In a previous post I wrote entitled The Economics Of A Migrant Nuclear Worker, I mentioned the ice condenser, but did not go into great detail as to what the system was. This is an article that explains to the best of my ability what the ice condenser is, what we do to maintain it and why we do it.
The ice condenser is a safety system designed to rapidly absorb steam and reduce containment pressure following a loss of coolant accident (LOCA) or Main Line Steam Break (MLSB). The ice condenser exists in a semi-circle around the containment of a nuclear reactor and is composed of 1,944 baskets 12.125″ in diameter and 48′ deep. These baskets combined hold roughly 2.5 million pounds of ice. In the United States there are 9 ice condenser units on 5 sites. DC Cook (AEP), McGuire (Duke), Catawba (Duke), Seqouyah (TVA), and Watts Bar (TVA). Each site has 2 units, but Watts Bar unit 2 is not yet operational. There are two other ice plants in the world, the Lovisa plant in Finland and the Ohi plant in Japan. D.C. Cook unit 1 was the first ice condenser plant built and it came online in 1975. Watts Bar unit 2 will be the last, which should come online next year. Watts Bar unit 2’s construction was put on hold in 1988 when it was 80% complete, construction resumed in 2007.
The ice condenser is arranged in a series of 24 bays, each containing 81 ice baskets in a 9X9 grid. since the ice condenser is a semi circle around containment the inside dimension of each bay is smaller than the outside dimension. There are two primary work areas for the ice condenser, upper ice and lower ice. From upper ice we weigh the baskets, empty the baskets, inspect them and refill them. In lower ice we unpin baskets that need to be unpinned (in order to help get a weight on them or to remove them), inspect the bottoms of baskets, inspect the lower inlet doors, and of course remove all the ice that is produced from emptying the baskets and the the ice that falls out of the baskets as we refill them.
Ice condenser containments were designed and sold by Westinghouse in the late 1970s and early 1980s. The advantage of an ice condenser containment is that a much smaller containment building can be built because the ice condenser can reduce the pressure the containment building needs to withstand. The customer saves a lot of money building a smaller containment building, and the community gets a reactor that is much less visible. Take a drive down Red Arrow Highway in Michigan and let me know if you can see the containment building. If they didn’t have a giant sign out front you would have no idea there was a nuclear plant back there.
The ice condenser was designed and sold as “maintenance free”, I’m writing this, so as you know this ended up not being the case. Ice over time sublimates, just like in an old freezer where ice builds up on the walls, the same thing happens in an ice condenser containment. Ice leaves the baskets in the middle and moves to the edges on the condenser. In the event of an accident, if left unchecked, this sublimation could cause the system to fail. This is the primary reason that these systems are not maintenance free. The goal is to have as close to an even amount of ice throughout the condenser. Baskets that are too light are emptied and re-filled, as are baskets that are too heavy.
By emptying the baskets that are too light and re-filling them, we keep the ice bed balanced to provide for an even melting during an accident. At DC Cook, the plant in which I work at, containment cooling projects which keep the inside of the containment building cooler during operation have slowed down the rate of sublimation, which in turn requires less total maintenance. How cold is the ice condenser? Roughly between 8 and 20 degrees Fahrenheit.
What we do:
This is how ice condensers are serviced by my employer, which maintains 3 of the ice plants in the US. While all 3 of these plants use a similar system, the plant I work at has some extra steps and a lot of extra documentation due to a major industry OE (operating experience).
While the unit is online we weigh every basket for the engineers to determine which baskets need to be serviced. They enter the data into a linear equation and essentially baskets that are underweight need to be emptied and refilled, and baskets that are substantially overweight need to be emptied as well, and slightly overweight baskets need to be dipped. Dipping a basket is essentially removing ice from the top of the basket, these dips are typically 6″ to 24″. Some baskets we are not able to get a weight on. All of the baskets that we try to weigh twice and can not achieve a weight on, we unpin once the unit goes offline. All of the baskets are held down in lower ice with a pin that is secured through the structural steel. We remove these pins to give the basket more range of motion and try to weigh them again. If we can not achieve a weight, engineering goes off of projections for the weight based off of that particular baskets previous valid weights and the trend of the baskets around it. The computer program they use stores all of the data we generate and trends the movement of ice over time.
We remove a film barrier (basically a heavy cloth that is bolted between the chillers to the ice bed, which creates a seal), then we unbolt the deck doors , which are bolted horizontally to the framing. Once these tasks are complete we can move these door assemblies (each bay door assembly of 8 doors weights 825lbs). We stack them 2 high, giving us 12 open bays. Now we can begin basket servicing. The baskets engineering selected to be serviced will be emptied, inspected with a camera down the entire basket, every ligament, every screw, every detail of the basket is inspected. We make any necessary repairs to the basket, then fill the basket. As bays are complete we shuffle the deck doors to open up more bays until we have serviced all of the baskets. In an average outage we service around 200 baskets, about 10% of the ice condenser. We then restore the ice condenser to how it was found.
Ice baskets are essentially a lattice mesh of 1″ square holes. with 1/8″ steel lattice. We empty the baskets by using modified air powered concrete vibrators. These vibrators break up the ice and knock it out of the lattice. A basket can take anywhere from 3 hours to 3 shifts to empty. We refill our ice by creating it in our ice machine, which pumps the ice which is a borated snow, into a 3″ line, up 70 feet and into the ice condenser, to a tool called a cyclone, which directs the ice into the basket we need to put the ice in. On a good shift our ice machine can fill 12 baskets.
We also have to chip ice off the walls of the condenser. This is the wall panel project. We empty 9 baskets, completely remove them, then send guys down into the ice bed to chip the ice off the walls, then replace the baskets. We typically do 3 bays an outage.
Who would have thought in a nuclear power plant there would be paperwork? lol. Everything we do has paperwork with it. Each basket we weigh has a minimum of 3 sheets of paper, X 1,944 baskets with some being weighed up to 4 times. It is always amusing to drop off our weighing surveillance paperwork to the work control department. We have to use a cart and several boxes. We have other tasks that individually use more paper, but since we have to weigh every basket, in bulk this is our largest paper consumer. I now spend countless hours reviewing paperwork during our ice condenser project, which typically lasts for 14 days of online work to weigh the baskets and around 25 days of work during the outage. Each piece of paper we generate is reviewed several times, first by the crew who did the work, second by a maintenance supervisor, third by a project lead from the utility company, fourth by engineering and a 5th time by work control.
Foreign Material Exclusion (FME):
The Ice Condenser is what is called a High Risk Foreign Material Zone. What this basically means is that if anything is accidentally dropped into the system, it may be impossible to recover and could be a major problem if the ice condenser had to function. Foreign material in the ice bed would be released when a flash of steam is ran through it. This foreign material could do two things: clog the lower inlet drains that allow water to exit the ice condenser into lower containment, or clog the sumps which are used to distribute the water once it gets to lower containment.
Needless to say keeping foreign material out of the system is highly important. Here’s the challenge: Our project involved over 130 people working around the clock and using LOTS of equipment. Most high risk foreign material projects require only a few tools, our project requires thousands of items. Every square inch of the ice condenser is an area where something can drop into the system. We have a state of the art computer based FME system and designated FME monitors to control what goes in and what goes out. We use lanyards on all of our tools, inspect everything before it comes in, and all personnel are trained in Foreign Material Exclusion. From time to time we do have items that end up falling or that we can’t initially find. We then make a plan for retrieval and find the lost items. The important thing is that our training, our FME system, and our culture ensure that when an item is lost or dropped we know about it and can document it and find it.
D.C. Cook has a major O/E (Operating Experience) due to foreign material in the ice condenser. The standards for maintenance on this system were nothing like they are today back in the 1980s and 1990s. In 1997 DC Cook shut down and stayed shut down for 3 years. One of the major factors to this shutdown was the ice condenser. Between poor documentation and foreign material in the ice bed, the ice condenser was deemed a major safety problem. All of the ice from both units was removed, all of the baskets and deck doors were replaced and the baskets were refilled. During this process over 6 55 gallon drums worth of foreign material were discovered. This was unacceptable. When the plant came back online the standards were greatly improved.
With the exception of project management, the entire staff for this project is contracted (including myself). I started working as a contractor in 2006. Due to the short nature of these jobs it is difficult to make being an ice condenser worker a career, most people either work other nuclear outages or have a normal job that allows them flexibility and/or time off during the outages. I personally work as a steam generator nozzle dam technician when I am not working ice condensers. Every season my company hosts a job fair and hires between 30 and 50 people depending on our returnee rate. All of these people need to be able to pass a drug test, background check, and training at the plant in order to work for us. When there is an upcoming job fair I will post it on the Action Economics Jobs page, so check back often!
We spread the new hires out across several crews to make sure that no individual supervisor ends up with all new people. Very few contracting groups bring in this amount of new to nuclear people at a time. We are constantly in training mode and we have a very strong program. People who start out in the nuclear industry working for the Ice Condenser project go on to be all stars working in a variety of positions across the country. People I have worked with have become project managers, nuclear plant operators, I and C technicians, Radiation Protection workers, Steam Generator project leads and more. For more information on nuclear power jobs check out the book The Essential Guide to Getting a Job in the Nuclear Power Industry: How To Secure Full-Time Employment or Contract Work.
Do you have any questions about the ice condenser? Ask and I will try to answer them to the best of my ability.For some extra reading on Ice Condensers, there is one book published that talks about the subject: Tritium on Ice: The Dangerous New Alliance of Nuclear Weapons and Nuclear Power.
NOTE 6/30/17: We will be having a job fair for ice condenser work at D.C. Cook for this Fall’s outage. The job fair will be held at the Holiday Inn Express at 3019 Lakeshore Dr, St. Joseph, MI on August 1st from 9AM – 5PM. The job should last 4 – 6 weeks.