Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Japan Earthquake: nuclear plants part 2

  1. Jul 22, 2017 #1301
    Far more interesting than expected.

    First and foremost; the water level. The water level in Unit 3 is remarkably high. The probe results from the other 2 units show water levels much, much lower.
    At 1:40 in the video we see the probe swimming close to submerged control rod drives.
    [COLOR=#black]. [/COLOR]

    Knowing that the CRD bottom ends are located at the top of the pedestal and that the bottom of the pedestal is openly connected to the outer pvc area, that means the entire unit 3 Primary Containment 'light bulb', must be flooded to an amazing level (Similar to the light blue area below but also including the outer area).


    Considering that the unit building is above the water table and sea level, they must be pumping water in to keep it flooded.

    The second interesting thing is the structure of the deposited fuel melt. Anybody experienced in lava or melt flows would be able to say whether these globular formations were formed in air or in water, i.e. did the fuel melt burn through the reactor onto a dry floor where it could burrow further or was it quenched as it entered an already flooded containment.

    Lastly, muon results. It will be interesting to see when they are released, how they match up with what appears to be visible fuel deposits in the Unit 3 pedestal area. That could give some information as far as the lack of results from the scans on units 1 and 2 i.e. is the fuel in units 1 and 2 higher or lower etc.

    Unit 3 being so heavily damaged but still holding that much water is definitely the most surprising thing to see though.
  2. Jul 23, 2017 #1302
    As I said earlier remember unit 3 was depressurized so it did not have a hot debris ejection. Hot debris ejections are extremely likely to cause containment damage under the conditions units 1/2/3 were operating in.

    The severe accident guidelines at the time direct operators to flood containment following a core breach to establish "Minimum Debris Submergance Level". This is to stop or prevent a core concrete interaction. So they did just that. Unit 3's containment appears to be in much better condition than 1/2. There is speculation that the fact that unit 3 didn't have a containment failure led to hydrogen gas buildup in the upper head of the drywell and the head "burping" as a way to release gas pressure which caused the different hydrogen explosion than we saw at unit 1. There is history of BWR drywell a burping through the drywell head seals during containment pressure testing under some more extreme conditions so this isn't unheard of.
  3. Jul 23, 2017 #1303

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    Closure bolts yield ?
  4. Jul 23, 2017 #1304
    Latest results and images seem to indicate Unit 3 did in fact sustain a substantial hot debris ejection.

    Going by the definition of:

    "an explosion through the bottom of the reactor [where nuclear] material scatters all across the basemat of the containment liner below the bottom of the reactor. If a core catcher was installed when the plant was built, the material will spread across the core catcher, and will not be able to eat through the concrete at the bottom of the containment. If there wasnt, it could potentially undergo a reaction with the concrete for a while, where it can eat through the containment basemat. The whole time this is occurring, if it is not cooled and water is not injected to the containment, then the containment will start to heat up, overpressurize, crack, and leak radioactive material directly outside." - Hiddencamper

    Tepco engineers and the new images seem to be suggesting this is exactly what happened. We know there was no "core catcher", which is why I raise the point about whether the pedestal area was flooded at the time the core ejected. Depending on how deep into the concrete the fuel ate and how damaged the basemat was by this and the earthquake explosions is very relevant. If the basemat is no longer water tight and the building is flooded to a height of 30 feet above sea level or so, there must be a lot of water flowing past the fuel into the environment. That would explain the groundwater contamination issues at the site.

    Again, whether this new found apparent fuel will match up with the muon scans and what that says about the negative results on 1 and 2 could have a lot to say about fuel locations and depths over there too. Very interesting time and I'm glad TEPCO is sharing so much information
    Last edited: Jul 23, 2017
  5. Jul 23, 2017 #1305

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    I see at least some CRD mechanisms still in place
    which infers not gross vessel failure
    so i doubt there was any significant 'ejection'

    though a smaller drip is sure plausible
    which into water would cause a steam explosion likely scattering the fuel.

    When they find the actual fuel its physical arrangement will give more insight to what went on that morning.
  6. Jul 23, 2017 #1306
    I'm out of town right now, but when I get home I'll link a TEPCO report that shows the Automatic Depressurization System actuated and depressurized the reactor preventing a hot debris ejection.

    A hot debris ejection is when the vessel is still pressurized. If you have a hot debris ejection while the suppression pool exceeds the heat capacity temperature limit and the pressure suppression pressure, then containment failure is virtually guaranteed. The fact that unit 3 is retaining this much water demonstrates that it is far less damaged than units 1/2 which can't even keep the pedestal region flooded.

    There are pressure trends which were recorded on unit 3 where you can clearly see the ADS actuation. The logic was made up because primary containment and suppression chamber pressure were so high that it dummies the low pressure ECCS running signal which is a permissive for ADS to actuate.
  7. Jul 23, 2017 #1307
    This link discusses the rapid pressure drop in unit 4. They pretty much disprove everything but the ADS which gives the trace they saw (and lines up with what I've seen in simulator scenarios)

  8. Jul 23, 2017 #1308
    Is the rv pressure status at the time of the core ejection relevant? If the new images are accurate then it still shows that the core ejected from the RV and splattered around the CRD room, pedestal and outer PVC area. Whether it was complete and immediate ejection (i.e. under pressure) or a depressurised slow and partial release would only be relevant to calculate potential basemat penetration. Seeing as the explosions and earthquake put the integrity of the basemat at the time of the meltdown into the unknown I dont think it matters.

    Well this is the really interesting part. The pedestal is open to the outer pvc at the bottom here:


    So the water level inside the pedestal must be the same outside in the pvc. Is the outer PVC designed to be water tight to that level? If so then maybe they just add relatively small amounts of water to keep the building flooded. If not then they must be pumping in massive amounts but need to in order to provide some sheilding around the fuel in a relatively problematic location.

    What does that mean for unit 1 and 2? That they dont need the same amount of forced flooding because their cores are lower down in the basemats closer too or actually screened by the water table?

    The persistent groundwater contamination is coming from somewhere. It's just interesting to figure out whether unit 3 is the best of the bunch or the worst.
  9. Jul 23, 2017 #1309
    The SRVs discharge steam in a relatively controlled fashion into the suppression chamber. While a 1000+ psig bottom head rupture will catastrophically expand in the containment system while its already beyond its pressure limit. The severe accident mitigation basis specifically discuss the importance of depressurizing the vessel to prevent a hot debris ejection while the containment is above the pressure suppression pressure because you will exceed the design limit of the containment. The SAGs also tell you to immediately depressurized the vessel once you recognize that adequate core cooling is lost because of this.

    There is a huge difference between depressurizing to the pool and rupturing the vessel at pressure.

    As for the water level, this is how bwrs are designed. After a core melt breaches the reactor you were supposed to flood the suppression pool until water backfills the under pedestal region and submerged the fuel (called the Minimum Debris Submergence Level or MDSL). This changed post Fukushima for most bwrs.... but that's irrelevant. The goal of a post vessel failure, whether it's due to core melt or LOCA, is to flood containment to backfill the vessel and resubmerge the fuel. The containment is designed to be essentially leaktight. After backfilling, you pull the reactor head off and dig the debris out from the top using the water as shielding.

    Now we know unit 1/2 aren't leaktight anymore due to the damage they had. Neither one is capable of submerging the undervessel grating. However unit 3 looks like it's still nearly leaktight so this may be an option.
  10. Jul 23, 2017 #1310
    According to this document (July 13, 2017) http://www.tepco.co.jp/en/press/corp-com/release/betu17_e/images/170718e0101.pdf the amount of water injected to each reactor is comparable (70, 67 and 70 m3/day for units 1, 2 and 3 respectively).

    EDIT: And as of July 20 is 72 m3/day for each unit.
  11. Jul 23, 2017 #1311

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    Where in the images do you see that?
  12. Jul 23, 2017 #1312


    User Avatar
    Staff Emeritus
    Science Advisor

    Various news organizations are publishing pictures purportedly being 'likely molten fuel'. Unfortunately, there is no scale or reference, so it is impossible to judge the size of what is considered to be molten fuel, or a mixture of molten steel and fuel. The orange/brown is likely corrosion products from steel. The dark or black material could be molten fuel.



    I would hope that the robot can grab some samples for analysis of the composition.
  13. Jul 23, 2017 #1313

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    Thanks Astro.
  14. Jul 24, 2017 #1314
    There's a new quick report (in Japanese) regarding the third day of investigation in Unit 3 (22 July)
    Pages 1-2 show new photos from inside the pedestal.
    Page 3 lists short conclusions: The investigation of 22 July aimed and succeeded to obtain information regarding the lower part of the pedestal and surroundings. Once molten and then solidified masses were observed, as well as fallen objects such as pieces of grating, and sediments/deposits. The analysis will continue.
    The photos can also be seen here:
  15. Jul 24, 2017 #1315

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    Thanks Sotan !

    Has anyone looked in that 294 megabyte zipped folder?
  16. Jul 24, 2017 #1316
    Hey Jim. I hadn't even noticed that! I couldn't have played it at work anyway, but now I am home and I just looked at it and it's awesome, much more suggestive than the still photos. The swimming robot did a great job, I just wished I understood more or what I see. Reminded me a little of Titanic - and also of those scenes in which a paleoanthologist says "see here how well this bone is preserved" and all I see is a rock. The large pipe corroded (melted) so bad that it is missing a few patches (01:37) through which we see a smaller pipe inside? The blue-greenish stuff that we've seen in another unit too (00:48). The mix of materials at 01:20. The amount of light, floating sediment that is raised by the robot's propellers.

    By the way the clickable folder name in the link given by Jim is something like link1 / link2, link1 is "play the movie" and link 2 is "download the file".

    Edit: Then I found the video images of 21st too. Totally as impressive! Take a look (I only know this way to access them : http://www.tepco.co.jp/tepconews/library/archive-j.html?video_uuid=f1ak69jq&catid=69619)
    Especially intriguing moments: 00:12 and 00:53, that can't be steel corrosion, too conspicuous and non-uniform? they rather look like blobs of material splattered over structures. What's that at 00:19, a little current of water - under water? Or bubbles? And at 00:44, a large irregular mass near the still nicely rounded pedestal wall.
    Last edited: Jul 24, 2017
  17. Jul 24, 2017 #1317
    My apologies for not posting links to the suspected molten fuel deposit videos, I assumed they were commonly known as mainstream media have been covering it quite a lot over the last few days. As I said in my earlier post, the notion that this is ejected fuel isn't based on seeing the video images, TEPCO spokesman Takahiro Kimoto is on record since last week indicating they suspect the images are of melt out fuel. "You can see something melted and solidified fallen from the inside of the reactor" he said. (Images and the quote source are here: www.cnn.com/2017/07/24/asia/fukushima-robot-nuclear-fuel-detected/index.html)

    Unit 2 inspection already showed very strong indications of gross melt out there so I don't think it should be any surprise for unit 3.

    As you say, as far as the ratio of fuel that exited the RV, I don't think the size of the hole in the CRD roof is relevant. Molten fuel being a liquid, once container penetration is achieved, the entire contents are able to exit the RV just as well through a small bore hole as a large one. Bottom loading control rods are a known inherent weak-point in this particular reactor design after all. News sources are calling the deposit sizes 'large' but whether that is from casual arm chair eyeballing or from actual TEPCO officials looking at known equipment and CRD remains captured in inspection videos and using them for scaling is unknown. If we get enough images to tile together as was done for Unit 2, we should have a clearer picture.

    @Hiddencamper , Thank you very much for such an informative reply. I had no idea that the PVC was inherently designed to be water tight that way. This explains so much about events during the accident with the helicopters dumping water on the reactor buildings and what they were trying to achieve. Very interesting.

    As far as this part:
    I understand that it is better to have RV depressurisation than pressurised RV rupture, but if the depressurised reactor then goes on to suffer a melt through anyway, what difference does it ultimately make? I am correct is assuming that it only matters as far as trying to calculate how much the concrete under the reactor will be damaged/attacked and how long you have to re-establish cooling before radioactive materials leak directly outside?
  18. Jul 24, 2017 #1318

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    Thanks Sotan

    my Windows is dyng of update poisoning, Microsoft's preferred means of demise. It downloaded the file but refuses to play it. Perhaps i'll try the other machine tonight.

    Splattered ? As in perhaps thrown against the walls by expanding steam ? That might explain a lot.

    I would be very interested to see upper head area of the drywell (or is PCV correct term? The big light bulb..) Do you recall any photos ? . .
  19. Jul 24, 2017 #1319

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    Got it to play
    THANKS, Sotan and Charles ..

    I see what you mean. Around 1:21 and 1:50 could pass for slag.

    but speaking for myself I have to resist the natural tendency to speculate whatever is most exciting . So at this point i wouldn't assert that it's less benign than melted plastic cable insulation. .

    Maybe someone who's been under a BWR will recognize the mechanical parts.

    Astro nailed it - gotta get a sample of the stuff.. Activation products in it will tell a lot.
  20. Jul 24, 2017 #1320
    I think it is at 1.18. That's something dense and solid hanging down (!) from CRD support.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?