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

Wednesday, April 20, 2011

Some technical and time details at Fukushima Daiichi

We've been waiting for a LONG time now to see some actual technical details emerge about just what kinds of systems the plants at Fukushima Daiichi had available for decay heat removal after the total loss of power (SBO) when the tsunami hit. It should be remembered that these plants don't have to subscribe to NRC rules that have for a number of years now covered plant modifications to mitigate SBO sequences leading to accidents; as a result, the systems installed at Fukushima Daiichi cannot be expected to be identical to any US plants of similar vintage. In fact, most US plants of nearly identical vintage differ between themselves in some respects, except of course for the mandated backfits dictated by the NRC.

The following chains of events include known facts, and interpretations. It's still too early to have all the information, but we can start to try to figure out some actual events that occurred in the first few days.

With that in mind, we now have available a message from the Japan Nuclear Technology Institute that gives us a couple of mechanical hints. This letter, available at the JANTI site is really intended to be a stern admonishment to the senior nuclear establishment in Japan (primarily the NSC) to get on-site and get contributing.

Anyway, the letter mentions the fact that No. 1 plant was using an isolation condenser system for decay heat removal, and that this system functioned for half a day before not being operable any further, and also indirectly states that this is longer than it was designed to operate for. This is a very old system for removal of decay heat when the core is still pressurized, and requires only DC power for valve operation in the short term, although according to a report we have here by Brookhaven National Laboratories the isolation condenser makeup system must be operable within about an hour to maintain water level. This same report implies that, and we're making an assumption here, if the plant is in its original configuration AND that configuration matches most plants of that overall design then the only other source for injection at high pressures is the FWCI or Feedwater Coolant Injection System that requires AC power. Loss of DC power will remove operability of the isolation condenser system as well. Decay heat removal at high pressure is thus limited in several ways if AC power is lost.

Looking at the timeline for No. 1 plant we see the report of loss of normal core injection (mandatory report) at 16:36 on the 11th. Containment vessel pressure started rising abnormally just after midnight the 12th; surely there was relief valve lifting during this time. At 14:30 that afternoon TEPCO started venting the containment to atmosphere but one hour and six minutes later a hydrogen explosion destroyed the building. At 20:20 that evening TEPCO began injection of seawater to the core on this plant. After this, primary pressure began to drop markedly for the first time. (Ref: JAIF / JANTI charts)

We might then assume that the isolation condenser system did its job at least temporarily; the JANTI letter seems to imply that the system functioned for 12 hours ("half a day.") We might guess that this would roughly correlate to battery power, so we looked for any relevant data we could find about how long they might last in this condition.. not knowing the actual capacity of this plant's batteries. A study performed covering the Browns Ferry plant by Oak Ridge National Laboratories indicates, for that plant, a battery life of about six or seven hours. That time.. six or seven hours.. roughly correlates to the increase in CV pressure, so that we might wonder if the operators, knowing loss of the isolation condenser system was approaching might have depressurized the plant deliberately to allow LPCI. We will have to wait until an official narrative appears.

Now, we turn to No. 2 and No. 3 plants, which are of a newer design. The JANTI letter indicates that these plants were using steam turbine driven coolant injection systems; these functioned for 3 days at Unit 2 and for 1.5 days at Unit 3. The available graphs don't give good data for pressure vessel pressures for these two plants early on; what we can say is that No. 3 plant started venting its containment at 20:41 on the 12th, which is BEFORE it reported a loss of cooling function. That occurred at about 5 AM on the 13th. At No. 2 plant, venting of the containment also started before the report of loss of cooling function; No. 2 started venting at 11 AM on the 13th, with the loss of cooling report coming at 13:25.

One has to guess that operation of the RCIC system, which is what we assume the letter to indicate was in use, eventually failed due to either loss of all water supply, or loss of battery power. By early afternoon on the 13th No. 3 was receiving seawater injection from external sources, and No. 2 followed a little over 24 hours later.

SO, NOW that we've been through all that, what do we glean?
-Battery failure is probably the next aggravating event in the fault tree after loss of all AC power, at least at No. 1 plant.
-It appears that the emergency cooling systems functioned longer than the owners / operators and nuclear establishment in Japan had predicted they would.
-The systems at No. 1 plant differ considerably from those at No. 2 and No. 3 plants, so that further safety implications to other plants world-wide as more details emerge will have to be carefully applied to / compared with only the appropriate plants. (For reference again, No. 1 plant is a BWR/3 reactor while No. 2 and No. 3 are BWR/4 reactors.)

Again.. we don't have a good handle on what the present battery capacity on the plants at Fukushima Daiichi is. We suspect at least on No. 2 and No. 3, looking at the events, that it's considerably longer than at No. 1. That will be a major factor in examining the sequence of events, in a causative way. What we're trying to do here is take the few new facts we've been given, and start down the path to understanding those critical 24 to 48 hours in which the whole situation at Fukushima Daiichi turned from unusual event to accident. We will provide updates and corrections to the information as new data appear.

1:00 PM Eastern Wednesday 4/20


  1. hey just heard latest news that 520 ton radioactive water is flown into sea ,is that correct?its told that 20000 times the dangerous level.is that true?

  2. @machlipatnam: Yes, that's the short term total from about April 1 onwards, if I understand the press release information correctly. However, the concentration is of the released water, and the concentration of course dilutes rapidly off shore. There will be more data to come as TEPCO conforms to government orders to quantify the releases.