Hacker News Reader

Comment by dragontamer | Hacker News Reader

IIRC, newer nuclear designs use the control rods to control the speed of the reaction, as opposed to purely being a safety measure.

View on HN

Yes, it's used heavily in France (with nuclear >70% of production it's a necessity there)

Interesting. I was under the impression that the control rods were always used to regulate the reaction. How were they used chiefly as a safety measure? Similarly what was used for controlling the rate of reaction when the primary purpose of the rods were purely for safety?

I don't know the details.

What I do know is that the older reactors weren't really the type that could "spin up" or "spin down". Control rods always existed on nuclear power plants, even the oldest designs.

-------

In any case: "Connecting" those rods up to a computer so that the reaction can get hotter or run cooler would be the basis for variable power output.

>would be the basis for variable power output.

This has been done literally forever. Why do you think nuclear plants run at 100% capacity 24/7?

> Why do you think nuclear plants run at 100% capacity 24/7?

The primary reason would be economic. Fuel costs IIRC are really low but capital-expenses (especially safety engineering / politics / convincing the people nearby that the plant is safe) is so costly.

When you have cheap-fuel but expensive machines, it makes sense to run closer to 100%. If you have expensive fuel but cheap machines (ex: Gas turbines), it makes sense to run in a "peaker" fashion.

-----

That being said: I've heard that older nuclear plants are "baseload" designs, in that they _CANNOT_ turn off or feasibly regulate their power output.

I mean, they obviously "can" turn off, they all have control rods for safety. But they're clearly not using them to regulate the power output.

>But they're clearly not using them to regulate the power output.

No, you heard wrong. They're often the last or second to last type of plant that's called upon to alter supply to match demand, but they don't run 100% 24/7.

They want to run them nearly 100% 24/7 for economic reasons though. Also what is the effect of regulating with control rods on the fuel, does it have an impact on how often the fuel needs to be changed?

>does it have an impact on how often the fuel needs to be changed

Yes, fissonable elements in the fuel are consumed in a manner that's proportional to output. So if you run at 25% output your fuel needs to be changed (example) every 4 years, if you run at 100% output your fuel needs to be changed every year.

But it is not 100% proportional, for example, 15% of the plant's output is required simply to run the plant's accessories. So it is advantageous to run at 100% output if you can.

Another comment in this thread, I can't find it now, said that lower output past a certain level did lead to having to change the fuel faster. I don't know who's correct but it feels strange that this would be a linear relationship, these kinds of things don't tend to work like that.

I recall reading that the fuel is 1/3 of the operating expenses, but I could be misremembering.

About 28% as of 2018[1], but there have been some advancements since then. Half of that cost is enrichment and fabrication, and disposal is a factor too. It costs a lot less to fuel nuclear plants than coal plants, but maintenance is more costly for nuclear. Pricing in externalities or requiring scrubbers makes coal FAR more expensive. Natural gas is really cheap right now so that is largely what is undercutting Nuclear for base load power. It's worth noting that historically natural gas production/use has been far more dangerous than nuclear and far less scrutinized by regulators.

[1]http://large.stanford.edu/courses/2018/ph241/wang-k2/

Most nuclear power plants in the US fleet are not variable-output for design reasons. Mostly, both the reactor (see my other comment about neutron poisoning) and turbines are designed for safe operation at a fixed output and cannot operate at "partial" power for extended periods (in the turbines I think the problem is vibration and how it interacts with the bearing designs). A lot of this basically comes from these reactors having been designed without the benefit of computer modeling (everything had to be hand-calculated around a specific target output), newer designs based on computational modeling and optimization techniques achieve much larger safe/sustainable power ranges. Unfortunately we have very few reactors of modern design in the US.

I'm not an expert but my general understanding is that neutron poisoning is the big issue. Most reactors are designed around a careful balance of operating temperature and radiation field to mitigate neutron poisoning by the fuel, and operating outside of that balance will cause the problem to become progressively worse until the reaction is no longer self sustaining. So reactors have a "design power" and they can operate outside of that design power only for limited periods of time, while transitioning in and out of operation. Operating outside of the design power for too long causes thermal output to get lower and lower until the reactor just needs to be shut down while the operators wait out the half-life of the short-lived fuel products that cause poisoning, and then they can start it up again. I think this can be weeks so it's carefully avoided.

You are right about the (mostly Xenon-induced) neutron poisoning being a important design factor, but there's still quite a lot of margin you can use to modulate power (reducing boric acid concentration in water is the key factor here, so it will works better when the fuel has been renewed recently than just before you need to refuel).

I can hear the confusion in your writing and am thinking the same thing as you. The comment you're replying to seems almost artificial. The words refer to nuclear power but there's no sense of any understanding behind them.

Well, I'm a programmer and will openly admit it! I have zero training in nuclear power!

Disclaimer - I only took one reactor course 17 years ago.

The rods soak up neutrons. the reactor is designed so that its power would increase if the rods weren't there. The rods are what keeps the k == 1 (basically how the neutrons change in time. K == 1 is no change).

So don't think of the rods as a throttle valve in your car, i.e. this throttle angle corresponds to this much torque and therefore power. The rods are best thought of as controlling the rate of change of reactor power.

This is why the reactor is built so that the neutrons increase when fully withdrawn - because you have to start it! If the reactor is being started, the rods are withdrawn until the power wanted is achieved. Then the rods are lowered.

Note this is a far more complicated problem; this is a very serious operation because at the very start, except for your neutron starter, you only have "prompt" neutrons that can very quickly cause an increase of power. Once the reactor has stabilized the "nuclear waste" inside it releases non-prompt neutrons that are part of the neutron budget and are easier to control.

So, while waste in the reactor is being accumulated and while the fuel burns up, the position of the rods to achieve stable power changes! Furthermore these changes are different in different parts of the reactor so each rod is individually changed (actually I think there are two sets of rods, not all of them are individually actuated)

I have a degree in Nuclear engineering, although I haven’t been practicing in a while. Your explanation is accurate and well described.

> actually I think there are two sets of rods, not all of them are individually actuated

commercial reactors have 5-10 “banks” arranged symmetrically, although only a subset of those are typically used during power operations [1], so you weren’t far off. Rods are rarely controlled individually, the only case that comes to mind is for identifying and then limiting the reactivity in a “leaker” fuel bundle (one where the cladding has been breached by a foreign object, so it is leaching contamination into the coolant - this is undesirable but expected).

[1]: edit for more information: PWRs and BWRs differ substantially here:

PWRs primarily use boric acid in the coolant to limit reactivity at the beginning of a fuel cycle, which is diluted out over the course of the cycle. One bank of rods is used for fine reactivity control. This gives them a generally even power distribution across the reactor.

BWRs primarily use their control rods, so start the fuel cycle with substantially more inserted. A significant part of designing the fuel layout for a cycle is also designing the control rod patterns to be used over the cycle, to control excess reactivity while maximizing fuel use. Power distribution across the reactor can vary wildly, so it’s a much more challenging engineering problem.

I just remembered I actually have a detailed visualization illustrating (modeled) power distribution in a BWR. You can easily see where control rods are inserted and how that has shifted the power around. https://www.kmr.me/posts/coremap/#content

Those are called "load-follow" designs. Not "newer designs". They are almost as old as the discovery of fission itself.

American nuclear submarines use them, as does France's power generating reactors.

I'm not a nuclear reactor engineer etc, though it was my understanding that control rods in nuclear reactor have always been used to moderate the reaction to control power output, as well as being a critical safety component.

See https://en.m.wikipedia.org/wiki/Control_rod

Reply on news.ycombinator.com