4 ms·

You can regulate a nuclear plant between ~50% and 100% output in a matter of minutes by adjusting the control rods. That's very limited compared to most other

You can regulate a nuclear plant between ~50% and 100% output in a matter of minutes by adjusting the control rods. That's very limited compared to most other options, but obviously much better than solar or wind.

isostatic 9y ago
And far worse than batteries
- seanmcdirmid 9y ago
  Batteries, like hydro reservoirs, can be charged from inflexible energy sources when they have surplus output. It really isn’t fair to compare them to energy sources as they can smooth over the inflexible ones.
  - nl 9y ago
    Since they are direct substitutes it seems reasonable to compare them to me.
    - seanmcdirmid 9y ago
      Again, batteries make sense with inflexible energy production methods. So if you want to use that coal more effectively, get a bunch of batteries or a lake up a hill to pump into.
- lmm 9y ago
  Not at practical capacities. We're talking about a pair of 1.6GW reactors, so the control rods can adjust total generation by 1.6GW in a matter of minutes. How many batteries would it take to supply 1.6GW for a reasonable amount of time (let's say 24 hours)? How much would that cost?
  - isostatic 9y ago
    100mw output of batteries costs $38m, so 16 of those costs $0.6b, 3% of the cost of Hinckley C, even assuming no more overruns. Ok that only gives you 1 hour of charge, but your requirement was changing in minutes. These batteries respond to load change in milliseconds.
    - lmm 9y ago
      > 100mw output of batteries costs $38m, so 16 of those costs $0.6b, 3% of the cost of Hinckley C, even assuming no more overruns. So 24 hours' worth would cost 72% of the cost of Hinkley C, and that's not including any amount of power generation. Looks like wind/solar + enough batteries to timeshift their output to match consumption would end up costing more. > Ok that only gives you 1 hour of charge, but your requirement was changing in minutes. These batteries respond to load change in milliseconds. The ability to change quickly is only part of the problem; if you want to run the country on wind or solar then you need to be able to handle several overcast or calm days in a row.
      - ben_w 9y ago
        > if you want to run the country on wind or solar then you need to be able to handle several overcast or calm days in a row. On a country scale yes, but not on a continental scale. Although in the case of solar, you probably just want to have peak output twice the size you’d expect for naive insolation, and it looks like it’ll be cheap enough soon enough for that to be economical.
hansthehorse 9y ago
No, you can't. I spent 22 years as a operations nuclear plant supervisor and I can tell you this can't be done. Two to four mw/min power increase, depending on several factors, is the norm. Power follows steam demand is an old saw in the industry. You slowly open the turbine valves and dilute the coolant in the reactor to keep things in balance. Control rods are kept fully withdrawn at all times at power.
- tormeh 9y ago
  So the nuclear reaction has constant power output, you just throw the energy you don't want away by using coolant instead of turning the turbines? That sounds.. inefficient.

isostatic 9y ago

And far worse than batteries

seanmcdirmid 9y ago
Batteries, like hydro reservoirs, can be charged from inflexible energy sources when they have surplus output. It really isn’t fair to compare them to energy sources as they can smooth over the inflexible ones.
- nl 9y ago
  Since they are direct substitutes it seems reasonable to compare them to me.
  - seanmcdirmid 9y ago
    Again, batteries make sense with inflexible energy production methods. So if you want to use that coal more effectively, get a bunch of batteries or a lake up a hill to pump into.
lmm 9y ago
Not at practical capacities. We're talking about a pair of 1.6GW reactors, so the control rods can adjust total generation by 1.6GW in a matter of minutes. How many batteries would it take to supply 1.6GW for a reasonable amount of time (let's say 24 hours)? How much would that cost?
- isostatic 9y ago
  100mw output of batteries costs $38m, so 16 of those costs $0.6b, 3% of the cost of Hinckley C, even assuming no more overruns. Ok that only gives you 1 hour of charge, but your requirement was changing in minutes. These batteries respond to load change in milliseconds.
  - lmm 9y ago
    > 100mw output of batteries costs $38m, so 16 of those costs $0.6b, 3% of the cost of Hinckley C, even assuming no more overruns. So 24 hours' worth would cost 72% of the cost of Hinkley C, and that's not including any amount of power generation. Looks like wind/solar + enough batteries to timeshift their output to match consumption would end up costing more. > Ok that only gives you 1 hour of charge, but your requirement was changing in minutes. These batteries respond to load change in milliseconds. The ability to change quickly is only part of the problem; if you want to run the country on wind or solar then you need to be able to handle several overcast or calm days in a row.
    - ben_w 9y ago
      > if you want to run the country on wind or solar then you need to be able to handle several overcast or calm days in a row. On a country scale yes, but not on a continental scale. Although in the case of solar, you probably just want to have peak output twice the size you’d expect for naive insolation, and it looks like it’ll be cheap enough soon enough for that to be economical.

hansthehorse 9y ago

No, you can't. I spent 22 years as a operations nuclear plant supervisor and I can tell you this can't be done. Two to four mw/min power increase, depending on several factors, is the norm. Power follows steam demand is an old saw in the industry. You slowly open the turbine valves and dilute the coolant in the reactor to keep things in balance. Control rods are kept fully withdrawn at all times at power.

tormeh 9y ago
So the nuclear reaction has constant power output, you just throw the energy you don't want away by using coolant instead of turning the turbines? That sounds.. inefficient.