4 ms·

Very likely, each fusion plant (if they ever exist) would require, ironically, a fission plant next door to provide the power required for the magnets. The mag

by hpcjoe 5y ago

Very likely, each fusion plant (if they ever exist) would require, ironically, a fission plant next door to provide the power required for the magnets. The magnets, and wiring won't be cheap. The power delivery and control won't be cheap. I'm not sure how this would be amortized into the cost of the power, without making it 2-3x (or more) more expensive than alternatives. We can build, and we need, nuclear plants now, to be able to generate cheap/plentiful electric power. And if we don't we're basically going to have to push the brakes on EV deployment. Or light up more NG/Oil plants to provide the power for those.

thehappypm 5y ago
This approach doesn't use magnets.
epistasis 5y ago
> We can build, and we need, nuclear plants now, to be able to generate cheap/plentiful electric power. And if we don't we're basically going to have to push the brakes on EV deployment. Or light up more NG/Oil plants to provide the power for those. The real power generation tech here will be wind and solar, not nuclear. We can build lots and lots and lots of it, easily and cheaply. And using it to charge vehicle batteries doesn't even require intermediate storage. Even construction powerhouses like China are deploying more than an order of magnitude more renewables than nuclear, because renewables will be the backbone of any future grid. Nuclear isn't getting any cheaper, yet wind solar and batteries are on exponentially decreasing cost curves. People on HN are far better able to understand exponential technology advancement than people in the energy industries. And hopefully we can all see how renewables plus batteries is going to make nuclear obsolete.
- elihu 5y ago
  I wish there was more interest in electrifying our major roads and freeways (i.e. allowing cars to charge while they're moving) in part to reduce the necessity of hauling giant batteries around everywhere, but also to shift EV power use from mostly-nighttime to mostly-daytime when more solar power is available. Another option is to get some high-capacity high-voltage DC lines connecting continents so that countries around the world can sell power during the day and buy it at night. We can keep the fossil fuel plants around as emergency backup. At this point, though, we're a long way even from being able to shut off the fossil fuel plants during the day while the sun is shining.
  - epistasis 5y ago
    Eh, storage is a solved problem for passenger vehicles. It's really only long haul that's a problem. Electrifying roads is too expensive, for too little benefit. Electrifying rail, however, would be a fine idea. Nearly all vehicles are parked for most of the day. We should incentivize workplace charging, to help with that. But really, we must drastically reduce our need for driving. We simply can't scale EV production quickly enough to make the impacts we need to. If we want to reduce car emissions by half by 2030, we need to be at 100% EV sales very very soon, and we are no where close to it.
    - elihu 5y ago
      I'm not sure that electrified roads would really be all that expensive relative to the benefits. I mean, if you only add electrification to a two mile segment every twenty miles on a few major interstate highways, it means that long haul trucking that uses those highways can be fully electric. That seems like a pretty big deal. If all the major roads and highways are electrified for 100% of their distance (like we would have had to have done in order to switch to electric, say, back in the 70's during the oil crisis when lead acid was the best battery tech available), that would be far more expensive and probably not worth it. Driving less, more workplace charging, more mass-transit, and electric rail are all good. It'd be great to see electric shipping too, but I'm not sure how to get there; you'd need multiple charging stations in the middle of major oceans or something.
Robotbeat 5y ago
NIF doesn’t use magnets. Pure inertial confinement. You’re thinking of Tokamaks. Those use superconducting magnets. Superconducting magnets like ITER uses don’t require energy to continue running as they, of course, are superconducting. They “only” have to dump heat put into them by the reaction radiation. The SPARC reactor by MIT is similar but much smaller by using Cuprate superconductors (high temperature superconductors, but here operated at much lower temperatures to increase the critical field) that allow much higher field strengths. They often have non-superconducting joints which cause a (sort of) small amount of additional heat that needs to be dumped, but it is possible to make such joints superconducting as well. Anyway, all such designs for commercial scale power from Tokamaks use fusion generated electricity to power the magnet cooling, and those electricity requirements are less than the electricity produced.

by hpcjoe 5y ago

thehappypm 5y ago

This approach doesn't use magnets.

epistasis 5y ago

> We can build, and we need, nuclear plants now, to be able to generate cheap/plentiful electric power. And if we don't we're basically going to have to push the brakes on EV deployment. Or light up more NG/Oil plants to provide the power for those. The real power generation tech here will be wind and solar, not nuclear. We can build lots and lots and lots of it, easily and cheaply. And using it to charge vehicle batteries doesn't even require intermediate storage. Even construction powerhouses like China are deploying more than an order of magnitude more renewables than nuclear, because renewables will be the backbone of any future grid. Nuclear isn't getting any cheaper, yet wind solar and batteries are on exponentially decreasing cost curves. People on HN are far better able to understand exponential technology advancement than people in the energy industries. And hopefully we can all see how renewables plus batteries is going to make nuclear obsolete.

elihu 5y ago
I wish there was more interest in electrifying our major roads and freeways (i.e. allowing cars to charge while they're moving) in part to reduce the necessity of hauling giant batteries around everywhere, but also to shift EV power use from mostly-nighttime to mostly-daytime when more solar power is available. Another option is to get some high-capacity high-voltage DC lines connecting continents so that countries around the world can sell power during the day and buy it at night. We can keep the fossil fuel plants around as emergency backup. At this point, though, we're a long way even from being able to shut off the fossil fuel plants during the day while the sun is shining.
- epistasis 5y ago
  Eh, storage is a solved problem for passenger vehicles. It's really only long haul that's a problem. Electrifying roads is too expensive, for too little benefit. Electrifying rail, however, would be a fine idea. Nearly all vehicles are parked for most of the day. We should incentivize workplace charging, to help with that. But really, we must drastically reduce our need for driving. We simply can't scale EV production quickly enough to make the impacts we need to. If we want to reduce car emissions by half by 2030, we need to be at 100% EV sales very very soon, and we are no where close to it.
  - elihu 5y ago
    I'm not sure that electrified roads would really be all that expensive relative to the benefits. I mean, if you only add electrification to a two mile segment every twenty miles on a few major interstate highways, it means that long haul trucking that uses those highways can be fully electric. That seems like a pretty big deal. If all the major roads and highways are electrified for 100% of their distance (like we would have had to have done in order to switch to electric, say, back in the 70's during the oil crisis when lead acid was the best battery tech available), that would be far more expensive and probably not worth it. Driving less, more workplace charging, more mass-transit, and electric rail are all good. It'd be great to see electric shipping too, but I'm not sure how to get there; you'd need multiple charging stations in the middle of major oceans or something.

Robotbeat 5y ago

NIF doesn’t use magnets. Pure inertial confinement. You’re thinking of Tokamaks. Those use superconducting magnets. Superconducting magnets like ITER uses don’t require energy to continue running as they, of course, are superconducting. They “only” have to dump heat put into them by the reaction radiation. The SPARC reactor by MIT is similar but much smaller by using Cuprate superconductors (high temperature superconductors, but here operated at much lower temperatures to increase the critical field) that allow much higher field strengths. They often have non-superconducting joints which cause a (sort of) small amount of additional heat that needs to be dumped, but it is possible to make such joints superconducting as well. Anyway, all such designs for commercial scale power from Tokamaks use fusion generated electricity to power the magnet cooling, and those electricity requirements are less than the electricity produced.