5 ms·
After reading the book "A Piece of the Sun - The Quest for Fusion Energy", I came to understand just how hairy these reactors have grown. A simple idea has nee
by robotrout 8y ago
After reading the book "A Piece of the Sun - The Quest for Fusion Energy", I came to understand just how hairy these reactors have grown. A simple idea has needed to have layer upon layer of complexity added to it, up until the point that I'm not sure it would be viable even if it did work in the lab.
Not only will each commercial reactor, should anybody attempt one, be extremely complex and therefore non-robust, but there's still the issue of neutron contamination, making the whole thing brittle and ready for the nuclear waste disposal team to rip out most of it in a few years time.
- Retric 8y agoA lot of the current complexity relates to the fact these are experiments not production systems. We want to be able to change a great deal of different things with high precision while gathering data, not simply generate power. Production systems would likely have much higher gas contamination for example and operate at a steady state for months. Also, ITER is actually rather small from a power production standpoint, 500MW thermal power ~= 150MW of electricity where we have plenty of fusion reactors several times that size. Luckily things get much easier when you scale fusion power.
- pfdietz 8y agoNot being production systems means entire subsystems are omitted. No tritium breeding blankets, no robotic maintenance systems, not extremely high efficiency tritium purification and recovery systems. Research reactors don't need the materials that could withstand the extreme neutron loading a production reactor would be exposed to. And they don't need to be reliable enough to operate with a high enough capacity factor to pay off the investment, as a production reactor would have to. As you scale up the power of a fusion reactor, the volumetric power density goes DOWN. This is because it becomes limited by the power/area through the first wall. Square/cube law in action.
- Retric 8y ago> no robotic maintenance systems ... neutron load. ITER has full remote handling as it's expected to get extremely radioactive. Scaling reactors up does not really mean increasing the neutron load on the walls as you want to keep that fairly steady per surface area. Tritium production has ~zero impact on operation. The breeding blankets are really simple, you take lithium and enclose it in a metal of some type. Replace after a few months. Also, a ~10x device is unlikely to be 50-50 DT as while hard to operate on pure DD fusion it's not that hard to get when you have some tritium in the mix.
- pfdietz 8y agoRetric: a production reactor would need robots capable of replacing the entire first wall every few years. ITER will not have anything like that.
- Retric 8y agoITER remote maintenance system (IRMS) is specifically designed to do just that: http://users.isr.ist.utl.pt/~mir/pub/SOFT_RHSystemsForITER_IRibeiro-PlenaryTalk.pdf http://users.isr.ist.utl.pt/~mir/pub/SOFT_RHSystemsForITER_I... It's not designed to replace everything, but I doubt commercial designs would go that far either.
- DennisP 8y agoMIT's ARC is designed to make that easy. The reactor opens right up and you just lift out the inner wall and drop in a new one. They've already tested joints in the REBCO tape, and found that they introduce very little electrical resistance.
- pfdietz 8y agoAnd not only extremely complex, but also very large. Fusion reactors have terrible power density compared to fission reactors. Complex + large = very expensive. These failings have been known for 35 years or more, and yet still they're largely ignored. If there is a thing called "pathological technology" (in analogy with pathological science) then fusion must qualify.