APR: your source for nuclear news and analysis since April 16, 2010

Wednesday, May 25, 2011

Fukushima Daiichi details... Wed. May 25

There are two news stories of interest coming out of Japan at the moment. We'll tackle them in our usual fashion.

First, TEPCO is now reporting failures at the bottom of all three reactor pressure vessels; this is being carried for the moment only on the major Japanese news media and has not appeared in TEPCO's press material as of yet. I reported on this blog a very long time ago the high percentage of likelihood during a long-term SBO event that damage at the bottom reactor pressure vessel head would occur. I believe I even mentioned this on the radio interview that Fintan Dunne performed, wherein I was a guest. It should be noted that TEPCO is reporting equivalent leakage from a single theoretical hole, and that the real failures are probably around control rod drive penetrations, from some cracks, and maybe around pipe penetrations so that there isn't one single large hole but a scattering of smaller failures.

A long time back this blog was the first to introduce to the general public the 'inside' term of "feed and bleed" to describe what was occurring at Fukushima Daiichi; that essentially being the cooling of the cores by continuous injection of water from new sources, with water loss occurring at some or other point or at multiple points. This is of course different from recirculating cooling, where water is moved in a loop. So, then, for readers of this blog this is not an entirely new revelation in the sense that for a long time we've been wondering exactly how the water was getting out of the reactors.

We knew where it was going long before this revelation (if it really is revelation) by TEPCO. First, we saw contaminated water in the turbine buildings. Then we saw it in trenches outside, and in the seawater. Of course, early on everyone suspected it was getting to these places after having started in the reactor buildings unless there were piping failures in the turbine building that were releasing water from the reactors (one example, discounted early, could have been stuck open MSIV or Main Steam Isolation Valves.)

In summary, then, it's not really a surprise, given the known facts about long-term SBO accidents (core damage in hours, core melt through pressure vessel as a possibility starting after ten hours or so, which varies by plant and by actual timeline of events) and the known release of water if there is in fact serious failure at the reactor pressure vessel lower heads, or possibly the area of the vessel just above them. TEPCO performed a number of steps early on that made us suspect that TEPCO suspected serious core damage; for example, the very early drywell flooding is something you see when the operators are trying to prevent or delay failure of the pressure vessel by its contact with melted fuel. Only later did drywell flooding come about again, this time as a part of the attempt to fully cool the cores; early on it was an accident mitigation strategy and not a recovery strategy. Events at Fukushima Daiichi, studies done over many years (primarily by ORNL), and operator actions at Fukushima Daiichi have pointed to serious failure in the pressure vessels or piping for some time now.

Now, the second news item; TEPCO is reporting data that indicates that there may have been a pipe failure at No. 3 plant in the HPCI (High Pressure Coolant Injection) system. TEPCO is denying at the moment that the earthquake caused a pipe failure; in fact, given the data we have right now we cannot assure that the HPCI system was the failure point. It appears that initiation of the HPCI system occurred shortly before a major drop in pressure in the primary plant. Some writers are saying that this is de facto proof that there was a pipe failure in the HPCI system, and are assuming the earthquake caused it. However, until the system itself is inspected this is not a sure bet; for example, no one has ruled out a stuck open relief valve or any other such possibility. (There are some anomalies in data recorded on relief valve tailpiece RTD's which we've noted here once or maybe twice; these are what you would use to detect a stuck open relief, or a relief that lifted at all for that matter.)

The reason this is important theoretically is that if it were true you would then have to consider the plant also in a LOCA scenario (Loss Of Coolant Accident) in combination with the SBO scenario. This seems a small difference but is a major difference in the analysis of how to handle the situation, and how to prevent an occurrence in the future. But before we jump to conclusions we need more solid data from TEPCO who, to repeat, is not saying that the earthquake damaged the system as of yet. We will certainly keep an eye out for clarification on this issue.

More to come later on today.

12:45 PM Eastern Wednesday 5/25

No comments:

Post a Comment