One of Plant Vogtle’s new $30 billion nuclear reactors is splitting atoms. What’s next?
Georgia Power announced Monday that the first of two nuclear reactors under construction at Plant Vogtle, in Burke County, has achieved “initial criticality,” meaning engineers have begun splitting atoms.
The construction of Plant Vogtle’s Units 3 and 4 has been mired in budget overruns and delayed for years. When it comes online, Plant Vogtle’s Unit 3 will be the first new operating nuclear reactor built in the U.S. since 1996.
To find out what Monday’s announcement means for the project, the Telegraph called Mark Nelson, founder of the energy consultancy Radiant Energy Group and a nuclear engineer by training. This interview has been edited for clarity and length.
Q: Describe what just happened at Unit 3.
A: Let’s talk about the process of starting a reactor. All the painful stuff is done, mostly: the years, the financing, the delays, the problems. So a big moment is when the reactor receives permission to load in fuel. That was done last year. The fuel is now in the reactor. That’s not the same as starting up. Now, starting up has several stages. For me, the most important stage is what just happened, which is called going critical, which means neutrons are multiplying. Chain reactions that can self-sustain are now occurring. There is not enough power to do anything with just the neutrons coming from splitting. Atoms have been detected in amounts that show that the chain reaction is ongoing.
Q: What’s the process that goes into this moment happening?
A: So the reactor is fueled, the reactor is closed, bolted shut. Control rods are slowly being pulled out. The control rods absorb neutrons without undergoing any nuclear reactions. They’re like sponges absorbing the neutrons that would be bouncing around, making more chain reactions. The control rod has been pulled out, thereby allowing some neutrons not to get uselessly absorbed. Instead, the neutrons are getting more uranium fuel and breaking it apart and making more neutrons. So that’s the criticality part.
After this point, there’s probably six months of checks — safety checks, performance checks, equipment checks — before the power is brought up to commercial operation. So the reactor is still at a minuscule fraction of its eventual power, almost no power at all. What they need to do is increase the reactions by many orders of magnitude, to increase the power by millions of times in order to start making heat that can turn into steam, which can turn the turbine.
Q: To use an inelegant metaphor, is it a little bit like a pilot light in a gas heater?
A: You know what? I think I like that. That’s not terrible. Or, this is maybe a little bit weird, but it’s like detecting a heartbeat for the first time.
Here’s another way I would want to say it. This is the moment when it goes from a cold, dead machine to something getting energy from the fundamental physical forces of the universe.
Q: Is it typically a dramatic moment inside the facility? Do you imagine there were cheers when this thing happened?
A: Absolutely there would have been cheers. It’s like a successful launch on a spaceflight, not yet arriving at the moon, if that makes sense. It means the reactor is now alive. It means it is now permanently alive.
Q: Why did this happen now rather than when it was supposed to have happened?
A: The longest delays are because we didn’t know how to build reactors in America. Then there were delays because some of the work that was done was done incorrectly and had to be fixed.
There was a breakdown between that company that supplies the nuclear reactor and the engineers and construction firms. That’s Westinghouse and the firms that were supposed to turn Westinghouse’s design into a power plant, build the power plant, and have it ready to install the equipment. That completely broke down. There was not a fully construction-ready design before construction started. The supply chain to deliver the critical parts, the modules of the reactor, was not in place. And the new construction techniques that were supposed to save time were too new for the construction firms. They didn’t know how to do these new things, and so it didn’t work well.
Then, very recently, the delays were for other reasons. There was vibration in one of the safety backup pipes. Sometimes when water runs through — I don’t know if you ever heard of water hammer, or if you’ve ever heard pipes in any old building go clacketa-clacketa-clack when starting to turn on water. They had a problem that was kind of like that — vibrations caused by operation in one of the tests, and they had to quickly make a fix to that part. So that delayed it for a few weeks.
Q: Is it smooth sailing from here on for Unit 3, or are there issues that could occur between now and when it comes fully online?
A: There can be issues. And historically, reactors have occasionally taken more than a year to enter commercial operation. In this case, there’s been so much time to do debugging, I mean, that’s part of this slowing, that we do expect it to be smooth. However, I expect the fourth one to be even smoother. This is not the first [Westinghouse reactor model] AP1000 to come on. Instead, it’s the fifth AP1000 to turn on in the world. So they do know a lot of what’s possible to go wrong as they’re getting started.
There will be an internal schedule that they hope to hit. That will be their ability to raise their reactor to, say, 1% power, then 10% power, then 100% power. So they need to hit the milestones of raising the reactor to power. Once you raise it to higher levels of power, you have to have the turbines going, too, because you have to dissipate all that heat. So they’re going to be working on raising the reactor power slowly on up to the different levels.
There is going to be a series of milestones to hit after this. Full power is one of the big ones. Synchronization with the grid is another one. And then the start of commercial operations, meaning selling electricity, is the final one.
Q: What is that process of raising the power levels like?
A: When I was in grad school for nuclear engineering at Cambridge, we had this activity in which we had this reactor simulator built into Microsoft Excel, and our homework assignment was to successfully start up this reactor — that’s this little graphic with little flowing water signs and stuff on a spreadsheet. And we had little controls that we can manipulate to start the pumps that move the water, to start the control rods, to start the sensors, to start all these different things. No matter what we tried, we couldn’t get the reactor to successfully start until we followed the instructions precisely. It took about 45 minutes to an hour to start up the reactor. And if we clicked one wrong button, it automatically shut down and we had to start over.
So there’s a bewildering number of steps. They have to do with the temperature of the water. They have to do with the substances put in the water to control reactivity in the reactor. They have to do with the pumps of many different systems working. It has to do with the neutron count, the detectors showing the rate of multiplication of the neutrons in the reactor. It has to do with the positions of the control rods as they’re slowly withdrawn from different parts of the core. It has to do with the steam generators being ready, heated up and ready to make steam. It has to do with a device called the pressurizer, which is a device that just keeps the pressure exactly as designed in the system, not too high, not too low. It has to do with the turbine, a big spinning machine. The lubrication system has to be pumped up and ready to go. The turbine has to be spinning at the right speed to connect with the grid — the turbine physically is linked to the electricity grid; if the grid feels the turbine coming on, the turbine feels the pull of the electricity grid.
Q: Is Unit 4 basically a doppelganger of Unit 3?
A: Exactly. It is identical.
However, that does not guarantee that if one has problems the other will, or if the other doesn’t have problems one will. For example, when two of these reactors started up at a site in Sanmen, China, one of them had troubles with one of their pumps that moves water through the core, carrying out the heat from their reaction to turn into steam. And one of the reactors was down for quite a while as they fixed the pump.
In general, it is much faster and more straightforward to do every procedure on the second of a twin reactor. It’s faster to build, faster to turn on, cheaper to do all those things. We should expect Unit 4 to be as fast or faster than Unit 3.