Japan Earthquake: nuclear plants Fukushima part 2

[Hiddencamper]

I'm not sure where the actual report is, but apparently Japanese regulators determined on their own that the tsunami was the cause of the accident at Daiichi, and the earthquake did not cause the event.

http://www.neimagazine.com/news/newsdaiichi-accident-caused-by-tsunami-not-earthquake-japanese-regulator-concludes-4322346

If anyone can find this report, please let me know!

 

[jim hardy]

I agree - the high water killed the electrics which pretty well doomed the plant.
It would have survived the earthquake with no tidal wave
but not the tidal wave with no earthquake.

Events followed reasonably closely a "Loss of all AC Power" scenario hypothesized years ago by Oak Ridge.
http://web.ornl.gov/info/reports/1981/3445600211884.pdf

 

[Sotan]

I read about that too, on several sites.

I believe the reports of the Japanese regulator (NRA) are those from http://www.nsr.go.jp/committee/yuushikisya/jiko_bunseki/20140718.html

(Unfortunately, they are only in Japanese.)

The first two of them (1-1 and 1-2) are about the causes of the loss of electric power in Unit 1.

The next two (2-1 and 2-2) contain a more general evaluation of the nuclear accident at Fukushima Daiichi (they actually contain the 7 items researched by NRA which are mentioned in the link given above by Hiddencamper in post #527 (1. Minor water leaks at unit 1, 2. Loss of function in emergency power supply at unit 1, etc., up to 7. Hydrogen explosion at unit 4.)

The last report (2.3) is a short material called "About the method for further research in the future".

As translating such materials in full would require more time than I have (I haven't even perused them properly at this moment), I can only hope they will be translated in English at some point in the near future. Or, as always, if you have more punctual points of interest, I could try to find those portions and attempt to present a summary.

 

[Sotan]

Today I took a look at those reports and I have to say that the short newspaper article linked by Hiddencamper actually can serve as a good summary - at least for non-specialists like me. The reports are simply much more documented, they take each item and discuss it in relation with the earlier findings of the Parliament commission that investigated the accident, they list the evidence and discuss it, but the conclusions given in the newspaper articles actually do convey the main content.

Also, I have to confess that the first two reports are extremely technical and, because they refer to electrical issues, I really don't understand much (I only have a mechanical engineering degree).

---------

But: the main reason I post this time is to say I am impressed by this new source of information I found while searching for the report wanted by Hiddencamper: the NRA site, the homepage of the Japanese nuclear regulating authority.

Every two weeks or so they have a meeting, for which they post the schedule, the discussion agenda, the materials used for those discussions, a detailed record of all the talks, as well as a video recording on Youtube. http://www.nsr.go.jp/committee/yuushikisya/tokutei_kanshi/

I browsed through the minutes of the meeting of June 6 (the last available - it seems that the minutes are posted a good while after the meeting is over) and it was actually very interesting. TEPCO has several representatives who that attend these meetings and who know what's in those materials they prepared. They are the first to speak, and they take each material which has been prepared for the meeting, and present it in detail, explaining what is on every page, every diagram. Then the invited experts and NRA members express opinions or ask questions, to which TEPCO representatives answer. This goes on until all topics are approached. The June 6 meeting was about TEPCO's progress in building the ice wall.

For somebody who really follows the events, these minutes are probably invaluable. Many of the facts that transpire in only several words in newspaper articles and news could be much better explained based on materials and discussions held here. Specialized journalists are probably reading these and then conveying what they can, but they probably have to deal with space limitations, plus how much of the public is really interested in details.

For example, I remember at some point that 1) some people were worried how will the ice wall affect the stability of the terrain (won't the soil sink?) and therefore the buildings in the Fukushima Daiichi complex. 2) TEPCO replied that they are confident such effects, if any, will be negligible. 3) people expressed concern regarding how well grounded was this conclusion and what were their arguments. Well, among the materials of the NRA meeting of May 26 there's an http://www.nsr.go.jp/committee/yuushikisya/tokutei_kanshi/data/0022_02.pdf that deals exclusively with this issue. Of course I cannot say I read and understood it all and I can say anything about the scientific conclusions they drew - but it does show that they are at least trying to do a serious job.

This was just one example... I still don't know how to deal with the amount of information, but I will browse through it and post when I find something interesting.

 

[Sotan]

And again I jump to something else (apologies!)

I found this very instructive video which some of you might have missed.
It does have some rather unrelated portions, but the first 20 minutes or so present the works for opening Reactor 5 at TEPCO's Kashiwazaki-Kariya plant for scheduled maintenance.
Also, at min. 46:40 or so there's a short appearance of Fukushima Daiichi, as they are restarting a reactor after maintenance.
It's been shown on TV first in 1994.
It's only in Japanese, but still, worth a look, at least it was for me.

 

[Sotan]

NHK English article of Jul 23:
http://www3.nhk.or.jp/nhkworld/english/news/nuclear.html

 

[Sotan]

Another NHK English article, Jul 23:
http://www3.nhk.or.jp/nhkworld/english/news/nuclear.html

These NHK reports are actually good summaries of the presentations made by TEPCO in the NRA meetings.

 

[zapperzero]

A some of you might have missed.

Don't think I've seen it before. Thanks.

 

[Sotan]

I haven't read the NAS Report itself, instead I found this comment by Dave Lochbaum, on the UCS site:

http://allthingsnuclear.org/nas-report-lessons-learned-from-fukushima-nuclear-accident-for-safety-of-u-s-nuclear-plants/ (July 25, 2014)

Worth a look.

 

[Sotan]

IMHO it can be about the usual papermill of the permissions and components. It might be weeks to get everything ready for the work - while that cooling would run on reserve, without any further reserves.

With that patchwork it'll have the 'main' and the 'reserve' for all the time of the papermill.

I would be worried only if the proper replacement would not be performed within a month or two.

Follow up on this issue:
http://www.tepco.co.jp/nu/fukushima-np/handouts/2014/images/handouts_140730_05-j.pdf (in Japanese)
They took out the valve and discovered it was deeply corroded on the inside.
It originally had some (rubber?) lining on in the inside, that was mostly gone, plus the metal was deeply corroded too, hence the piercing of the valve body and the 3 mm diameter hole.
The cause they indicated is the high speed of water through the valve in that region; the valve was kept open only 30% for a long time, a rather unusual regime, which led to high fluid speed in that area. That damaged the rubber lining and opened the way for corrosion. They replaced this valve with the similar one from Fukushima Daini plant Unit 3 which was in good condition. They will check this valve more often in the future.

They dismantled another similar valve in the vicinity of this one, and they found some damage to the rubber lining; they repaired that and put the valve back, but they plan to replace this one too, with a new one, in the future.

The cooling of the SFP at Unit 5 in Fukushima Dainichi has thus been restored using the usual circuit.

 

[Hiddencamper]

TEPCO releases their next report on unexplained occurrences at Fukushima Daiichi

http://www.tepco.co.jp/en/press/corp-com/release/2014/1240140_5892.html

(if you recall, the first major report came out in dec 2013 http://www.tepco.co.jp/en/press/corp-com/release/2013/1233101_5130.html)

The key findings from this report: Unit 3 RCIC likely tripped due to high turbine exhaust pressure. Unit 3's HPCI likely stopped injecting earlier than originally expected due to low steam supply pressure, which meant that fuel damage likely started a few hours earlier and that there was greater damage at unit 3. This may help close some of the gaps as to why unit 3 had so much hydrogen compared to units 1/2.

 

[Sotan]

Good and quick find, Hiddencamper - that English summary report.

In case somebody is interested, the full report and the load of related materials are available at http://www.tepco.co.jp/cc/press/2014/1240099_5851.html

But they are only in Japanese.

 

[zapperzero]

The report also determined that the detected presence of neutrons near the facility's main gate on March 13-14 was the result of spontaneous nuclear fission of easy- to-decay (short-life) actinides in Units 2 and 3 as the result of the fuel melting, and not a return of the reactor core to recriticality (sustained nuclear fission).

I would like to see the reasoning behind this. Also, I don't recall anyone claiming that sustained nuclear fission was taking place at that time. That seems an odd thing to disprove - what might need disproving is the claim that there were short fission bursts taking place because of the core debris being (re)flooded.

Too, the actual summary report does not use nearly the same assured tone, to wit:

"It is possible that intermittent increases in nuclear pr
essure after forced depressurization at Unit 2 (issue list
"Unit 2 - 7") may have been due to hy
drogen and water vapor discharged while
fuel was melting. In conjunction,
the neutrons observed near the main
gate may have been generated by th
e spontaneous fission of actinides
discharged while fuel was me
lting at Units 2 and 3. "

http://www.nrmc.jp/en/report/detail/__icsFiles/afieldfile/2014/08/06/er9.pdf
has more detail

It's funny how they claim there is no correlation between gamma ray spikes and neutron spikes, but then on page 28 they show a gamma+neutron spike starting at or near the time when water injection from fire engines starts.

 

[mheslep]

I would like to see the reasoning behind this. ...

I dunno, but the gamma signatures from the fission of generated actinides, versus that from critical U235 would be like a fingerprint.
http://www.lanl.gov/orgs/n/n1/panda/11. Nuclear Data for NDA.pdf

 

[jim hardy]

I would like to see the reasoning behind this.

I remember being glued to the computer back when this was going on. (I'm retired from a plant so was very interested).

If i recall
that indication came from neutron detectors at the entrance gate . That seemed strange to me for in my plant we didn't have any way out there . The level was extremely low, almost individual counts.
It was about the same time as water was being added and some venting, to best of my recollection.
Wind was in the direction from plant toward entrance gate - that i do remember checking.

so at the time i assumed it was some neutron emitting fission fragments carried out through the vents, generated when water hit hot and crumbly fuel elements. Or else hot particles on somebody's muddy boots.


Does that jibe?

 

[Hiddencamper]

Consider that unit 3 underwent an automatic depressurization system (ADS) blowdown, and with this new data, we know the fuel was already damaged/melted when ADS actuated. The ADS system would have discharged a lot of fission products and noble gasses to the suppression pool.

Unit 3 did have containment venting a short while after the ADS blowdown:

http://www.cas.go.jp/jp/seisaku/icanps/eng/03IIfinal.pdf

Although depending on specifically what you read, its possible that venting may not all have gotten to the stack. Some of it may have went to unit 4, some may have gone into the reactor building. The venting was from the suppression chamber. This would have created a direct pathway for radioisotopes to migrate from the damaged fuel, out the safety relief valves to the suppression pool, and out the suppression pool to the environment.

This isn't a sure thing, just an idea looking at what we know.

 

[zapperzero]

I dunno, but the gamma signatures from the fission of generated actinides, versus that from critical U235 would be like a fingerprint.
http://www.lanl.gov/orgs/n/n1/panda/11. Nuclear Data for NDA.pdf

I am not sure the gamma detectors they had were all THAT sensitive/discriminating.
Too, one does not exclude the other.

 

[mheslep]

I am not sure the gamma detectors they had were all THAT sensitive/discriminating.
Too, one does not exclude the other.

I only know detectors that can discriminate gamma (obviously) exist. No information what was on site.

 

[etudiant]

The recent TEPCO statements do seem to provide a coherent sequence for the failure of the three reactors.
Clearly the reactor systems functioned largely as designed and postponed the meltdowns, but not long enough for the operators to regain control. This raises the question whether anything could have averted the disaster once the tsunami had struck, had the operators had better knowledge.

 

[jim hardy]

This raises the question whether anything could have averted the disaster once the tsunami had struck, had the operators had better knowledge.

With hindsight there's some things that could have been tried, for example ignoring the vessel's cooldown limits and cooling the reactor very quickly to below 100C..
Nobody knew how big the tidal wave would be. Who'd thermal shock a reactor vessel for a "maybe" ?

But without electric power they were trapped . Lots of heat to get rid of and no way to move it.

Probably they'd have been able to only postpone things a matter of hours or very few days.

just my thoughts...

old jim

 

[Hiddencamper]

I think, if they did not lose DC batter/control power at units 1/2, then maybe, they might have been able to do something.

24 hours should be more than ample time to restore power or at least implement a decay heat removal solution. But they spent so much time dealing with no reactor indications, trying to figure out the state of their cooling systems, the inboard valves for the IC at unit 1 appears to have isolated on loss of control power, and shortly after that they were dealing with core damage, high rad levels, explosions, personnel shortages.

Seriously, I think if they had control power at units 1/2, they would have kept core cooling capability of the unit 1 IC, and rather than spend time trying to figure out the status of unit 1/2, they could have devoted resources to an ultimate heat sink solution similar to the Daini site.

On a separate note, the Us industry does not assume total DC control power failure for any scenario, including the new Fukushima based scenarios or for the AP1000. I think this is a significant gap, and that there needs to be training on how to deal with this. The extensive damage mitigation guidelines do have steps for performing specific tasks without control power, like hooking batteries up to instruments to get readings, but a total cd power loss would be extremely challenging to deal with as you would be blind to the majority of your indications.

Just my thoughts.

 

[jim hardy]

The extensive damage mitigation guidelines do have steps for performing specific tasks without control power, like hooking batteries up to instruments to get readings, but a total cd power loss would be extremely challenging to deal with as you would be blind to the majority of your indications.

That's the one plant guys have nightmares about. Without DC you can't start your diesels, either.

At my plant we knew just where to connect engine driven DC welders. But you're right it wasn't in procedures . Well, more correctly it wasn't when i retired twelve years ago.

 

[jim hardy]

Hmmmm they're working on it.
http://www.gpo.gov/fdsys/pkg/FR-2012-03-20/pdf/2012-6665.pdf

golly I'm out of date ! It even mentions coronal mass ejection...

 

[Hiddencamper]

We can start our DGs once without control power. I actually simulated the emergency start procedure today for one of my license qualifications. But we are just one plant. Manually open the air start solenoids, while you have a guy standing by the generator breaker and service water breaker to close them in and at least pick up the bus. Let the mechanical governor hold the engine rpm until you can bring the battery charger on service.

Obviously it's a very complicated procedure. If 9/11 and Fukushima brought anything good, it's the fact that we actually have this procedure and others like it, and we train on it.

 

[jim hardy]

Manually open the air start solenoids,

Now THAT's interesting.

I wonder if they've added such a bypass at my prior plant.

Our diesels were air cooled by huge radiators and the fans driven mechanically by engine shaft so they needed no service water
had hours of fuel available by gravity feed day tank
the air start reservoirs were sized for five start attempts
the starting air electric compressors could run from either bus
and we had a small one cylinder diesel compressor to recharge the start reservoirs if things ever got that bad...
...but we needed battery.

How'd you flash your field? Pilot exciter?

 

[Hiddencamper]

Engineering evaluated it and then we later did a test on a generator to show it works. There is enough residual magnetic field for one flash. It's not pretty but it can be done.

 

[nikkkom]

On a separate note, the Us industry does not assume total DC control power failure for any scenario, including the new Fukushima based scenarios or for the AP1000. I think this is a significant gap

I suppose I should feel happy. Only 3 years past Fukushima and some in US nuclear industry start to think they need to be prepared for SBO.

I guess only short 20 years to wait more and US plants will actually have procedures in place to survive SBO. Just about time when most of them will be closing down.

(angry unproductive rant mode off)

 

[jim hardy]

Is effort better spent on preventing it?

Either way, one is preparing for "what if's".
Those "what if's" are assigned likelihoods by something called "Probabilistic Risk Assessment" (PRA) and dealt with in their order of probability. Those below some threshold are ignored.

So the key is a good and honest set of PRA "What If" risk assessments.
When those 900 year ago tidal waves in the region were discovered the PRA didn't get adjusted for them.
Those basements housing diesels and switchgear should have got wrapped in something like a submarine hull, or moved up the hill.
But they didn't.

If you want to make something a certainty just publicly declare it an impossibility.
I guess that's a strength of PRA, nothing is considered impossible .

When PRA was first getting started some wag at our plant wrote a procedure for "loss of gravity" and submitted an engineering request for springs to push the rods into the core, and seatbelts for operators. We all enjoyed a good laugh...
He was an ex -submariner. Submarines can turn upside down... so their "loss of gravity" weighting in PRA reflects that fact.

Point being - with the plants we have that are designed around the concept of emergency power being available, i'll choose even a pound of prevention over the unforseen consequences of a necessarily piecemeal redesign.
After all - we only need twenty more years out of them.

old jim

 

[jim hardy]

A step forward in passive system thinking..
especially the middle video

http://nuclearstreet.com/nuclear_po...n-the-ap1000-reactor-081302.aspx#.U-uaFWdASih

 

[etudiant]

DC power was also an imperative for the Bell phone system, so they used large, custom designed round lead acid batteries. These gave very long and very reliable service. I'm surprised the Bell know how has not diffused into the nuclear industry.

 

[Rive]

DC power was also an imperative for the Bell phone system, so they used large, custom designed round lead acid batteries. These gave very long and very reliable service. I'm surprised the Bell know how has not diffused into the nuclear industry.

It's not about the knowledge needed for building such batteries but about the knowledge that those batteries should be built.

 

[jim hardy]

We had four of them each somewhat larger than this:

larger image here:
http://assets.safetyfirst.nei.org.s3.amazonaws.com/wordpress/wp-content/uploads/2011/11/IndianPointBatteryPower.jpg

 

[mheslep]

We had four of them each somewhat larger than this:

larger image here:
http://assets.safetyfirst.nei.org.s3.amazonaws.com/wordpress/wp-content/uploads/2011/11/IndianPointBatteryPower.jpg

Would love to drop a crow bar across those battery terminals.

 

[Astronuc]

I suppose I should feel happy. Only 3 years past Fukushima and some in US nuclear industry start to think they need to be prepared for SBO.

I guess only short 20 years to wait more and US plants will actually have procedures in place to survive SBO. Just about time when most of them will be closing down.

(angry unproductive rant mode off)

The US nuclear industry has been thinking about station blackout (SBO) for decades.

Reevaluation of Station Blackout Risk at Nuclear Power Plants (NUREG/CR-6890) - 2005
Note: Vol. 1, Analysis of Loss of Offsite Power Events: 1986-2004 (PDF - 10.97 MB)
◦Vol. 2, Analysis of Station Blackout Risk (PDF - 5.93 MB)
◦Vol. 3, Resolution of Comments (PDF - 3.87 MB)

http://www.nrc.gov/reading-rm/doc-collections/nuregs/contract/cr6890/

This report is an update of previous reports analyzing loss of offsite power (LOOP) events and the associated station blackout (SBO) core damage risk at U.S. commercial nuclear power plants. LOOP data for 1986–2004 were collected and analyzed. Frequency and duration estimates for critical and shutdown operations were generated for four categories of LOOPs: plant centered, switchyard centered, grid related, and weather related. Overall, LOOP frequencies during critical operation have decreased significantly in recent years, while LOOP durations have increased. Various additional topics of interest are also addressed, including comparisons with results from other studies, seasonal impacts on LOOP frequencies, and consequential LOOPs. Finally, additional engineering analyses of the LOOP data were performed. To obtain SBO results, updated LOOP frequencies and offsite power nonrecovery curves were input into standardized plant analysis risk (SPAR) models covering the 103 operating commercial nuclear power plants. Core damage frequency results indicating contributions from SBO and other LOOP-initiated scenarios are presented for each of the 103 plants, along with plant class and industry averages. In addition, a comprehensive review of emergency diesel generator performance was performed to obtain current estimates for the SPAR models. Overall, SPAR results indicate that core damage frequencies for LOOP and SBO are lower than previous estimates. Improvements in emergency diesel generator performance contribute to this risk reduction.

And we reassess when some major event occurs.

I remember covering LOOPs about 30+ years ago, and it was not a new concept then, but something that had been around since at least the 60s.

http://nrcoe.inel.gov/resultsdb/LOSP/

Loss Of Offsite Power
It is recognized that the availability of ac power to commercial nuclear power plants is essential for safe operations and accident recovery. A loss of offsite power (LOOP) event, therefore, is considered an important contributor to total risk at nuclear power plants. In 1988, the NRC published NUREG-1032 to report on an evaluation of the risk from actual LOSP events that had occurred at nuclear power plants within the United States up through 1985. NUREG/CR-5496 documents a similar study whose primary objective was to update the LOOP model parameters, frequency, and recovery time, using plant event data from 1980 - 1996. An additional objective was to re-examine the engineering insights concerning LOSP events.

http://www.nei.org/Master-Document-...gh-the-Decades-History-of-US-Nuclear-Energy-F (only back through 1992)Clearly there were a number of deficiencies at the Fukushima Daiichi site. It should have been clear that the region was at risk of a 30 m tsunami or so - because it had happened before. On the other hand, much of the coastline north and south was not protected against such a tsunami.

http://en.wikipedia.org/wiki/1964_Alaska_earthquake
http://wcatwc.arh.noaa.gov/about/64quake.htm

Leaving the diesel generators and oil storage tanks so vulnerable is to me mind boggling.

Leaving the switch gear in a vulnerable location is also mind boggling.

Ignoring studies that indicated the need to identify such vulnerabilities is unconscionable.

 

[jim hardy]

Would love to drop a crow bar across those battery terminals.

Somebody at another plant got an aluminum ladder across their battery... once a DC arc is started it's hard to extinguish. Made a real mess of that battery room.

We didn't allow aluminum ladders on the property.

But it would be exciting from a safe distance.