Japan Earthquake: Nuclear Plants at Fukushima Daiichi

[NUCENG]

It says so right in the report:

and in another place:


Note it says "reactor building" not "containment". It struck me as very odd, too.

Seems to lend credence to the idea that some damage to piping took place very early on. I doubt fuel uncovery alone could increase doses as much, but then, what do I know?

Elsewhere in the report, there is an odd passage about checking the status of the IC by observing the steam that came out.



In my mind's eye, this plays out as operators peeking beyond a corner and seeing a wisp of steam, too concerned by what the dosimeter was showing to actually walk up to those pipes. I dunno... maybe I watched too much Hollywood.

The IC condensers are in the reactor building outside of primary containment (Drywell) The tubes inside the ICs contain steam from the RPV, condensing and returning to the RPV driven by natural circulation. After the operators stopped IC operation and fuel damage occured, the first thing released was the so-called "gap release source term" nobel gases and other volatiles that escaped from fuel pellets and collected in the gap inside a fuel rod. As damage progressed massive amounts of hydrogen and further gas releases from fuel pellets gave more non-condensible releases into the RPV. Those non-condensible raqdioisotopes would have been in the steam in the pipes to the ICs. In addition, pressure rose inside the containment, leakage would have released more radiation into secondary containment. (Remember, pressure was more that twice the design limit for containment.)

The indications on the instruments for the IC before the tsunami hit showed that the IC was working until it was turned off. Based on that and the explanation of the radiation above I don't see a need to assume IC damage occurred during the earthquake.

 

[clancy688]

But the operator checked sometime before 1800. zapperzero said here that the valve was closed at 1818. According to the TEPCO analysis, fuel didn't get uncovered before 1800.
The timing doesn't fit. Why a gap release when the core is still covered?

concerning the fact that at around 17:50 on the same day, they could not check the IC's condenser tank water level because the radiation dose in the vicinity of unit 1's reactor building was high.

 

[tsutsuji]

http://www3.nhk.or.jp/news/genpatsu-fukushima/20120107/0500_osensui.html 140 tons of low contaminated water (most likely rain water) were found in a different tunnel. This tunnel is not connected to the sea. Tepco has a plan to check about 100 locations among the underground tunnels that connect to the waste treatment facility buildings where the high contaminated water is stored.

But the operator checked sometime before 1800. zapperzero said here that the valve was closed at 1818. According to the TEPCO analysis, fuel didn't get uncovered before 1800.
The timing doesn't fit. Why a gap release when the core is still covered?

I think the valve was closed at 18:25. And Tepco's simulation said fuel uncovering started at 17:46 :

From your text, we can place this detection at somewhere between 17:19 and 17:50.
So we can suppose containment breach? Is it coincidence that the simulation says 17:46 is when water level reached TAF?

 

[clancy688]

I think the valve was closed at 18:25. And Tepco's simulation said fuel uncovering started at 17:46 :

I think they said that fuel uncovering started around 3 hours after SCRAM. Which was at 14:46.

I'm using this TEPCO-analysis for all of my statements. But I don't know if it's the most recent one.

Anyway, it states that fuel unovering started around 3 hours after SCRAM, but the depicted graphs show that TAF was reached at or a couple of minutes after 18:00.
Moreover, on page 4 there's a depiction of the core support plate 4.8 hours (19:30) after SCRAM with only 2% core damage in the lower middle of the core.

So now I have one question - when do cladding damage / gap releases occur (state of the core: ~3 hours after SCRAM)? In the moment the core is uncovered? Or later? 30 minutes? 1 hour?

So even if the core was uncovered moments or several minutes before the operator checked for the shell side IC pools, is it still possible for the fuel to rupture and for the fission elements to reach the shell side? Especially when the IC was effectively out of action, with the inner side valves (nearly) fully closed?

I still don't see any way for fission products to reach THAT specific place in such a short amount of time and over nearly fully closed piping with no or only minor natural circulation.

So my conclusion would be that those fission products reached the shell side before the IC shut down, hence before the Tsunami hit. And that leads to the assumption that the fuel may've been damaged during the earthquake.
Little fissures in the fuel rod cladding should be enough to let radioactive noble gases escape.

 

[Pakman]

When talking about how could high radiation appear in the reactor building, I think we miss the true matter: how do we know there was high radiation?

As we know from the report the dosimeters that were used by operators has full scale of 2,5 mkrSv/h of gamma radiation. It's only in 25 times more than natural background radiation. To imagine how insignificant it is, realize that it will take take for 5 years to achieve the emergency doze limit of 100 mSv. In Sophie, Bulgaria, natural radiation level is more than 1,5 mkrSv/h and people lives there.

Is it supposed to be so low radiation in the reactor building rooms behind the airlock even under normal operation? Really so? It's hard to believe in it. So there's the question what is the purpose of using such detectors when entering the building? Could it show another result but being overscaled? By myself, I use quite the same detectors when I go to market. Did they go to market? Or may be they thought they are working in the flower garden?

Coming back to Chernobyl, there were used dosimeters with top of 10 mSv/s (per second! not hour) and it was not enough. That is the suitable size of measurement, because the track of time in severe accident is determined by the time remaining to onset of melting. And it is hours, not years. Exactly such a dosimeter should be used to figure out whether or not you achieve doze limit being working for an hour or a half in the reactor building.

 

[clancy688]

Is it supposed to be so low radiation in the reactor building rooms behind the airlock even under normal operation?

Are the IC pools behind an airlock? I really don't know. But they are not in the primary containment. Are there airlocks for secondary containments?

You may be right regarding the actual doses. But that doesn't change the fact that there was radiation where it was not supposed to be at a time when the core still should've been undamaged.

And if the operators were equipped with dosimeters with a top scale of 2.5 uSv/h, then normal radiation levels must be well below that. What's the point in wearing a dosimeter if it goes off at radiation doses considered to be normal every five minutes?

 

[rmattila]

Would anybody by a chance happen to have information, how much the water level in the reactor normally sinks after closure of the MO-3 valves? According to the transient recorder data published by TEPCO, the measured level in the reactor dropped by 400 mm after the 15:04 closure of the valves. Since the main circulation pumps were already stopped at that time, and there should not have been any transients affecting the level measurement impulse tubes, it seems that about 8 cubic meters of water really did exit the reactor before the level stabilized, which feels rather a large amount to fill the IC tubes.

I have no experience on the isolation condensers and have so far been unable to verify if level drop of this magnitude is a normal phenomenon or not.

 

[tsutsuji]

Have you read Note 31:

note 31: If radioactive substances are generated in the reactor pressure vessel, radiations such as gamma rays are not only spread into the reactor building even if the reactor pressure vessel and the primary containment vessel are not damaged,

What do you think they mean ? I thought they mean that if there are high radiations in the RPV or in the PCV, such as during a meltdown, the PCV's wall is not thick enough to stop all the gamma rays. Is that correct ? If this is correct, then the next question is: are the doors thick enough to stop the gamma rays that come from inside the reactor building ?

 

[Pakman]

So now I have one question - when do cladding damage / gap releases occur (state of the core: ~3 hours after SCRAM)? In the moment the core is uncovered? Or later? 30 minutes? 1 hour?

As I understand this, at 17:20 it was the attempt to enter the reactor building, which was immediately renounced due to "high" radiation. At 17:50 it was reported to the headquarter. So, most likely the actual time of radiation mesurement is 17:20. By all means the core was covered at this time.

The quake could damage the cladding, yes. It could easely be checked by viewing the alarm list if there is high radiation alarm to the main steam (it flows towards turbine for 40 sec after SCRAM). I don't have this list under my hand righ now.

The reason for increased radiation at the reactor building airlock could be the SVR's operation. It drains inventory from RPV, which is in the drywell, to suppression pool which is out in the room with less thicker concreate walls than drywell has. So the radiation from the thorus inventory could affect the measurement at the airlock.

 

[tsutsuji]

Are there airlocks for secondary containments?

As Tepco wrote (about unit 1 on March 11) :

Government and other authorities were notified at 23:40 of survey results showing rising radiation dose levels inside turbine building (1.2mSv/h in front of turbine 1st floor north-side airlock and 0.5mSv/h in front of turbine 1st floor south-side airlock).
http://www.tepco.co.jp/en/press/corp-com/release/betu11_e/images/110810e21.pdf page 9

23:00 - Radiation dose levels rise inside the turbine building (1.2 mSv/h in front of turbine 1st floor north side airlock, 0.5 mSv/h in front of 1st floor south side airlock) due to the influence of rising radiation in the reactor building.
http://www.tepco.co.jp/en/press/corp-com/release/betu11_e/images/110810e21.pdf page 21

I guess the answer to your question is "yes". But I would be glad that Tepco provides a map showing where all these airlocks are located. What do they mean by "turbine building airlock" ? Is it an airlock between the turbine building and the outside, or between the turbine building and the reactor building ?

 

[Pakman]

Would anybody by a chance happen to have information, how much the water level in the reactor normally sinks after closure of the MO-3 valves?

I see you're tightly in subject, neighbor! I counted 10 cubic meters. About. You think this is for both IC or only for B?

However, I bet nobody else here even know what tubes you are talking about.

 

[Pakman]

I guess the answer to your question is "yes". But I would be glad that Tepco provides a map showing where all these airlocks are located. What do they mean by "turbine building airlock"? Is it an airlock between the turbine building and the outside, or between the turbine building and the reactor building ?

BWR has quite contaminated steam-water circuit loop, including the turbine flow, so the turbine set in maintenance-free room with airlock to outside. Almost like secondary containment of the reactor building. Therefore one doesn't simply walk into turbine hall.

 

[Pakman]

What's the point in wearing a dosimeter if it goes off at radiation doses considered to be normal every five minutes?

Who knows how far is our knowledge? What's the point in fail-saving the emergency cooling system due to DC power loss, which most likely means something unexpected is happened?

 

[rmattila]

I see you're tightly in subject, neighbor! I counted 10 cubic meters. About. You think this is for both IC or only for B?

As both trains were operating until 15:03-15:04 and then both of them were closed by shutting the MO-3 valve, I suppose both of them would continue taking steam until they fill up to a stable level. But I have no idea of the amount of space available for the steam to get condensed into water.

 

[rmattila]

However, I bet nobody else here even know what tubes you are talking about.

Well, theoretically (=if the level measurement reference line is very close to the reactor temperature) it would be possible that while the IC is operating and the reactor pressure is dropping, the hot condensate at the top of the level measurement reference tube would flash to steam, causing the level measurement to show too high a value, and after the IC is closed and the pressure rises again, the reference tube would fill up and the level measurement would drop to the correct value.

But i doubt it, since the pressure transients caused by operation of the IC were not very large. However, as I said, I have never seen an IC operating, so I can't say for sure.

 

[Gary7]

Interesting story about a plan to use muons to take a virtual x-ray of the reactor cores.

http://www.yomiuri.co.jp/dy/national/T120107003539.htm

 

[jim hardy]

"" the hot condensate at the top of the level measurement reference tube would flash to steam, causing the level measurement to show too high a value, and after the IC is closed and the pressure rises again, the reference tube would fill up and the level measurement would drop to the correct value. ""

that's one reason it's recommended to depressurize quickly in station blackout.
depressurizing reduces reactor temperature
which reduces heatflow from vessel to containment, ( which has no cooling in blackout, )
where the 'reference legs' are located.
the containment temperaturre needs to be kept low
both for sake of electronics in there and you don't want the containment's metal wall to expand from the heat, buckle and crack
not to mention reference legs.

reference legs depend on steam moving into condensate pot, condensing and running back out the sloped sense pipe toward vessel, to keep condensate pot full.
so the condensate pot will be at saturation temperature.
that's how you check them - feel and if they're not scalding hot they're "gas bound", ie filled with noncondensables that keep the steam out.
we had vents atop ours to let out the noncondensibles.

the cooler water in reference leg can carry dissolved noncondensibles, just like carbonated soda does.
If the reference legs got hot enough to boil out some of the water or drive out dissolved noncondensibles
at a time when there was significant noncondensibles in the vessel (hydrogen? )

i believe the reference legs would become gas-bound and not refill.

just an old instrument guy's estimate.

 

[Pakman]

The reason for increased radiation at the reactor building airlock could be the SVR's operation. It drains inventory from RPV, which is in the drywell, to suppression pool which is out in the room with less thicker concreate walls than drywell has. So the radiation from the thorus inventory could affect the measurement at the airlock.

Without denying written above, as a simpler explanation may be the lack of ventilation in the reactor building. Did you know if you shut the vent hole in the bathroom and turn on the shower, radiation inside rapidly increases up to 2 mkSv due to Radon? Such was certainly here but in a more rigid form.

Everything about the attempt to enter the reactor building rests on the fact that the dosimeters were not suitable for this job, even there wasn't any damage to the equipment inside the building, and certainly if there was. Why then the report does not blame operators for this? This is a real bug, unlike the lack of knowledge of the fail-safe automatic functioning imposed by the report.

 

[jim hardy]

we might be over-focused on dosimeters. i don't know what they had.

surely they had survey meters good for a few R/hr?

the Inpo report mentions water deeper than their boots, i'd have turned back too.
water can carry reeally nasty stuff, recall the electrical workers who got radiation burns from walking through water. presumably dissolved beta emitters because field above water wasn't too bad.

decisions have to be judged by the facts they were based on, ie what was known at that time & place.

i'd wager too much credence was put on the reactor level instrument.

 

[rmattila]

i'd wager too much credence was put on the reactor level instrument.

I've been wondering about the level measurement systems since the very beginning, even started a thread in March to get some justification for the trust that seemed to be put on the level instrumentation.

A rule of thumb is that 10 kg/s decay heat evaporation rate (typical about 1 hour after shutdown) will cause the level in the tank to drop about 3 cm per minute (=about two meters per hour), and as there was no cooling for several hours, it should have been obvious that the core will have been uncovered during the evening of March 11.

(And, by the way, I am still somewhat uncertain whether the GE level instrumentation will work at all, when the level drops below -1200 mm).

Diversification of the low level instrumentation with a binary signal based on floats is one way to improve reliability of data concerning the level drop in situations where impulse pipes might not work properly (rapid depressurisation etc.) My understanding is that such plant modifications/improvements have already been implemented at some European BWRs, and are currently being carried out at least in the Finnish plants.

 

[Pakman]

decisions have to be judged by the facts they were based on, ie what was known at that time & place.

It's a fact that at that time&place the reactor cooling status was unknown. Is it not enough to enter the reactor building by any cost? Especially because there was no high radiation alarm to the reactor building rooms after quake and before SBO.

By the way, abut deep contaminated water it was in the turbine hall.

 

[zapperzero]

It's a fact that at that time&place the reactor cooling status was unknown. Is it not enough to enter the reactor building by any cost? Especially because there was no high radiation alarm to the reactor building rooms after quake and before SBO.

By the way, abut deep contaminated water it was in the turbine hall.

Well they attempted to enter the building after the tsunami, when they had lost all power so I'm not sure those dosimeters that could have triggered an alarm were even online.

 

[Pakman]

Right because of this I wrote above there was no radiationn alarm BEFORE SBO.
You better read exactly what is written.

 

[zapperzero]

Right because of this I wrote above there was no radiationn alarm BEFORE SBO.
You better read exactly what is written.

I did read.

No radiation alarms implies no major pipe breaks or leaks before the tsunami, I guess. How does that tell me anything about the radiation field in the reactor building hours later?

 

[r-j]

What does SBO stand for?

 

[swl]

SBO = Station Black Out.

Loss of power to the power plant.

 

[Joffan]

What does SBO stand for?

Station black-out. No electrical power either from external sources or from site backups. I think it is usually limited to mean AC power availability but I believe in the case of Fukushima many DC sources (batteries) also failed (before exhaustion). Real experts here may elaborate on (or contradict) this response.

 

[NUCENG]

Station black-out. No electrical power either from external sources or from site backups. I think it is usually limited to mean AC power availability but I believe in the case of Fukushima many DC sources (batteries) also failed (before exhaustion). Real experts here may elaborate on (or contradict) this response.

Correct. SBO starts with a Loss of Offsite Power (LOOP) and failure of the Safety Related Emergency Diesel Generators (AC). In the case of Fukushima the Offsite AC supplies were lost during the earthquake Onsite EDGs failed due to flooding from the tsunami. Damage occurred to the fuel tanks and flooded some of the EDG rooms in the Turbine Building basements. Finally some of the AC switchgear which would allow EDG power to be applied to the safety equipment were also flooded.

However Fukushima accidents went one step further and some of the DC switchgear for the DC batteries were also damaged. So in that respect it was a beyond SBO accident.

 

[matthewdb]

BWR has quite contaminated steam-water circuit loop, including the turbine flow, so the turbine set in maintenance-free room with airlock to outside. Almost like secondary containment of the reactor building. Therefore one doesn't simply walk into turbine hall.

No, not even close. The airlock is for going into primary containment. The secondary containment and the turbine building are ventilated to the outdoors.

The turbine is not normally approached under power due to gamma from N-17. It has a bioshield around the turbine and main steam lines but the rest of the turbine hall can be entered under power.

 

[Pakman]

No, not even close.

Ok, I just guessed. So it needed to change "airlock" to "tightly closing door" in what I've written above. But then the talking was about the reactor building.

The secondary containment and the turbine building are ventilated to the outdoors.

Through the vent stack, I believe? What happens with level of radiation inside the building when ventilation stops?