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

Thursday, October 6, 2011

Davis-Besse vessel head replacement

As reported on the NRC blog, Davis-Besse has been shut down for replacement of its pressure vessel head. This plant is a Babcock & Wilcox pressurized water plant which went into commercial operation in 1978.

This plant has had extremely bad press because of a deficiency discovered during an NRC mandated inspection of its pressure vessel head. Below, an illustration of a plant like Davis-Besse (Babcock & Wilcox raised loop) showing the location of the pressure vessel head - a heavy, bolted lid which seals off the upper end of the reactor pressure vessel and through which are many holes to allow for installation of control rod drive mechanisms and for instrumentation.


Background: After a number of licensees (that is, utility companies operating nuclear plants) had reported cracking in vessel head penetrations to the NRC, the NRC issued a bulletin in 2001 which was bulletin 2001-01 calling for data concerning the various licensees' inspection programs to detect this type of possible failure. Davis-Besse addressed the requests of this memo during its first available refueling outage in February 2002, at which time a serious flaw was found in the vessel head which had been caused by a boric acid deposit. This threatened the integrity of the vessel head, and much bad press surrounded the discovery of this defect. The head could not be saved, and another had to be either made or found.

Fortunately -- at least, temporarily -- for Davis-Besse, there was a practically identical head available, unused, in Michigan. Consumers Power Company's Midland nuclear station, a two-unit Babcock & Wilcox plant, had been cancelled back in the middle 1980's due to a downturn in power demand generally and, specifically, a large number of on-site problems (diesel generator building sinking, improper welds in No. 1 plant's vessel, cracked support bolts for the reactor vessels, and not least, a well publicized $120,000 fine against Consumers Power for deliberately violating quality assurance requirements and for design and construction code violations.) This plant still had on site one vessel head which Davis-Besse could acquire and apply to its plant. However, since this head was of the same material as the original, it was susceptible to all the same potential corrosion problems as the first, so that Davis-Besse was required also to order a new head that would be applied later on.

Below, the unfinished (and actually never finished) Midland nuclear power plant as it looked in May, 1978. Press photo in APRA collection.



To make a long story short, the former-Midland head went on in 2004, and six years later upon inspection showed a few tiny cracks. Not serious ones, but cracks nonetheless. This was probably not particularly surprising to many in the industry. These cracks (which were repaired) led FirstEnergy to move the head replacement up from 2014 to the present time.

The NRC blog notes that over half of the pressurized water reactors in the United States have actually had their pressure vessel heads replaced. Thus, this is by no means an unusual procedure.

I will be making another post soon covering some details of pressure vessel construction that might be of interest as background for this type of maintenance.

4:36 PM Eastern Thursday October 6, 2011
ATOMIC POWER REVIEW

5 comments:

  1. As I remember, the Midland plant was designed not only as a power generating station, but it was also supposed to supply process steam to a nearby DOW chemical plant. This would have been quite unique in the industry. Too bad the management of the project was apparently less than adequate.

    How is the new Davis-Besse reactor head different than the previous one? You mentioned the Midland head as being of the same material as the first. So the new head is of different material? The first head had a stainless steel liner on the inside. The boric acid in the coolant leaked through this liner, and then the acid ate away at the carbon steel head. Is the new head all stainless steel or something?

    ReplyDelete
  2. @t.: You're familiar with Midland? Interesting; most folks aren't. Yes, that was one of the designs that would have supplied steam not directly for power generation and it was indeed to a DOW plant. It appears that Consumers Power sure didn't assemble the right group to build this plant... and picked the wrong time to do it.

    The first two heads were SA 533B steel, clad inside with 308 stainless, according to NRC records. The head nozzles were Ni-Cr-Fe Alloy 600 which of course is better known as Inconel 600. The coolant leaked though cracks in the Inconel stub tubes, not through the 308 clad. Suspect also was the weld material sealing the stub tube penetration through the clad surface, which was Alloy 182.

    The only confirmed detail I can find from FirstEnergy is that the stub tubes (nozzles) are of a different material than Inconel 600 which is not susceptible to stress corrosion cracking. The NRC's present blog says: (A brand new head made from a different metal that is much less susceptible to corrosion was originally supposed to be installed in 2014.)

    ReplyDelete
  3. The link below (near the bottom of the page) says: "The longer term solution for many plants has been to seek replacement using Alloy 690 base metals and Alloy 52 and 152 for weld filler metals."

    http://www.nrc.gov/reactors/operating/ops-experience/pressure-boundary-integrity/overview.html

    Many of the new replacement steam generators are being made with Alloy 690 tubes, so this new alloy is apparently more resistant to stress cracking in either acidic (RCS) or caustic (secondary side of SGs) environments.

    Regarding Midland, I was in college when the plant was in construction. I remembered one of my professors talking about how the plant was going to be integrated into the DOW chemical plant. That fact stuck with me.

    ReplyDelete
  4. @t: Ah, so you had the answer all along! :)

    ReplyDelete
  5. "Ah, so you had the answer all along!"

    You just reminded me that the original Davis-Besse problem was with the nozzle and that it was Alloy 600 material. A quick Google search gave me the NRC link and much more information.

    If a football-sized hole can be made in one of these heads and still not create a Loss of Coolant Accident, it just goes to show how much margin there is in the pressure vessel design. TEPCO has implied that they think some molten fuel actually left the reactors at Fukushima. I guess we'll eventually find out, but I hope it was all contained.

    ReplyDelete