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

Friday, June 15, 2012

Water Hammer - guest post by A. David Rossin

Background: A number of us have been discussing, for some months now, the chance that some of the water leakage at the Fukushima Daiichi plants which experienced the meltdowns might be completely external to the reactor vessels, and might be due to the deleterious effects of water hammer (sometimes in the old days known as 'steam hammer' under variable conditions.) This happens when a dead end of pipe, or piping with a restriction or perhaps a check valve is suddenly pressurized; water, being for all practical purposes totally incompressible, can cause serious damage when it is propelled suddenly with high velocity to the end of a piping run which effectively is dead headed. It has been postulated that this kind of effect, perhaps in conjunction with thermal shock (the operators did have to resort to the use of any water volume available, including sea water) could have resulted in leakage that detracted from flow that should have been cooling reactor fuel at any or all of the three plants that experienced core melting. While this is merely conjecture, I will add that is is sensible and informed conjecture - but also conjecture that can't be proven until TEPCO can get far deeper into the plants than it is now.

What follows is a very interesting reflection on this phenomenon by A. David Rossin. "Dave" Rossin has become one of my e-mail acquaintances (who I cannot wait to meet in person) through the American Nuclear Society; Dave is in fact a past ANS President. His experience is incredible and quite varied, and a proper bio would be a half hour or hour long TV documentary. I'm proud to present his piece below in entirety and unedited.


Water Hammer – Steam Hammer

A. David Rossin 6-10-12

A couple of years after I started work at Argonne in 1955 I was assigned to the ALPR project. This was the small 3 Mw thermal BWR which was designed to be for a remote Alaskan DEW Line outpost. It was later renamed SL-1. Remember?

I was “Assistant Reactor Physicist.” After we had run it for a couple of months, we scheduled a set of tests to verify its safety and to test the performance of some safety systems. The ALPR had a spray ring inside the vessel above the top of the core. Its purpose was to provide emergency cooling water to cool the core if it was necessary to quench the steam, depressurize the vessel and flood the core. In some ways it was the forerunner of the HPCI, the high-pressure injection system of the first BWR’s.

I was in the control room, which was separated from the reactor building by a gap, some steel and12 feet of rock and sand shielding.

The plan was for the operator was to SCRAM the reactor, slow down and then cut off the recirc pump, let boiling subside, and then activate the emergency spray cooling system.

Several seconds after the pump was turned on the banging started. The noise level was terrible! The control room shook violently, and we figured that the reactor and its structure did too. The emergency pump was cut off ASAP. The heaviest banging soon stopped but a lower level of noises kept up for a while.

I recall that an “older” engineer (he must have been 53 or 54) announced that it must have been “steam hammer.” He told some of his boiler stories, and we all recalled visits to our parents or grandparents’ apartments in the winter when the radiators had gone on and there were couple of minutes of steam hammer before things settled down and they got hot. Heavy cast-iron radiators seemed to sit just below windowsills.

“Steam Hammer”, “Water Hammer” - the two names were both used.

What was obvious was that when water was injected by the emergency pump into the spray ring, the ring pipe was plenty hot, and before high-pressure water flow could be established into the ring and out the many holes, water would flash to steam at many places and the banging was the result. Much later, when the ring itself was examined, damage was found on the supports that held the ring onto the inside wall of the pressure vessel. But as I faintly recall, the ring itself, though kind of discolored, was pretty much as it had been dimensionally. No new cracks or bends.

We decided that the emergency cooling function had to be re-evaluated. Meanwhile the reactor would only be operated at very low power for physics measurements and training purposes.

I went back to Chicago and was assigned to work with Bernie Spinrad as his Reactor Physics Division Assistant Director, Monthly Report Editor and Errand Boy, etc.

It was a couple of years later, serving on the Argonne Reactor Safety Review Committee that the water hammer issue came up again somewhere else. By then, ALPR had been transferred to Combustion Engineering to operate for the Government and named SL-1. To the best of my knowledge, no operator had tried to operate the emergency core cooling system. I believe it had been locked out. It had nothing to do with the later tragedy that killed three men and ended the life of SL-1.

When I was at Commonwealth Edison after 1972 I visited Dresden and Quad Cities. I talked at length with the plant managers. One of them set up a meeting for me with guys from GE. They explained how they had engineered their HPSI and LPSI systems to avoid steam/water hammer, and provided papers on a couple of test programs. Although the issue was looked at again at NSAC while the Peach Bottom PRA was being done, the guys there found it satisfactory and it did not run high in any of their risks analyses.

Every couple of years I get a dream about water hammer. I think about conditions at the time it is needed being outside the envelope of temperature, pressure and flow conditions for which its operation was designed and tested. I just hope I never wake up to find myself reading about it in the paper the next morning.


Thanks to Dave Rossin for this story - which gives our readers here a first person introduction to this phenomenon; we may need this for background as the Fukushima Daiichi investigations continue. Below, from "Boiling Water Reactors" published for the AEC in 1958 (by Andrew W. Kramer), we see the ALPR (later, SL-1) site.

The largest cylindrical structure in the illustration is the reactor building, which housed not only the reactor but the turbine generator as well. The control room was located in the adjacent support building.

6:20 PM Eastern 6/15/2012


  1. This is a great story. For me stories are the best teaching device. Thanks.

  2. This is a really helpful article which should be required reading for everyone who writes specifications.
    The surprising behaviour demonstrated by a small reactor run in off standard conditions should open eyes to the potential for operating excursions. I'd hate to see the results of similar bahaviour in a large power reactor.
    It is too bad that there apparently was not a systematic follow up on this experience. We might have some better understanding of the phenomenon and safer reactors today in that case.