5 ms·

ITER has been working on this problem for almost a decade now, and with more money available to them. What are the odds some startup can beat them to the punch?

brobdingnagians 7y ago
Happens all the time with other kinds of startups. The main thing with this is having a good skillset and a good idea.
- ajross 7y ago
  A good skillset and a good idea is the "main thing" if you want to make a software product. What you need with fusion is fundamentally new science, not just an idea to pound into shape. That's rather a taller order than a disruptive crowd-scale gig play in the power industry.
pinewurst 7y ago
How can one presume that the multi-national, highly political, very long term ITER program is the definitive source of truth?
starpilot 7y ago
Well, GF is using a liquid metal liner to compress their plasma: https://www.youtube.com/watch?time_continue=77&v=k3zcmPmW6dE&feature=emb_title https://www.youtube.com/watch?time_continue=77&v=k3zcmPmW6dE... Oddly, I remember a UFO TV show describing the alien powerplant as relying on liquid metal in some way. No idea if that is meaningful.
JumpCrisscross 7y ago
> ITER has been working on this problem for almost a decade now ITER has been working on one implementation of magnetic confinement. It's an international effort, bringing to the battle the advantages of scale and disadvantages of multilateral bureaucracy. Since ITER was formed in 2007 [1], the bleeding edge has advanced beyond Tokomaks. Advanced computational methods birthed the Wendelstein 7-X [2]. Improved superconducting magnets are compacting designs, reducing costs--and thus accessibility to start-ups--through mass reduction. It's almost likely a start-up beats ITER. And that's fine! I don't know if any of these designs would be where they are if ITER hadn't been funded. [1] https://en.wikipedia.org/wiki/ITER#Organization_history https://en.wikipedia.org/wiki/ITER#Organization_history [2] https://en.wikipedia.org/wiki/Wendelstein_7-X https://en.wikipedia.org/wiki/Wendelstein_7-X
- kirrent 7y ago
  You're right about better superconductors leading to more compact and cheaper reactor designs, but many, if not most, of those designs are still tokamaks. To say that the bleeding edge has moved beyond tokamaks is just wrong. To take your example of Wendelstein 7-X, the only sources I can find give an expected Q factor of 0.1. JET was getting 0.75 ages ago. ITER will hopefully hit 10. While I'm excited about stellarators because they're very interesting to research, they probably won't be catching up to tokamaks any time soon.
  - snakeboy 7y ago
    >ITER will hopefully hit 10. As a layman, is it right to think that this would mean viable fusion has arrived at that point? I mean, 10x energy out/in is a victory, no? Is there another factor that stops ITER from being immediately used and replicated for energy generation?
    - kirrent 7y ago
      Well, DEMO, the demonstration power station to follow ITER is estimated to have a Q of 25 and output of 2GW. Whether that's viable will then be less about the physics and more about the cost that they can get subsequent designs down to when they have to compete with high capacity factor offshore wind, battery firmed utility solar, pumped hydro, etc. in 2040. It'll be a fascinating engineering challenge. I hope they can make it happen, but I doubt it.
    - larkost 7y ago
      The next big problem after hitting positive energy production will be sustainability. My understanding is that the walls of the reactor get hit with enough stray neutrons that they degrade (often in radioactive ways). Unlike in fission reactors where you can put in a lot of absorptive shielding (like lead), in a fusion reactor you have to keep the magnets close to what they are working on. Note, I am not talking about really long term radioactive waste here, more like things that are dangerous for less than 100 years. Still problematic, but not for millions of years like fission.
    - 7y ago
audunw 7y ago
For General Fusion they have a completely different approach. It's not a Tokamak like ITER. It remains to be seen if it works at scale though. Commonwealth Fusion also thinks they can move faster than ITER with a Tokamak reactor, by utilising new and better superconductors. I suppose the design of ITER was locked down before those became available in large quantities.
wongarsu 7y ago
ITER is taking the safest approach, and being a giant international cooperation makes them probably as flexible and nimble as a boulder. A startup betting on a novel, cheaper approach could beat them as long as they accept a much higher risk of failure. And while a government wouldn't invest $100M in something with say 10% chance of success, compared to $6B for ITER that's a worthwile bet for everyone else
- 2muchcoffeeman 7y ago
  > And while a government wouldn't invest $100M in something with say 10% chance of success, compared to $6B for ITER that's a worthwile bet for everyone else Maybe governments should. That’s doesn’t seem like a large amount of money in the scheme of things and I’d rather the tech not belong to a startup.
  - mlinsey 7y ago
    In an ideal world, government would invest in many more low-probability/high-impact engineering projects. But this doesn't align well with political incentives - 100M on a bet that will return 50X 10% of the time is great for a private investor, but a politician won't get 50X the votes from a successful project as an unsuccessful one, instead there's a 90% chance the project will be portrayed as a boondoggle.
    - EpicEng 7y ago
      Also, those relatively small investments of $100M will add up to a lot over time. At some point people are going to get a bit sour on the idea of acting like an angle investor with tax dollars.
    - perl4ever 7y ago
      The problem with investing in a lot of low probability things is that when, say, 100 of them all fail, how do you know whether that's because they were low probability, or zero probability? Unfortunately, there's no limit to the number of very low probability things you can invest in, so nobody has the budget for them all.
ArtWomb 7y ago
I'll link to the NAS report on fusion findings. US benefits from ITER partnerships work both ways. But also recommends investments in smaller, less expensive, more innovative (and therefore competitive) designs. http://www8.nationalacademies.org/onpinews/newsitem.aspx?RecordID=25331 http://www8.nationalacademies.org/onpinews/newsitem.aspx?Rec... Just to put things in perspective. ITER budget is estimated ~$25B. Seeking to produce >500MW with a 12T magnetic field. Massive radius. Twice any previous tokamak size. Just look at the pictures ;) JET budget was ~$5B total. And achieved 25MW with a 5T coil. But without actually "burning plasma" (Max Q < 1.0) Princeton Plasma Physics Lab run by DoE budget is >$100M per annum. So this represents a significant challenge, and not just in fund raising. The most promising space might be in the design of "mini-magnets" https://nationalmaglab.org/news-events/news/lbc-project-world-record-magnetic-field https://nationalmaglab.org/news-events/news/lbc-project-worl...
dnautics 7y ago
technically, they are persuing a completely different mechanism. In my opinion it's totally bonkers, but in principle there is no reason why the spend/return should be correlated.

brobdingnagians 7y ago

Happens all the time with other kinds of startups. The main thing with this is having a good skillset and a good idea.

ajross 7y ago
A good skillset and a good idea is the "main thing" if you want to make a software product. What you need with fusion is fundamentally new science, not just an idea to pound into shape. That's rather a taller order than a disruptive crowd-scale gig play in the power industry.

pinewurst 7y ago

How can one presume that the multi-national, highly political, very long term ITER program is the definitive source of truth?

starpilot 7y ago

Well, GF is using a liquid metal liner to compress their plasma: https://www.youtube.com/watch?time_continue=77&v=k3zcmPmW6dE&feature=emb_title https://www.youtube.com/watch?time_continue=77&v=k3zcmPmW6dE... Oddly, I remember a UFO TV show describing the alien powerplant as relying on liquid metal in some way. No idea if that is meaningful.

JumpCrisscross 7y ago

> ITER has been working on this problem for almost a decade now ITER has been working on one implementation of magnetic confinement. It's an international effort, bringing to the battle the advantages of scale and disadvantages of multilateral bureaucracy. Since ITER was formed in 2007 [1], the bleeding edge has advanced beyond Tokomaks. Advanced computational methods birthed the Wendelstein 7-X [2]. Improved superconducting magnets are compacting designs, reducing costs--and thus accessibility to start-ups--through mass reduction. It's almost likely a start-up beats ITER. And that's fine! I don't know if any of these designs would be where they are if ITER hadn't been funded. [1] https://en.wikipedia.org/wiki/ITER#Organization_history https://en.wikipedia.org/wiki/ITER#Organization_history [2] https://en.wikipedia.org/wiki/Wendelstein_7-X https://en.wikipedia.org/wiki/Wendelstein_7-X

kirrent 7y ago
You're right about better superconductors leading to more compact and cheaper reactor designs, but many, if not most, of those designs are still tokamaks. To say that the bleeding edge has moved beyond tokamaks is just wrong. To take your example of Wendelstein 7-X, the only sources I can find give an expected Q factor of 0.1. JET was getting 0.75 ages ago. ITER will hopefully hit 10. While I'm excited about stellarators because they're very interesting to research, they probably won't be catching up to tokamaks any time soon.
- snakeboy 7y ago
  >ITER will hopefully hit 10. As a layman, is it right to think that this would mean viable fusion has arrived at that point? I mean, 10x energy out/in is a victory, no? Is there another factor that stops ITER from being immediately used and replicated for energy generation?
  - kirrent 7y ago
    Well, DEMO, the demonstration power station to follow ITER is estimated to have a Q of 25 and output of 2GW. Whether that's viable will then be less about the physics and more about the cost that they can get subsequent designs down to when they have to compete with high capacity factor offshore wind, battery firmed utility solar, pumped hydro, etc. in 2040. It'll be a fascinating engineering challenge. I hope they can make it happen, but I doubt it.
  - larkost 7y ago
    The next big problem after hitting positive energy production will be sustainability. My understanding is that the walls of the reactor get hit with enough stray neutrons that they degrade (often in radioactive ways). Unlike in fission reactors where you can put in a lot of absorptive shielding (like lead), in a fusion reactor you have to keep the magnets close to what they are working on. Note, I am not talking about really long term radioactive waste here, more like things that are dangerous for less than 100 years. Still problematic, but not for millions of years like fission.
  - 7y ago

audunw 7y ago

For General Fusion they have a completely different approach. It's not a Tokamak like ITER. It remains to be seen if it works at scale though. Commonwealth Fusion also thinks they can move faster than ITER with a Tokamak reactor, by utilising new and better superconductors. I suppose the design of ITER was locked down before those became available in large quantities.

wongarsu 7y ago

ITER is taking the safest approach, and being a giant international cooperation makes them probably as flexible and nimble as a boulder. A startup betting on a novel, cheaper approach could beat them as long as they accept a much higher risk of failure. And while a government wouldn't invest $100M in something with say 10% chance of success, compared to $6B for ITER that's a worthwile bet for everyone else

2muchcoffeeman 7y ago
> And while a government wouldn't invest $100M in something with say 10% chance of success, compared to $6B for ITER that's a worthwile bet for everyone else Maybe governments should. That’s doesn’t seem like a large amount of money in the scheme of things and I’d rather the tech not belong to a startup.
- mlinsey 7y ago
  In an ideal world, government would invest in many more low-probability/high-impact engineering projects. But this doesn't align well with political incentives - 100M on a bet that will return 50X 10% of the time is great for a private investor, but a politician won't get 50X the votes from a successful project as an unsuccessful one, instead there's a 90% chance the project will be portrayed as a boondoggle.
  - EpicEng 7y ago
    Also, those relatively small investments of $100M will add up to a lot over time. At some point people are going to get a bit sour on the idea of acting like an angle investor with tax dollars.
  - perl4ever 7y ago
    The problem with investing in a lot of low probability things is that when, say, 100 of them all fail, how do you know whether that's because they were low probability, or zero probability? Unfortunately, there's no limit to the number of very low probability things you can invest in, so nobody has the budget for them all.

ArtWomb 7y ago

I'll link to the NAS report on fusion findings. US benefits from ITER partnerships work both ways. But also recommends investments in smaller, less expensive, more innovative (and therefore competitive) designs. http://www8.nationalacademies.org/onpinews/newsitem.aspx?RecordID=25331 http://www8.nationalacademies.org/onpinews/newsitem.aspx?Rec... Just to put things in perspective. ITER budget is estimated ~$25B. Seeking to produce >500MW with a 12T magnetic field. Massive radius. Twice any previous tokamak size. Just look at the pictures ;) JET budget was ~$5B total. And achieved 25MW with a 5T coil. But without actually "burning plasma" (Max Q < 1.0) Princeton Plasma Physics Lab run by DoE budget is >$100M per annum. So this represents a significant challenge, and not just in fund raising. The most promising space might be in the design of "mini-magnets" https://nationalmaglab.org/news-events/news/lbc-project-world-record-magnetic-field https://nationalmaglab.org/news-events/news/lbc-project-worl...

dnautics 7y ago

technically, they are persuing a completely different mechanism. In my opinion it's totally bonkers, but in principle there is no reason why the spend/return should be correlated.