:41AM,
Mar 13 (successfully completed at 09:20AM, Mar 13. After that, we began
injecting water containing boric acid that absorbs neutron into the reactor
by the fire pump from 09:25AM, Mar 13.
Taking account of the situation that the water level within the pressure
vessel did not rise for a long time and the radiation dose is increasing,
we cannot exclude the possibility that the same situation occurred at Unit
1 on Mar 12 will occur. We are considering the countermeasure to prevent
that.

At the same time, TEPCO only has plant parameters starting from June 2011 on their website.

EDIT: found a semi-useful plot on page 16 of this report:
http://www.nsr.go.jp/english/data/dai-ichi_NPS_handouts2.pdf
the red line is the actual pressure, blue line is the imagination of report authors, augmented with some software, so you can ignore it. There is a small unexplained pressure dip near the start of the plot, but other than that, it seems that it was indeed depressurized through operator actions.

 

[LabratSR]

At the same time, TEPCO only has plant parameters starting from June 2011 on their website.

Try here

https://fdada.info/EDIT: Found this there

https://fdada.info/docdata/accident_analysis/ES-Unit3-01.pdf

 

[turi]

You are right:
EDIT: found a semi-useful plot on page 16 of this report:
http://www.nsr.go.jp/english/data/dai-ichi_NPS_handouts2.pdf
the red line is the actual pressure, blue line is the imagination of report authors, augmented with some software, so you can ignore it. There is a small unexplained pressure dip near the start of the plot, but other than that, it seems that it was indeed depressurized through operator actions.

Aren't the blue and red dots the measured values and the red and blue lines the (software aided) approximations?

 

[Hiddencamper]

Iirc reactor 3 depressurized all by itself? The plots are somewhere in the mega-thread.



Obviously such situations can be handled by other means. For example you could have a vent path that is under operator control and feeds into the same filter. I don't see how you can argue against the necessity of a passive venting system by pointing out that there are failure modes it doesn't address. Shall we not install curtain (lateral) airbags in cars, because there are also lots of head-on collisions?

For a reactor to depressurize itself after its SRV accumulators have depleted means that the vessel was breached.

The SRV (safety relief valves) in GE BWRs are designed to open once or twice against 3/4 of containment design pressure. The first lift is assumed to be of all valves in relief mode, due to a load reject or MSIV fast closure. The second lift is ONLY the ADS (automatic depressurization system) valves, which then stay open until corespray comes in service to hold pressure low. About 1/2 of the valves in a BWR are ADS valves. (This is likely a little different for the unit 1 BWR, as some BWRs use EMRVs and ADS valves, but the overall concept is similar).

The accumulators are typically either 20-25 gallon (for normal SRVs) or 55 gallon (for SRVs that also utilize the ADS feature).

These air accumulators, against no containment pressure, only have a handful of lifts each. Many plants have backup air bottles, which can be used to refill and get up to 100 lifts out of the SRVs in their relief move. With no electrical power, there is no way to refill the SRV accumulators through normal means. The air lines into containment automatically isolate on a loss of power, a level 1 water level (about 2 feet above the fuel), high drywell pressure, or a loss of air pressure (they use air pressure pilot valves to hold them open).

 

[Hiddencamper]

All this talk of rupture disc problems runs the risk of downplaying the other issues that delayed venting, and also runs the risk of making it sound like venting failed at all the reactors, as opposed to the apparent reality that it was mostly reactor 2 where the failure to vent story was allowed to play out in full. Thats certainly the only reactor where they seem rather unsure as to whether the rupture disc ever ruptured, and where no torus-scrubbed venting is thought to have taken place.

Having said that, the delays to venting at reactors 1 and 3 obviously caused additional delays in pumping water into those reactors.

Certainly to my mind the problem with the rupture disc-based system seems to be due to the fact that the initial design considerations for this system were heavily focussed on the idea of containment failures due to over pressurisation. Only in later theoretical loss of coolant accident analysis papers did other modes of containment failure, such as temperature-related failure of seals and penetrations, get due attention. Really the entire system seems inappropriate not just for venting under situations where other containment failures kept containment pressure below the level necessary to rupture the discs, but also for situations where there is a desperate need to reduce containment pressure much earlier on in order to ensure that the RPV itself can be depressurised via SRV's in order to allow pumping in of water.

Need to also remember the SRVs require DC power and pressurized air to operate in their relief mode, regardless of containment pressure. There were cases at units 2 and 3 of SRVs drifting closed or failing to open, between loss of DC power or pressurized air.

Typically, venting containment during a casualty is to help you flood the containment more than flood the core.

 

[nikkkom]

For a reactor to depressurize itself after its SRV accumulators have depleted means that the vessel was breached.

The SRV (safety relief valves) in GE BWRs are designed to open once or twice

For me, an outsider, this is a shocking revelation. A revief valve which is *not* designed for at least hundreds of actuations?

The accumulators are typically either 20-25 gallon (for normal SRVs) or 55 gallon (for SRVs that also utilize the ADS feature).

These air accumulators, against no containment pressure, only have a handful of lifts each.

A relief valve which requires *consumables* to work??

 

[westfield]

I'm really referring to the standby gas treatment system, which is a combination of HEPA filters and charcoal beds. <snip>.

I'm confused. Why are you referring to the SGTS filters?

If we are talking venting via the "hardened vent" systems at fukushima I don't understand why SGTS filters are involved in the conversation. The "hardened vent" systems don't run through the SGTS filters at fukushima daiichi, they go straight to the stacks, unfiltered apart from the scrubbing from the torus water in the case of the SC "hardened vent" path or no filtering whatsoever in the case of the drywell "hardened vent" path.

The inadequacy of the SGTS in an emergency venting scenario is precisely the reason why "hardened vent" systems were retrofitted to these types of plant. It was realized early on that the ducting of the SGTS systems in these types of plant would be highly likely to fail under an emergency venting scenario and would fill secondary containment with steam and combustible gases.

 

[Hiddencamper]

For me, an outsider, this is a shocking revelation. A revief valve which is *not* designed for at least hundreds of actuations?



A relief valve which requires *consumables* to work??

They are dual function valves. The relief function is used for manual or automatic control to open the valve, and uses air pressure and DC power. You can control pressure in almost any range with the relief mode, and can blow the reactor down with these. ADS (automatic depressurization system) works by using the relief mode solenoids to lift the valve and blow down the reactor. The relief mode solenoids is actuated by either logic systems, which respond to overpressure, or manually by throwing a control switch to energize the valve solenoid directly.

The safety mode is spring loaded and typically actuates about 100 PSIG above the relief mode. The spring mode is completely passive, but can only maintain pressure around its setpoint. It only reduces reactor pressure by no more than 100 PSIG from it's lift setpoint.

For my BWR, my lowest SRV has a logic that lifts it at 1103 PSIG, and reseats it at 926 PSIG. This is the automatic relief mode, and the plant's control systems will maintain my reactor pressure between those two points for me while I manage other, more important parts of the accident (like starting or overriding ECCS, getting feedwater back, making the turbine safe, getting aux steam running, or restarting feedwater). Now if my relief mode fails, or I lose DC power/air, this same valve's safety mode (spring) lifts around 1165 PSIG, and seats when you no longer have adequate force to overcome spring pressure. It will lift to maintain my pressure between 1165 and about 1065. It's more harsh on the equipment, because it results in more valve lifts, and increases the risk of a valve getting stuck open, but it will keep the pressure vessel safe.

If I want to or need to blow down the reactor, I need air and DC power, but if I don't have those, the safety mode will protect the vessel from exceeding its ASME code limit (typically in the 1300-1400 PSIG range), even with a full power ATWS.

The reason this is considered acceptable, is because for the design basis accident, you assume you only need two valve lifts. The first lift is on the initial load reject/MSIV closure, all valves lift once in the relief mode. The non ADS valves are assumed to have utilized their entire air inventory. Then the loss of coolant/loss of feedwater/loss of high pressure injection accident would eventually pick up the logic to activate ADS, and all the SRVs that have an ADS mode would use their remaining air inventory to bring pressure down to minimum. As pressure drops, the core spray system would actuate and spray the core. Core spray actually reduces pressure greatly, and as long as core spray is running, the pressure remains low enough for all the low pressure ECCS systems to inject to the core.

tl;dr - MSIV fast closure + loss of feedwater = first SRV lift. failure of high pressure ECCS (single failure required for design basis accident) leads to low-low-low alarm level 1 water level, which automatically performs an ADS blowdown. When pressure gets low enough, the core sprayers start up to keep pressure low enough for low pressure coolant injection to run. Instrument air to refill the valves is non-safety, and is not available during the accident.

If you read GE's design basis documents, they assume 1 lift on non-ADS SRVs, 2 lifts on ADS-SRVs, then their requirements for maintaining safe shutdown are 1 core spray and 1 coolant injection ECCS to maintain the core safe once its been depressurized. After you've blown down, the SRVs no longer are required to perform a safety function.

Obviously, you can get several lifts of an SRV without refilling the air, but it will depleate over time (whether or not you use it), and it obviously depletes much faster if you use it. The ADS backup air bottles allow you to refill those accumulators for more lifts, or to blow the reactor down, if you didnt need to blowdown early in the event, but now conditions have changed and you do need it

 

[Hiddencamper]

I'm confused. Why are you referring to the SGTS filters?

If we are talking venting via the "hardened vent" systems at fukushima I don't understand why SGTS filters are involved in the conversation. The "hardened vent" systems don't run through the SGTS filters at fukushima daiichi, they go straight to the stacks, unfiltered apart from the scrubbing from the torus water in the case of the SC "hardened vent" path or no filtering whatsoever in the case of the drywell "hardened vent" path.

The inadequacy of the SGTS in an emergency venting scenario is precisely the reason why "hardened vent" systems were retrofitted to these types of plant. It was realized early on that the ducting of the SGTS systems in these types of plant would be highly likely to fail under an emergency venting scenario and would fill secondary containment with steam and combustible gases.

I'm trying to discuss what the plant ALREADY has installed to meet its design basis requirements. SGTS is not inadequate for design basis accidents, its only inadequate in an extended total loss of power with damage to your permanently installed plant systems. This means a filtered vent is not required to maintain the public safe during design basis accidents. In no case during a DBA would you need a passive filtered vent to make the plant safe. The installation of a passive filtered vent does not help you at all for any design accident, and provides very little if any net benefit. From an engineering/reactor designer perspective its more of a warm fuzzy, because you already have nuclear safety grade equipment which performs that function. (Now if we were designing a new plant, you sure as **** can bet that I would design a passive filter in, but talking about existing plants, you already have something for that)

Now for beyond design basis accidents, you have to assume going into the BDBA that all your permenently installed equipment failed. This makes sense, because in order to get into a BDBA, you had to lose all your onsite equipment. So in that case, can you actually honestly believe that the passive filtered vent will function when all your other safety grade equipment failed? The most likely strategy to succeed (in my opinion) is one which utilizes off-site portable equipment to spray and wet scrubbing. I don't see how you can, in engineering space, claim a passive filter along with all the valves, discs, etc, is any more likely to be functional than an active system. And with the EPRI study on decontamination factors for wet spray/scrubbing and the suppression pool being available (the suppression pool is a major scrubbing source, and on its own achieves the appropriate level of DF), you can meet the same quantitative goal with > 1000 DF using portable equipment which you can guarantee will function post accident.

That's my view on it as a plant design engineer.

 

[zapperzero]

This means a filtered vent is not required to maintain the public safe during design basis accidents. In no case during a DBA would you need a passive filtered vent to make the plant safe.

Circular reasoning much?

The installation of a passive filtered vent does not help you at all for any design accident, and provides very little if any net benefit.

The historical record shows that so-called "beyond design basis" accidents do happen to NPPs, with alarming frequency even (~1% of population). The assumptions built in the design basis need a bit of challenging, iow.

Now for beyond design basis accidents, you have to assume going into the BDBA that all your permenently installed equipment failed.

Nonsense. Lots of equipment was and still is functional inside Fukushima 1. Useful stuff, even - isolation condensers, vent stacks, SGTSs for at least units 1 and 3 and many other such things. And yet, as soon as the tsunami swept through, the plant was in a BDBA (or so Tepco would have us believe).

So in that case, can you actually honestly believe that the passive filtered vent will function when all your other safety grade equipment failed?

It is not a matter of belief. You can ensure the passive filtered vent will only fail in harsher conditions, by designing it properly. Fewer moving parts, simpler control logic...

The most likely strategy to succeed (in my opinion) is one which utilizes off-site portable equipment to spray and wet scrubbing.

You are assuming you can get it onsite in time. Again, recent history shows that's not always the case - causes can be as trivial as a padlocked gate, or a worker who is not able to go in an unknown radiation field.

I don't see how you can, in engineering space, claim a passive filter along with all the valves, discs, etc, is any more likely to be functional than an active system.

I can assign some non-zero probability to the event of the active system losing power and/or control...

And with the EPRI study on decontamination factors for wet spray/scrubbing and the suppression pool being available (the suppression pool is a major scrubbing source, and on its own achieves the appropriate level of DF),

What if the pool water level drops too low for effective scrubbing for some reason? Say, I dunno, too high of a temperature and pressure?

you can meet the same quantitative goal with > 1000 DF using portable equipment which you can guarantee will function post accident.

Err... what? Are you seriously suggesting that having to bring a filter from somewhere else is just as reliable a strategy as already having it onsite?

That's my view on it as a plant design engineer.

It is a bit troubling.

 

[Hiddencamper]

Circular reasoning much?


The historical record shows that so-called "beyond design basis" accidents do happen to NPPs, with alarming frequency even (~1% of population). The assumptions built in the design basis need a bit of challenging, iow.


Nonsense. Lots of equipment was and still is functional inside Fukushima 1. Useful stuff, even - isolation condensers, vent stacks, SGTSs for at least units 1 and 3 and many other such things. And yet, as soon as the tsunami swept through, the plant was in a BDBA (or so Tepco would have us believe).


It is not a matter of belief. You can ensure the passive filtered vent will only fail in harsher conditions, by designing it properly. Fewer moving parts, simpler control logic...


You are assuming you can get it onsite in time. Again, recent history shows that's not always the case - causes can be as trivial as a padlocked gate, or a worker who is not able to go in an unknown radiation field.


I can assign some non-zero probability to the event of the active system losing power and/or control...


What if the pool water level drops too low for effective scrubbing for some reason? Say, I dunno, too high of a temperature and pressure?


Err... what? Are you seriously suggesting that having to bring a filter from somewhere else is just as reliable a strategy as already having it onsite?


It is a bit troubling.

The IC was NOT functional (if it was, unit 1 would not have failed in a few hours, as the IC had some ridiculous amount of cooling available to it). The IC would have provided adequate core cooling to unit 1, and could have been easily made up with a simple fire pump truck. Remember, the IC has to be cycled on and off under normal conditions to prevent exceeding the 100 degree F per hour cooldown rate (I think its 40 deg C/hr for Japan) for the vessel. It was cycled off when power was lost. The internal isolation valves are DC valves, and the external valves are AC valves. The operators typically cycle the IC discharge outboard isolation open and closed to turn the IC on and off. On the loss of power, the IC was already in the off position, and it is believed that some of the other valves may have went closed under an invalid isolation signal during the flooding of the safety related MCCs. The operators erroneously determined the IC was functioning when it wasnt. They possibly could have sent an operator to the IC outboard isolation valves, manually opened them, and got cooling, but they didnt even think to try (unfortunately they were not well trained on the system, and nobody in the control room at the time had any experience using it). So yes, it may have been capable of helping, but it was not functional at the time of the event, and may not have been available even.

SGTS was not functioning at Fukushima (need power to open the dampers, need power for the pre-heaters and dehumidification). First, the SGTS has fail close dampers on the containment isolation side. these dampers are controlled with fail close hydramotors. Hydramotors are throttleable or 2 position hydraulic actuating units for positioning valves and dampers. When a hydramotor loses power, its relief solenoid loses power, which drains pressure from the accumulator, and causes the valve or damper to fail to a specified state on loss of power. SGTS containment isolation valves fails closed on loss of power.

The vent stack is not an active component.

The rupture disks failed to break, those are passive components (which goes to show that you cannot count on your on-site equipment)

As for the suppression pool level, in a normal accident you have RHR to remove heat from it. In a beyond design basis accident, you can lose level if the suppression pool itself (torus) breaks. You can deal with this by flooding the basement using portable or fire pumps, which, while you lose containment capability, you still have wet scrubbing capability. A passive vent wouldn't help you in this case as your pressure boundary broke. As for temp/press, remember that pool is an enclosed system. Inventory has no place to go while the system is sealed. It's not going to just disappear for Mark I/II containments (for Mark III containments, it can lower due to the very large volume of the containment. Mark III plant designs utilize passive gravity fed suppression pool makeup systems to deal with that, which will automatically dump when pool level drops about 4 feet, to ensure the drywell vents are adequately covered). The only time level should lower is when you are venting the wetwell, which does require a hardened vent, but the effluents have already been scrubbed by the pool, and by venting the containment you can now make up the pool to maintain your scrubbing.

As for the 'circular reasoning'. I don't see circular. During DBAs (things that are IN the design basis), your active filtering is all that you need. You don't NEED a passive vent. That's not circular at all, its saying active is already installed and works, passive could also work but you don't need it. Passive filters only have added benefits for beyond design accidents, which I argue you might not even have them because whatever nasty accident took out your active systems could have damaged your passive filter as well.

 

[zapperzero]

The IC was NOT functional

http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_111122_03-e.pdf
says otherwise. Both trains available and functioning, but not at full capacity

The operators erroneously determined the IC was functioning when it wasnt.

According to the document above, the operation was confirmed by observing steam coming out of the appropriate place.

They possibly could have sent an operator to the IC outboard isolation valves, manually opened them, and got cooling, but they didnt even think to try.

" At 21:30. the operator conducted open op
eration of valve 3A and confirmed
generation of steam. "
(from the same cited document)

SGTS was not functioning at Fukushima (need power to open the dampers, need power for the pre-heaters and dehumidification).

I... what? The point I was making was that it was not damaged in any way - yet did not get used in the event.

The rupture disks failed to break

What rupture disks?

As for the suppression pool level, in a normal accident you have RHR to remove heat from it. In a beyond design basis accident, you can lose level if the suppression pool itself (torus) breaks. You can deal with this by flooding the basement using portable or fire pumps, which, while you lose containment capability, you still have wet scrubbing capability. A passive vent wouldn't help you in this case as your pressure boundary broke. As for temp/press, remember that pool is an enclosed system. Inventory has no place to go while the system is sealed. It's not going to just disappear for Mark I/II containments (for Mark III containments, it can lower due to the very large volume of the containment. Mark III plant designs utilize passive gravity fed suppression pool makeup systems to deal with that, which will automatically dump when pool level drops about 4 feet, to ensure the drywell vents are adequately covered). The only time level should lower is when you are venting the wetwell, which does require a hardened vent, but the effluents have already been scrubbed by the pool, and by venting the containment you can now make up the pool to maintain your scrubbing.

You don't like temp/pressure? Fine. Let's say an earthquake damaged a steam downcomer, so that there is now a big crack in it, above the water level? Now you can't scrub your steam, although there is plenty of water.
I have a lot of doubt about your claim that the wetwell provides sufficient scrubbing, too. I seem to remember dramatic spikes in readings of the counters at plant boundary, corresponding to venting operations.
The operators of the plant were not convinced either, as I recall there was much wringing of hands before venting was even attempted, as there was explicit concern at TEPCO over the pace/effectiveness of the evacuation effort. Venting was delayed too much, actually.

As for the 'circular reasoning'. I don't see circular. During DBAs (things that are IN the design basis), your active filtering is all that you need. You don't NEED a passive vent. That's not circular at all, its saying active is already installed and works, passive could also work but you don't need it. Passive filters only have added benefits for beyond design accidents, which I argue you might not even have them because whatever nasty accident took out your active systems could have damaged your passive filter as well.

You are basing this belief on what, exactly?