Japan Earthquake: Nuclear Plants at Fukushima Daiichi

[SteveElbows]

Unit 3 SFP Video: http://photo.tepco.co.jp/en/date/2012/201204-e/120416-02e.html

Cheers.

On first viewing, videos 3 & 5 offer us some views that were not apparent from the previously released photos.

Has anyone been able to identify any of the debris we see in the 5th video?

 

[clancy688]

As the steam plumes on the right-after-the-explosion picture suggests the containment integrity was broken at that time. As I recall after the explosion the RPV remained on ambient pressure.

So my bet:
The PCV released pressure and the hydrogen stocked up in the secondary containment till it exploded: this was the first 'bang'.

Then the RPV damaged and the overheated water in it turned to steam. This steam (and some more hydrogen) built the big cloud. Fortunately the wind moved the cloud toward the ocean, so the site remained relatively safe: also the steam separator stopped the worst of the solid fuel debris.

That isn't consistent with pressure data (March 14th):

Time                    RP(A)   RP(B/C) D/W       S/C      FNT    BHT     WL(A)  WL(B)   DW R/H         Ref
                        MPa abs MPa abs MPa abs   MPa abs  gr.C   gr.C    meter  meter   gr.C

23:00, March 14th       N/A     0.297   0.370     dscale    N/A    N/A     N/A   -2.30
22:40, March 14th       N/A     0.290   0.360     dscale    N/A    N/A     N/A   -2.30
22:00, March 14th       --approximate time for a fire at unit 3 or 4 -----------------
21:04, March 14th       N/A     0.270   0.360     dscale    N/A    N/A     N/A   -2.30
20:03, March 14th       N/A     0.284   0.360     dscale    N/A    N/A     N/A   -2.30
19:30, March 14th       N/A     0.283   0.335     0.500     N/A    N/A    -1.90  -2.30
19:00, March 14th       N/A     0.284   0.360     dscale    N/A    N/A     N/A   -2.30
17:00, March 14th       N/A     0.362   0.440     0.440     N/A    N/A     N/A   -2.20
16:00. March 14th       - webcam shows smoke coming from unit 3-----------------------
16:00, March 14th       N/A     0.409   0.480     0.475     N/A    N/A     N/A   -2.20
15:00, March 14th       N/A     0.399   0.480     0.470     N/A    N/A     N/A   -2.20
14:00, March 14th       N/A     0.382   0.460     0.450     N/A    N/A     N/A   -2.20
13:00, March 14th       N/A     0.352   0.430     0.430     N/A    N/A     N/A   -2.20
12:00, March 14th       N/A     0.292   0.380     0.380     N/A    N/A     N/A   -2.20
11:30, March 14th       N/A     0.291   0.360     0.380     N/A    N/A     N/A   -1.80
11:25, March 14th       N/A     0.292   0.360     0.380     N/A    N/A     N/A   -1.80
11:15, March 14th       N/A     0.316   0.380     0.390     N/A    N/A     N/A   -1.60
11:01, March 14th       ----------------- explosion in reactor building---------------
09:05, March 14th       N/A     N/A     0.450     N/A       N/A    N/A     N/A    N/A
09:00, March 14th       N/A     0.409   0.490     0.475     N/A    N/A     N/A   -1.50
08:00, March 14th       N/A     0.421   0.500     0.490     N/A    N/A     N/A   -1.00
06:50, March 14th       N/A     N/A     0.650     N/A       N/A    N/A     N/A    N/A
07:00, March 14th       N/A     0.439   0.520     0.500     N/A    N/A     N/A   -3.00
06:00, March 14th       N/A     0.282   0.425     0.400     N/A    N/A     N/A   dscale
05:00, March 14th       N/A     0.282   0.365     0.345     N/A    N/A     N/A   dscale
04:00, March 14th       N/A     0.260   0.340     0.325     N/A    N/A     N/A   -2.80
03:00, March 14th       N/A     0.235   0.315     0.305     N/A    N/A     N/A   -2.30
02:00, March 14th       N/A     0.178   0.275     0.275     N/A    N/A     N/A   -2.25
00:30, March 14th       N/A     0.152   0.255     0.255     N/A    N/A     N/A   -2.2

Containment pressure did fall after the explosion, but it rose again to higher levels. Not the behaviour I'd expect from a leaking containment. But there's also the possibility that the sensor, as so many others, already went to hell at that point.

Taken from:

http://gyldengrisgaard.dk/fukmon/uni3_monitor.html

 

[SpunkyMonkey]

Has anyone been able to identify any of the debris we see in the 5th video?

We seem to see main horizontal beams of the FHM that ran east-west across the pool. If we refer to the same video.

 

[SpunkyMonkey]

Containment pressure did fall after the explosion, but it rose again to higher levels. Not the behaviour I'd expect from a leaking containment.

But that the containment was leaking after the explosion is obvious, for example "www.youtube.com/watch?v=iV_r5Ifp8-I&hd=1#t=44s" . What one has to consider is that probably after the blast, melting fuel kept off gassing and more corium kept forming, boiling water ever-more vigorously and thereby increasing the pressure.

The containment was not blown wide open, just enough small leak paths were blasted open to have allowed a mushroom-cloud of steam to escape in an instant of sudden explosive force. After the explosion, the force propelling depressurization falls and pressures build back up as corium keeps off gassing and boiling water.

 

[Rive]

That isn't consistent with pressure data (March 14th):

Indeed. Very thanks for the data. Then the explosion-like sudden decompression can be excluded.

However I think it's safe to assume that something had a leak there after/around the time of the explosion: the data shows steady downward trend without venting attempts.

Then maybe some more hydrogen stocked in the lower parts of the building can be an explanation?

... small leak paths...

I think small leak paths would not be enough for that big mushroom cloud. An overheated RPV would, but small leaks wouldn't.

 

[SteveElbows]

But that the containment was leaking after the explosion is obvious, for example "www.youtube.com/watch?v=iV_r5Ifp8-I&hd=1#t=44s" . What one has to consider is that probably after the blast, melting fuel kept off gassing and more corium kept forming, boiling water ever-more vigorously and thereby increasing the pressure.

The containment was not blown wide open, just enough small leak paths were blasted open to have allowed a mushroom-cloud of steam to escape in an instant of sudden explosive force. After the explosion, the force propelling depressurization falls and pressures build back up as corium keeps off gassing and boiling water.

Yeah, Id go along with that, although I am somewhat suspicious of the quality of various measurements too.

 

[SteveElbows]

They did vent reactor 3 again several times (or at least once) after the explosive event didn't they?

 

[MadderDoc]

Cheers.

On first viewing, videos 3 & 5 offer us some views that were not apparent from the previously released photos.

Has anyone been able to identify any of the debris we see in the 5th video?

Yes, we are again seeing parts from the NW corner of the FHM bridge, corresponding to the part closest to Quince in this photo of FHM2.
http://photo.tepco.co.jp/library/120228/120228_02.jpg

The FHM3 is quite similarly built, but as westfield notes it is not identical. A relevant difference here is that the FHM3 bridge has extra diagonal reinforcement beams built into the bottom section of the northern face of the bridge, as shown in this video (at about 2:25)
http://video.sina.com.cn/v/b/48514912-1822430771.html

The identifiable parts which we have seen of the FHM appear to be no longer attached to each other, nor to other parts to which they were attached. Separation appear to have occurred by breakage of interlinking bolts. I fear we will find this FHM to be thoroughly broken up and fubar.

 

[jim hardy]

But there's also the possibility that the sensor, as so many others, already went to hell at that point.

one wonders what were they using for pressure indication? Got to put ourself in their shoes at the time:

we speculated at the time that with no electric power it was likely a mechanical pressure gage on the fire truck pumping into reactor that they used, hence readings would reflect the pressure drop along piping connecting the two. They'd be high when there is flow and accurate only at no flow. Operators would know to reduce flow to get a true reading, but the poor guy who's copying numbers for the record might not be so savvy. "Just read this gage and write it here every fifteen minutes"

I seem to remember looking at JStolfi's remarkable graphs and seeing correlation between flow and pressure. But that was a year ago.. don't trust my memory, 'cause i dont.

 

[MadderDoc]

<..>
Containment pressure did fall after the explosion, but it rose again to higher levels. Not the behaviour I'd expect from a leaking containment. But there's also the possibility that the sensor, as so many others, already went to hell at that point.
Taken from:
http://gyldengrisgaard.dk/fukmon/uni3_monitor.html

Those are earliest published data, and not the best available. Since then Tepco has assembled better and more complete data sets,
which yields better insight also in the period around the unit 3 explosion:
http://www.tepco.co.jp/nu/fukushima-np/f1/images/syusei_level_pr_data_3u.pdf

The impression of a rather sudden pressure drop concurrent with the explosion remains, however.

Data indicate to me that leading up to the event there was no longer any clear barrier between the atmosphere of the RPV, and the atmosphere of the primary containment. The whole system was well past design limits, the core was in the process of melting down, and the primary containment was probably already leaking to some degree. The sudden pressure drop in connection with the explosion is consistent with an assumed abrupt increase in leak area, resulting in a blow out of hydrogen and steam. One would expect such a blow out to be decreasing over time, as accumulated energy is being released from the system. In the case of unit 3, a rather vigorous evolution of steam was seen right after the explosion at 11 am, however this steam evolution swiftly subsided to become invisible before noon. (Only later in the afternoon unit 3 resumed steaming, apparently coincident with a renewed build up of pressure within the containment)

 

[jim hardy]

One would expect such a blow out to be decreasing over time, as accumulated energy is being released from the system. In the case of unit 3, a rather vigorous evolution of steam was seen right after the explosion at 11 am, however this steam evolution swiftly subsided to become invisible before noon. (Only later in the afternoon unit 3 resumed steaming, apparently coincident with a renewed build up of pressure within the containment)

indeed if water that's pumped in as liquid comes out as steam through a constant leak area, one would expect strong flow^2 relation to pressure.

 

[zapperzero]

It's heartening to see discussion of this topic, but also frustrating to know that the dedicated thread is lost. I wonder if the FHM really is in pieces. It wouldn't be much of a surprise if it were so, as it takes a lot of force to crumble concrete into small pieces like that.

It seems as if the #3 RPV was leaking hydrogen and steam at a fairly rapid rate even before the explosion. I was very skeptical regarding the theory of hydrogen going from Unit 3 into the others, but I am almost convinced, now.

 

[SteveElbows]

we speculated at the time that with no electric power it was likely a mechanical pressure gage on the fire truck pumping into reactor that they used, hence readings would reflect the pressure drop along piping connecting the two. They'd be high when there is flow and accurate only at no flow. Operators would know to reduce flow to get a true reading, but the poor guy who's copying numbers for the record might not be so savvy. "Just read this gage and write it here every fifteen minutes"

I find it unlikely that this was the source of pressure readings, since readings at various reactors were taken at moments when fire-trucks for reactor water pumping were not available, and we also have drywell and wetwell pressure readings.

Batteries were used to power existing instrumentation.

 

[Yamanote]

The latest plan for the fuel removal of unit 4 is now available:

http://www.tepco.co.jp/en/press/corp-com/release/betu12_e/images/120416e0201.pdf

 

[Yamanote]

And there are also some videos of SFP #3:

http://photo.tepco.co.jp/en/date/2012/201204-e/120416-02e.html

 

[tsutsuji]

http://www.tepco.co.jp/nu/fukushima-np/images/handouts_120416_04-j.pdf report on the recent nitrogen injection system failures (pressure swing adsorption)

http://www.tepco.co.jp/nu/fukushima-np/images/handouts_120416_03-j.pdf report on the recent contaminated water leak after reverse osmosis

http://www.tepco.co.jp/nu/fukushima-np/images/handouts_120416_02-j.pdf report on unit 4 SFP cooling system leak

http://www.tepco.co.jp/nu/fukushima-np/images/handouts_120416_05-j.pdf The plan to remove the remaining part of unit 4 reactor building roof

 

[Tepconium-311]

These posts made me think of a scenario.

If it did so before the FHM fell, was it thrust at an almost horizontal angle like a javelin?

Then the explosion-like sudden decompression can be excluded...
...I think small leak paths would not be enough for that big mushroom cloud. An overheated RPV would, but small leaks wouldn't.

I find it unlikely that this was the source of pressure readings, since readings at various reactors were taken at moments when fire-trucks for reactor water pumping were not available, and we also have drywell and wetwell pressure readings.

Readings were manually taken in intervals, yes?
Then it would be unlikely that explosion pressure/vacuum spike was recorded, right?
Could we then actually really exclude existence of such a spike?

And now my questions:
Could there have been a pressure surge ("explosion") in RPV/drywell which lifted concrete reactor seal a bit for a moment, propagating hydrogen explosion from drywell to maintenance floor 5, so pressing reactor concrete seals back into place, mostly undamaged except two or three points ("white steam leaks")?
If so, wouldn't pressure wave have been directed in a way that would lift up FHM from below sideways, and then push it horizontally like a javelin into the pool?

 

[thebluestligh]

These posts made me think of a scenario.
Readings were manually taken in intervals, yes?
Then it would be unlikely that explosion pressure/vacuum spike was recorded, right?
Could we then actually really exclude existence of such a spike?

And now my questions:
Could there have been a pressure surge ("explosion") in RPV/drywell which lifted concrete reactor seal a bit for a moment, propagating hydrogen explosion from drywell to maintenance floor 5, so pressing reactor concrete seals back into place, mostly undamaged except two or three points ("white steam leaks")?
If so, wouldn't pressure wave have been directed in a way that would lift up FHM from below sideways, and then push it horizontally like a javelin into the pool?

The AEC transmitted a scenario where the head bolts stretched, which GE didn’t think was what happened.

"All the three units (IF] -3) experienced containment overpressure in the early phase of the accident, which caused excessive leakage from flange, ailocks and other peentrations. Hydrogen produced by Metal Water reaction in the core damage process must have escaped to the reactor building via this containment leakage path and accumulated on the top and caused deflagration/detonation there. Further, containment was steam inerted and a large fraction of hydrogen escaped -by “hardened” venting."

http://www.scribd.com/doc/89386167/AEC-Hydrogen-Analysis-Pages-From-C142015-02B

 

[SpunkyMonkey]

I think small leak paths would not be enough for that big mushroom cloud. An overheated RPV would, but small leaks wouldn't.

How does an overheated RPV allow the big mushroom cloud to escape? Assuming just leaks, a steam explosion in the containment would force out a larger amount of steam than the slower steady lease from just boiling water. Imagine a plastic bottle you can squeeze and there's a steam generator in it with a steady amount of steam ejecting from leaks in the bottle. Then you suddenly squeeze the bottle with great force (simulating the explosion) and poof, a much bigger dose of steam shoots out the same leaks. No?

That aside, in "www.youtube.com/watch?v=VZoBb-8oxdg&hd=1" , the well cap is assumed to've been momentarily lifted, which is denoted as it falls back in place.

 

[Rive]

How does an overheated RPV allow the big mushroom cloud to escape?.

I mean: if all (or: most) of the overheated and pressurized water in the RPV is suddenly released (through a BIG hole) then it would be more than sufficient explanation for the secondary, mushroom-like cloud.

Just (relative) small leaks could not be able to carry out the necessary energy I think.

To turn it around: if that cloud were from the RPV then the release path would not be able to close itself and keep a relative high (or even just any over-) pressure after the explosion.

One more turn: I think we should assume that the energy of the secondary mushroom was also based on chemical reaction, and not on (or at least not mainly on) steam explosion.

(Sorry for my English.)

Ps.: Another possible explanation is that after the primary hydrogen explosion the escaping steam/hydrogen mixture burned for some seconds. That would require just a relative small hole, but it also would mean sufficient energy for a mushroom.

 

[MadderDoc]

I mean: if all (or: most) of the overheated and pressurized water in the RPV is suddenly released (through a BIG hole) then it would be more than sufficient explanation for the secondary, mushroom-like cloud.

I can see the need for a hole of a certain size, but not why it would have to be a hole in the RPV. A hole in the primary containment would seem quite fine for the job seeing the containment contained the main portion of the overheated water in the system. Indeed pressure readings indicate that the pressure in the PCV was higher than in the RPV leading up to the event. That could indicate that there was already a big hole connecting the RPV and the PCV, meaning any sudden leak to the upside would tap from both sources.

Just (relative) small leaks could not be able to carry out the necessary energy I think.

To turn it around: if that cloud were from the RPV then the release path would not be able to close itself and keep a relative high (or even just any over-) pressure after the explosion.

Perhaps we could have some reason to not trust the accuracy of the readings after the explosion? One failure mode could be the drying out of reference legs, which would if I understand result in faulty (too high) readings.

One more turn: I think we should assume that the energy of the secondary mushroom was also based on chemical reaction, and not on (or at least not mainly on) steam explosion.

(Sorry for my English.)

Ps.: Another possible explanation is that after the primary hydrogen explosion the escaping steam/hydrogen mixture burned for some seconds. That would require just a relative small hole, but it also would mean sufficient energy for a mushroom.

Considering the state the reactor was in, it stands to reason that any steam release from the containment could have included hydrogen possibly available for combustion, thus (only) adding heat to the mixture in the cloud. However, there would seem to me to be quite sufficient energy contained in the overheated water in the RPV/PCV, so I think perhaps it is an unnecessary assumption to make that combustion of hydrogen in the rising cloud played a significant role for its evolution.

 

[SteveElbows]

They have a different kind of robot which they are going to use to survey more of the torus of reactor 2 on the 18th:

http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_120417_03-e.pdf

I note that the building layout diagrams showing where it will go have a bit more detail than the ones they normally use. The resolution is barely good enough to read the labels but quite a number of them can be worked out if you zoom in on the pdf.

 

[jim hardy]

This INPO report from last November may be of interest to newcomers.

http://www.nei.org/resourcesandstat...t-the-fukushima-daiichi-nuclear-power-station

A Unit 3 timeline begins on p91 (that's p91 of the report which is p94 of the PDF).

It appears they were trying to lower primary containment pressure by venting in the hours before the explosion.

 

[tsutsuji]

They have a different kind of robot which they are going to use to survey more of the torus of reactor 2 on the 18th:

http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_120417_03-e.pdf

I note that the building layout diagrams showing where it will go have a bit more detail than the ones they normally use. The resolution is barely good enough to read the labels but quite a number of them can be worked out if you zoom in on the pdf.

I appreciate the fact that the sump pits are shown on the right hand diagram. It enables to narrow the possible locations for the pedestal opening ("slit").

 

[Yamanote]

Indeed, pretty cool and detailed drawings this time, would like to walk through the plant to see how it looks in reality! Years ago I did such kind of drawings for other types of big industrial plants, but I have never been to an NPP.

 

[tsutsuji]

http://www3.nhk.or.jp/news/genpatsu-fukushima/20120418/index.html On 18 April the robot will do such things as inspecting from the catwalk the vicinity of the inspection manholes, where it is possible that there is damage, or measuring the surroundings' radiations.

 

[SteveElbows]

Two images from the robot, showing manhole covers:

http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_120418_04-e.pdf

 

[SteveElbows]

The press photo page version of those images include some radiation detection levels:

http://photo.tepco.co.jp/library/120418_02/120418_11.jpg

http://photo.tepco.co.jp/library/120418_02/120418_12.jpg

 

[Cire]

I have seen quite a bit of speculation that there is a large vertical crack in thereactor pressure vessel (RPV), but cannot figure out how they could tell. There has been no entry of the Dryell (Primary Containment) due to rad levels and I am not aware of cameras that would survive longtime exposure in the drywell during normal operations and they certainly haven't installed cameras since the accident. If pressure was high enough to cause a break it would likely occur in piping or rod penetrations of the RPV. Breaches due to core melt would be on the bottom of the vessel. It just doesn't seem to me that a vertical crack in the RPV can be any more than speculation.

At least I hope so.

There are many more easier ways to fail the interfaces to the RPV then to crack the actual containment vessel. I think we'll find failed piping and as you mention rod penetrations etc and a mostly intact RPV.

Just my guess.

 

[tsutsuji]

http://www3.nhk.or.jp/news/genpatsu-fukushima/20120418/index.html [unit 2] On 18 April, the robot checked 90% of the upper part of the suppression chamber but until now no major damage or deformation was found. The north and south inspection manholes were checked but water leaks or traces of leaks were not found, and they looked comparatively clean. Otherwise, Tepco says that leaks might be located in the parts that connect the PCV and the suppression chamber, but their status has not been checked yet and Tepco has yet to analyse the pictures taken on 18 April. The highest radiation was 120 mSv/hour in the north-western region. Saying that "It is a high value, but it is still possible to plan short time human operations for important tasks that cannot be performed by robots", among other things, Tepco is taking the necessary steps to grasp the radiation distribution.

 

[LabratSR]

I stumbled across this on the TEPCO Photo page.

General Plan and Start of Main Work of the Cover for Fuel Removal of Unit 4

http://www.tepco.co.jp/en/press/corp-com/release/2012/1201934_1870.html

 

[Rive]

Regarding the radiation in the torus room: would it reduce the airborne radiation if they cover the water surface with oil?

 

[MadderDoc]

These posts made me think of a scenario.
Readings were manually taken in intervals, yes?
Then it would be unlikely that explosion pressure/vacuum spike was recorded, right?
Could we then actually really exclude existence of such a spike?

Right, we have no continuous recordings, and not even a lot of discrete readings. Strictly the pressure readings we have can only say that the reactor pressure in unit 3 appeared to be lower after the event than it did before.

And now my questions:
Could there have been a pressure surge ("explosion") in RPV/drywell which lifted concrete reactor seal a bit for a moment, propagating hydrogen explosion from drywell to maintenance floor 5, so pressing reactor concrete seals back into place, mostly undamaged except two or three points ("white steam leaks")?

I hope you don't mind some correction in terms of understanding. The RPV/drywell/wetwell,( in short the reactor) of unit 3 is strictly a steel container, it is nowhere sealed with concrete. It's got interesting things inside it, and it has an interesting shape, but looked at from the outside one must visualise it as a mere steel container, having a closed surface made of steel. A leak in it implies there's a hole in that steel surface.

However true enough there's concrete packed up all around the reactor, for support, for radiation shielding, and for making the building structures around it, but none of all that concrete is there to keep the reactor tight. Specifically the concrete slabs stacked on top of the reactor, which I think you are referring to, are not there to keep any pressure at bay. You should, in principle, be able to operate the reactor quite well without them. So, to get back to your point, an assumed falling back of those top concrete slabs after an explosive pressure surge from the reactor could not be assumed to reseal the hole in the reactor below, from which that pressure surge would have to be coming from.

If so, wouldn't pressure wave have been directed in a way that would lift up FHM from below sideways, and then push it horizontally like a javelin into the pool?

The hall mark of the unit 3 explosion is exactly the vertically rising cloud over the building. It stands to reason that this vertical rise represented a considerable upwards mass flow over a short period. Anything that stood in its way should have experienced the impulse. So I can well imagine your scenario, it is possible that the FHM was lifted up/sideways. Only from what I have seen of the FHM it appears to be more beaten up than I'd expect if it had been just pushed sideways into the pool.

 

[Rive]

This might be relevant regarding U3 explosion: http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_120419_03-e.pdf

 

[SteveElbows]

This might be relevant regarding U3 explosion: http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_120419_03-e.pdf

Aha, very interesting.

Its not terribly surprising given that we have previously seen attention paid to high radiation levels from the rail on the floor in front of this equipment hatch.

( Last november http://www.tepco.co.jp/en/nu/fukushima-np/images/handouts_111122_02-e.pdf )

However I do not recall Tepco talking about the 'crevice between shield plug and reactor building' before, so I think this new information is only going to increase suspicion that TEPCO do not reveal every detail as soon as they discover it.

My knowledge is not sufficient to judge whether this damage would occur due to explosion (e.g. explosion within containment) or whether damage to seals due to heat or pressure would be sufficient to cause such a gap & leak.