Japan Earthquake: Nuclear Plants at Fukushima Daiichi

[Cire]

...If the fuel melted and the RPV breached early on, as they have admitted in #1, then yes the temps and pressures they've been issuing for weeks do seem pretty suspect, don't they?

I don't believe TEPCO has stated the RPV has failed in unit 1. They had a core melt event, that's not the same as the RPV being breached. Three mile island showed us how robust the RPV can be in a fuel melt event.

If the bottom of the RPV melted out you wouldn't be getting temperature readings from sensors at the bottom of the RPV.

 

[tsutsuji]

I did see those from back in march, I believe from satellite and I should have clarified a bit.

In case you want to see the whole series of helicopter borne thermal images released by the ministry of defense, they are available at http://www.mod.go.jp/j/approach/defense/saigai/tohokuoki/temp.html

 

[Bodge]

I don't believe TEPCO has stated the RPV has failed in unit 1. They had a core melt event, that's not the same as the RPV being breached. Three mile island showed us how robust the RPV can be in a fuel melt event.

If the bottom of the RPV melted out you wouldn't be getting temperature readings from sensors at the bottom of the RPV.

So if the majority of the core remains in the RPV, are we still at risk of catastrophic explosions at 1-3 ?

 

[razzz]

I don't see a drawing in there labeled "8-1" but it sounds like you're talking about page 18. The drawing that shows vents directly from the drywell to the refueling floor originated with the Areva presentation that came out weeks ago. See page 20

http://www.fairewinds.com/sites/default/files/AREVA%20Fukushima.pdf

The AREVA presentation has gotten a lot of attention here and elsewhere (mostly elsewhere) but that drawing continually escapes criticism for some reason.

I double checked, it is still page 20 of 30 as my Foxit .pdf Reader produces it. Labeled 'Figure 8-1 Reactor Well Seals'
I take it, above the drywell cap, air is able to move freely from the vents shown.

The plant lacked a way to deliver coolant and the SFPs blew up and the reactor meltdowns are just collateral damage makes no sense esp. concerning Unit 4.

http://tec-sim.de/images/stories/fusfpfail.pdf"

 

[Gary7]

Just regarding the USS Ronald Reagan, I looked for information regarding the ballistic radiation readings and subsequent food contamination and three-day emergency, but I could not find any. There was an early report of radioactivity in the drinking water, which was later shown to be a false positive. There is an excellent AP article that describes events on the Ronald Reagan here.

http://www.1310news.com/news/world/article/202681--uss-ronald-reagan-fought-contamination-amid-effort-to-aid-disaster-stricken-japan?ref=topic&name=index.php&title=

More details here

http://waronterrornews.typepad.com/home/2011/03/japan-uss-ronald-reagan-update.html

Furthermore, all of this transpired on the 13th of March (Sunday). If I recall correctly, the explosion at Unit #3 occurred on Monday the 14th. So the radiation that the USS Ronald Reagan encountered had to have come from Unit #1.

 

[SteveElbows]

It sounds like we may learn how much of the steam at reactor 2 is from fuel pool as opposed to suppression chamber in a few weeks:

http://www3.nhk.or.jp/daily/english/19_04.html

TEPCO says it is laying power cables for a cooling system for Number 2 reactor's spent fuel pool. A heat exchanger will be brought into the facility early next week to start operating the cooling system by the end of this month.

Workers entered the Number 2 reactor building on Wednesday for the first time since a hydrogen explosion on March 15th. They tried to check radiation levels but left the building after 14 minutes because it was filled with steam, making further work impossible.

The utility says the vapor appears to be coming from the damaged suppression chamber as well as from the fuel pool itself.

Senior TEPCO official Junichi Matsumoto says he believes cooling the spent fuel pool will help reduce steam inside the reactor.

 

[pdObq]

Section 3 has documents which relate to control rods and some other stuff mixed in.
For example this is the file for reactor 1:
http://www.tepco.co.jp/nu/fukushima-np/plant-data/f1_3_Keihou1.pdf

[...]

Also what looks like a 15:00 visualisation of the rods all showing the -99 value. The -99 one is on nearly the last page of the document, and is somewhat obscured by being printed in the same place on the same page as another type of data. I expect its nothing/normal but I just want to be more certain.

On some pages it says "-99 = Missing control rod position". Don't know how to interpret that.

 

[TCups]

I double checked, it is still page 20 of 30 as my Foxit .pdf Reader produces it. Labeled 'Figure 8-1 Reactor Well Seals'
I take it, above the drywell cap, air is able to move freely from the vents shown.

The plant lacked a way to deliver coolant and the SFPs blew up and the reactor meltdowns are just collateral damage makes no sense esp. concerning Unit 4.

http://tec-sim.de/images/stories/fusfpfail.pdf"

The "Presentation" content regarding siphoning from the spent fuel pools seems like new and potentially important information. I don't think this topic or possibility has been previously considered or discussed in this thread, has it?

http://tec-sim.de/images/stories/fusfpfail.pdf

 

[pdObq]

I seem to remember having read in one of the documents posted in this thread that one of the cooling systems operates, in automatic mode, between a couple setpoints (water levels in the RPV, iirc).

Yes, just for those too lazy to take a look into those documents , here are a few graphs attached from http://www.ornl.gov/info/reports/1981/3445600211884.pdf (p. 38ff (p. 50ff), see also post #7747), that show what happens when everything works as it should (it's a computer simulation, just in case someone is wondering).

One can see nicely how the RCIC cycles on and off keeping the water level between the upper and lower level, and also how safety relief valves open above a certain pressure and close again, thus cycling the RPV pressure.

[Since apparently one can upload only up to 3 attachments, it's a big collage. It was resized when uploading to forum max size. Oh well... I guess it's still readable.]

 

[pdObq]

And two weeks later they found them dead of internal bleeding in the turbine basement. I thought this an interesting cause of death, because they could have said they died of blast damage from the explosion, and no one would have given it a second thought, however internal bleeding likely came from some other cause.

Why? Isn't internal bleeding exactly what one would expect from a strong shockwave running through a poor fellow?

See, e.g. http://en.wikipedia.org/wiki/Blast_injury (bolding by me):

In general, primary blast injuries are characterized by the absence of external injuries; thus internal injuries are frequently unrecognized and their severity underestimated. [...] Blast lung refers to severe pulmonary contusion, bleeding or swelling with damage to alveoli and blood vessels, or a combination of these. It is the most common cause of death among people who initially survive an explosion.​

See also here, http://www.bt.cdc.gov/masscasualties/blastessentials.asp , under primary blast injuries:

Injury from over-pressurization force (blast wave) impacting the body surface: TM rupture, pulmonary damage and air embolization, hollow viscus injury.​

I am not really familiar with all that "medical slang" but that sounds a lot like "internal bleeding" would summarize it reasonably well for medical laymen like me.

 

[Cire]

So if the majority of the core remains in the RPV, are we still at risk of catastrophic explosions at 1-3 ?

I don't believe so. The core geometry is mangled and the fuel is sitting in borated water which has poisoned the fuel and reactor.

I suspect the largest problem now is that of washing out the core. If they can capture the outflow and process out the contaminants, cool it and pump it back in then the situation will be much more stable..

With the 16+ pressure relief values on the RPV I don't believe its possible to over-pressurize it.

With water in the bottom of it or the dry well flooded I don't see how you can melt through it either.

 

[pdObq]

You know what I don't understand? TEPCO is pumping water in, using existing piping. If they can do that, they can also push a flexible borescope in, without altering their setup or opening additional valves or anything. Why aren't they?

They might not like what they find or they might not want us to know.
Or management really is FUBARED and radiation induced cognitive impairment is taking its toll.

Come on guys, what do you expect from people who want to spray water into the SFPs with riot police trucks, who don't accept robots flown in from Europe specifically designed for nuclear accidents, and instead use two bomb disposal robots with a radiation meter taped to one of them.

 

[pdObq]

A few questions regarding the temperature sensors to figure out how trustable the temperature readings still are.

1. Does anyone know what kind of temperature sensors they are using? (Thermocouple, resistive, semiconductor, ...)
2. How is the sensor signal read out from the sensor to the control room? Are there amplifiers, AD converters, signal conditioning electronics and stuff like that located in the drywell close to the sensors?
3. How are the sensors mounted at their various locations? (Thermally conductive glue?)
   Is it conceivable that they can detach due to heat and humidity?

Concerning unit 1, if the sensors were just floating in the air, would it make sense that steam from the steamy atmosphere in there condenses on them and keeps them at roughly 100 degC ?

 

[mikefj40]

Just picked this up on the Twitter... Purportedly new images of tsunami flooding from TEPCO http://www.tepco.co.jp/tepconews/pressroom/110311/index-j.html

TEPCO slideshow: http://cid-0b14c7ab35e39ebd.photos.live.com/play.aspx/%5ENFNPP1%20Tsunami%20Flooding

Forgive me if these have already been posted. They're new to me.

 

[mikefj40]

They're new to me.

Most of them, not all...

 

[TCups]

Just picked this up on the Twitter... Purportedly new images of tsunami flooding from TEPCO http://www.tepco.co.jp/tepconews/pressroom/110311/index-j.html

TEPCO slideshow: http://cid-0b14c7ab35e39ebd.photos.live.com/play.aspx/%5ENFNPP1%20Tsunami%20Flooding

Forgive me if these have already been posted. They're new to me.

The interesting thing, to me, is the perspective. It is easy to underestimate the size of the wave and the amount of water inundating the NPP. Look at the pipes going to the vent towers. They are big enough to drive a SUV through! I didn't appreciate the size and weight of the pipe sections before.

 

[mikefj40]

I didn't appreciate the size and weight of the pipe sections before.

I'd conclude that the earthquake and tsunami didn't break the connection to the stacks. Venting was observed on the 12th and 13th. Post #7684.

The highly kinetic environment caused by the explosions broke the stack ventilation.

I have to add AFAIK and FWIW...

 

[NUCENG]

OK the link to the japanese page which links to all the pdf's is this one, I haven't noticed whether its available in english yet:

http://www.tepco.co.jp/nu/fukushima-np/index10-j.html

Some of it is presented in an infuriating way, and some of the files are rather large.

Section 3 has documents which relate to control rods and some other stuff mixed in.

For example this is the file for reactor 1:

http://www.tepco.co.jp/nu/fukushima-np/plant-data/f1_3_Keihou1.pdf

It starts with a lot of printed logs that get very busy when the earthquake hits. Later in the document there are some tables of data that are presented in an annoying way. Finally after many pages the control rod-related data starts, and is presented in several different ways. The report formats varies considerably between reactors, some have more japanese than others, and some reactors have more snazzy visualisations of the control rod positions and what looks like other important reactor core data. Lots of the data is repetitive as it covers the normal operating hours before earthquake.

In any case, I presume that nothing bad happened to control rods, but as there are a few glitches in the data at certain moments in time, Id love someone knowledgeable to review this. For example with reactor 1 data there is a bunch of numbers but at one brief point the word BAD features. Also what looks like a 15:00 visualisation of the rods all showing the -99 value. The -99 one is on nearly the last page of the document, and is somewhat obscured by being printed in the same place on the same page as another type of data. I expect its nothing/normal but I just want to be more certain.

Results of a quick scan:

No significant problems associated with rods. all SCRAMs were completed and guaranteed shutdown as far as I can see. from alarms logs, and graphs.

Unit 1 had high pressure injection for about 12 minutes as reactor pressure dropped due to combined cooling from injection and isolation condenser. SRVs sis not cycle during the period covered by graphs.Bothe EDGs started and successfully loaded. No data at time of tsunami.

Unit 2 - see previous post

Unit 3 Multiple SRVs cycled unlike Unit 2. Neither HPCI nor RCIC appears to have auto-initiated or they were secured by operators. No coolant injection until 1602 hrs when RCIC was started. Both EDGs started and were loaded successfully and tripped at about 1538 (tsunami) Data has gaps: No data for SRV B and D and no flow information for RHR and RHRSW.

Very important: If level 0-reference is Top of Active Fuel Unit 3 core uncovery began at about 1600 on 3/11. Because multiple SRVs were cycling and HPCI apparently did not start, the coolant discharged through SRVs was more than RCIC could make up so level continued to drop after RCIC started. It would be assumed they were able to recover level after the time of these graphs or it would have been unit 3 needing to be vented first.

Unit 5 Data is incomplete: no data from time of scram or time of tsunami. Data does show both EDGs started.

There is a great deal of mor information available on that site that could use an interpreter. My next goal is to check the computer lists of alarms to match it up to information in graphs.

 

[MiceAndMen]

I'd conclude that the earthquake and tsunami didn't break the connection to the stacks. Venting was observed on the 12th and 13th. Post #7684.

The highly kinetic environment caused by the explosions broke the stack ventilation.

I have to add AFAIK and FWIW...

And this picture shows the staircase on the south side of Unit 4 survived the tsunami as well.

http://www.tepco.co.jp/en/news/110311/images/110519_1_10.jpg

I know that has been discussed before, but I don't remember where exactly.

The ground level of the tanks in this photo is about 10 meters O.P.

http://www.tepco.co.jp/en/news/110311/images/110519_1_1.jpg

... making this photo very scary looking indeed...

http://www.tepco.co.jp/en/news/110311/images/110519_1_5.jpg

 

[westfield]

In their scenario, no-one is pumping water into the drywell. I don't know... the core may be out of the drywell, or not.

You know what I don't understand? TEPCO is pumping water in, using existing piping. If they can do that, they can also push a flexible borescope in, without altering their setup or opening additional valves or anything. Why aren't they?

Deploying a borescope would most certainly require them to rejig the pipe configuration and would definitely require an entry point to be opened for the borecam umbilical.


Regardless, I don't believe it would be possible given how many physical constraints there would be in the pipe system - pumps(even stationary) and filters are two things that immediately preclude using a borescope - seems highly unlikely any water injection lines into an RPV would not have at least one filter.

If it was even possible to do that what would they see? The inside of a sparge ring or the annulus between the RPV and shroud? Not that useful.

If they thought it was possible would they would risk getting an instrument stuck in their only water injection line?


Clearly what they need is a wireless, rad hard, temperature tolerant , waterproof, steam proof, camera "pill" that can somehow get through pumps and filters undamaged and that can transmit data wirelessly through tons of water, concrete and steel. Perhaps they havn't had the chance to pop down to Akihabara yet ;)

 

[zapperzero]

Deploying a borescope would most certainly require them to rejig the pipe configuration and would definitely require an entry point to be opened for the borecam umbilical.

If it was even possible to do that what would they see? The inside of a sparge ring or the annulus between the RPV and shroud? Not that useful.

Rejig how? Definite yes on the entry point, but since the connection they are using is outside the reactor building, I don't see that as a problem. Any pumps there may be are stopped, the whole jig is letting 9 cubic metres per hour through and I'm not sure I understand why there must be filters on that line. That may be just me being stupid, of course.

They might see a broken ring seal, meaning they have been pumping water into the drywell instead of the RPV all along. They might see a glow from the corium and that would be informative. They might not, and that would rule out some scenarios too. They might see there's nothing recognizable left of the shroud. If they manage to retrieve the scope, they might get some interesting trace chemicals to analyze.

 

[zapperzero]

p 15. Narrow range and wide range reactor pressure. After scram and void collapse pressure dropped, After Main steam isolation valve closure pressure rose to SRV cycling setpoint.
p 16 SRV F was cycling. gradual change in cycling rate shows decay heat decreasing.
p 21. RWL slow change up to 1800 mm and dow to 300 mm under operation of RCIC and SRVs
p 34. SRV F continues to cycle.
p. 35. Nothing significant.

Iirc, just one SRV cycling means it was operating automatically, because normal procedure is for the operator to open alternate SRVs, and so avoid local heating of the pool. Correct? Any guess on cause/consequences?

EDIT: this is reactor 2 data and the working assumption of TEPCO is the torus is broken in #2. This NRC document
http://www.nrc.gov/reading-rm/doc-collections/nuregs/staff/sr0933/sec3/108.html
says

Continued steam blowdown into the pool will increase the local pool temperature. The condensation rates at the turbulent steam/water interface are eventually reduced to levels below those needed to readily condense the discharged steam. At this threshold level, the condensation process may become unstable; for example, steam bubbles may be formed and shed from the pipe exit, oscillate, collapse, and give rise to severe pressure oscillations which are then imposed on the pool boundaries.

EVEN LATER EDIT, IMPORTANT: don't let Dymytry near that doc. He'll get the conniption fits if he sees it.

 

[NUCENG]

Deploying a borescope would most certainly require them to rejig the pipe configuration and would definitely require an entry point to be opened for the borecam umbilical.


Regardless, I don't believe it would be possible given how many physical constraints there would be in the pipe system - pumps(even stationary) and filters are two things that immediately preclude using a borescope - seems highly unlikely any water injection lines into an RPV would not have at least one filter.

If it was even possible to do that what would they see? The inside of a sparge ring or the annulus between the RPV and shroud? Not that useful.

If they thought it was possible would they would risk getting an instrument stuck in their only water injection line?


Clearly what they need is a wireless, rad hard, temperature tolerant , waterproof, steam proof, camera "pill" that can somehow get through pumps and filters undamaged and that can transmit data wirelessly through tons of water, concrete and steel. Perhaps they havn't had the chance to pop down to Akihabara yet ;)

If they are adding water through the core spray lines the borescope would need to be able to go through a spray nozzle. If they go in through the RHR system they need to go through the Recirc system jet pumps. If they go in with RCIC or HPCI they would need to go through the feedwater header holes. They would have a long way to go after they get into be able to get to the area below the core support plate. The most direct path would be through the jet pumps to the area of the lower plenum. All of these systems are in the lower level of the reactor building close to the torus in an extremely high radiation area.

A borescope expedition may be possible in spite of the difficulties. But here is the real question. Depending on what they see, what could they do differently than what they are already doing?

 

[Cire]

A few questions regarding the temperature sensors to figure out how trustable the temperature readings still are.

1. Does anyone know what kind of temperature sensors they are using? (Thermocouple, resistive, semiconductor, ...)
2. How is the sensor signal read out from the sensor to the control room? Are there amplifiers, AD converters, signal conditioning electronics and stuff like that located in the drywell close to the sensors?
3. How are the sensors mounted at their various locations? (Thermally conductive glue?)
   Is it conceivable that they can detach due to heat and humidity?

Concerning unit 1, if the sensors were just floating in the air, would it make sense that steam from the steamy atmosphere in there condenses on them and keeps them at roughly 100 degC ?

1. Thermocouple.

2. Likely. The amplifiers are probably subject to high levels of radiation atm.

3. The sensors are mechanically connected to the pipes/vessel etc.

4. Possible.

Don't forget the sensors also where damaged when they subject to temperatures greatly exceeding their maximum rating. I don't believe the values they report now can be trusted with any certainty.

 

[Cire]

If they are adding water through the core spray lines the borescope would need to be able to go through a spray nozzle. If they go in through the RHR system they need to go through the Recirc system jet pumps. If they go in with RCIC or HPCI they would need to go through the feedwater header holes. They would have a long way to go after they get into be able to get to the area below the core support plate. The most direct path would be through the jet pumps to the area of the lower plenum. All of these systems are in the lower level of the reactor building close to the torus in an extremely high radiation area.

A borescope expedition may be possible in spite of the difficulties. But here is the real question. Depending on what they see, what could they do differently than what they are already doing?

You're not getting an image sensor anywhere near those high levels of radiation unless all you want to see is white noise.

That implies you have to snake the worlds longest fiber optic bundle through, valves, piping, debris into an environment that would bake said fiber bundle.

Nobody is going to look inside the RPV until radiation levels drop to the point that people can access the dry well.

 

[westfield]

Rejig how? Definite yes on the entry point, but since the connection they are using is outside the reactor building, I don't see that as a problem. Any pumps there may be are stopped, the whole jig is letting 9 cubic metres per hour through and I'm not sure I understand why there must be filters on that line. That may be just me being stupid, of course.

snip>.

This is boring - Can we just leave this at the most obvious reason and the reason you didn't try and argue : they don't want to risk compromising the only cooling they have just to take an image of dubious usefulness inside the RPV? Thats more than a good enough reason to kill this discussion without getting too technical.

We could get into the piping complications of which there are potentially many but why bother, they are not going to try and send a borescope through that line. They will find a far more suitable entry point if\when they do try and image the inside of the RPV.

If you cannot see why there would very likely be filters on a water line leading into an RPV and you cannot understand why a borescope cannot get past most pump impellers, stationary or otherwise, then it's prob best to drop the idea.

 

[Rive]

You're not getting an image sensor anywhere near those high levels of radiation unless all you want to see is white noise.

'Radiation hardened cameras' (google it) are available for nuclear industry. But Westfield is right, no use for such cameras now, they have to bring the reactors in cold shutdown first, and then secure everything before visually checking the internals of the containment/RPV. Such pictures will be needed only for the decommissioning, which will not happen in the next few years.

Unit 5 Data is incomplete: no data from time of scram or time of tsunami. Data does show both EDGs started.

It was on outage, so no SCRAM was needed?? :-)

 

[MoonofA]

'Radiation hardened cameras' (google it) are available for nuclear industry. But Westfield is right, no use for such cameras now, they have to bring the reactors in cold shutdown first, and then secure everything before visually checking the internals of the containment/RPV. Such pictures will be needed only for the decommissioning, which will not happen in the next few years.

If there is an accessible 150mm line going into the drywell or reactor vessel this little machine would be the right thing to use.
http://www.khgmbh.de/wEnglisch/fernhantierung_kleinesinspektionsfahrzeug.php?navanchor=1010051

It was specially build for such purposes. These robots were offered by Germany to the Japanese at the start of this mess, but the offer was not taken.

I have copied parts from an image movie which shows the various nuclear accident robots Germany has for just such purpose. You can watch it here:
http://www.youtube.com/user/MoonofA#p/u/0/dbdwUnUvmFw

Edit: The French have similar equipment available on 24/7 standby

 

[zapperzero]

This is boring - Can we just leave this at the most obvious reason and the reason you didn't try and argue : they don't want to risk compromising the only cooling they have just to take an image of dubious usefulness inside the RPV? Thats more than a good enough reason to kill this discussion without getting too technical.

We could get into the piping complications of which there are potentially many but why bother, they are not going to try and send a borescope through that line. They will find a far more suitable entry point if\when they do try and image the inside of the RPV.

If you cannot see why there would very likely be filters on a water line leading into an RPV and you cannot understand why a borescope cannot get past most pump impellers, stationary or otherwise, then it's prob best to drop the idea.

Yeah. Okay. I'm too stupid to understand an impeller pump, a bundle of fiber optic is just the kind of object you would use to plug a pipe shut, you have an opinion on whether there likely are filters and pumps on that line and it's way more valuable than mine because...

I can also see the usefulness argument. Who cares if there's still core in the RPV?

But yea, let's stop this here, the SNR is bad even without feelings getting in the way.

 

[razzz]

Via http://ex-skf.blogspot.com/"

From Yomiuri Shinbun (9:14PM JST 5/18/2011):

中部電力は１８日、運転停止作業中に冷却水に海水が混入するトラブルが起きた浜岡原子力発電所（静岡県御前崎市）５号機に隣接した補助建屋の排気ダクトの出口で、ごく微量の放射性核種「ヒ素７６」を検出したことを明らかにした。

Chubu Electric announced on May 18 that a minute amount of arsenic-76, radionuclide, was detected at the exhaust duct of the ancillary building to the Reactor 5 reactor building at Hamaoka Nuclear Power Plant, where the sea water got mixed up in the reactor coolant while the reactor was being shut down.

　周辺環境への影響はないという。

The company says there is no ill effect on the environment.

　同社は、トラブルの影響で、海水中に存在する「ヒ素７５」が原子炉内で放射化したものとみている。

The company thinks arsenic-75 present in the seawater turned radioactive inside the reactor.