Jeff Bezos funds General Fusion

[Ivy Matt]

Courtesy of mvanwink5 at Talk Polywell:

http://blogs.forbes.com/clareoconnor/2011/05/05/amazon-billionaire-bezos-backs-nuclear-fusion-in-19-5-million-round/

Amazon.com founder Jeff Bezos has contributed to a $19.5 million round of funding for Canadian nuclear fusion startup General Fusion, marking the web billionaire’s first major investment in nuclear energy.

Bezos is backing the British Columbia-based nuclear power developer via his personal investment firm Bezos Expeditions, according to General Fusion’s announcement. Neither company has disclosed how much of that $19.5 million came from Bezos. Other investors in the Series B round include Cenovus Energy and Braemar Energy Ventures.

For those who are unaware of General Fusion, their website is here. Steampunk indeed.

I’m not sure how Bezos expects to fit one of those things into his rocket. ;-P

[Tulse]

Boy, of all the possible “alt.fusion” contenders, I would have thought General Fusion would be at the bottom of the list. The approach requires a huge capital outlay to build the device, it is mechanically very complex, it still ends up with radiation issues, and as far as I know it is very far from creating a working prototype. In other words, it is everything that FF, Polywell, and Tri Alpha (among other approaches) are not.

If I had $20 million, instead of putting it all into one highly speculative basket, I’d give $5 million each to those three different approaches, with a promise of additional $5 milllion to the first group to show breakeven. So much of this basic research is done on a shoestring, and given that all these approaches require far less capital than tokomaks or ICF in order to produce useful information, I’d think spreading around a bit of seed money to lots of approaches would be a more efficient way to produce the desired goal.

[AaronB]

Wow. Maybe they’re closer to achieving the fusion dream than we’d thought. 😉 LPP would gladly take $5M in seed money, and do a lot of good things with it. Forget $5M, we’d take $1M and see how close we could come to breakeven!

[zapkitty]

AaronB wrote: Wow. Maybe they’re closer to achieving the fusion dream than we’d thought. 😉

…
… …
… … …

AaronB wrote: LPP would gladly take $5M in seed money, and do a lot of good things with it. Forget $5M, we’d take $1M and see how close we could come to breakeven!

What could be causing investors to shy off of aneutronics?

Polywell seems to hover on the edge of a government intelligence black hole… or at least Nebel and company have left that impression.

Tri-alpha is secretive but I’m sure having the name Jeff Bezos would result in a red-carpet guided tour complete w/ plushy CBR souvenir.

FF is an open book.

Musing aloud, the only outstanding issues with FF that wouldn’t be solved with net power would be capacitor wear and, to a lesser extent, electrode wear.

LPP seems to have hedged against ‘trode wear by increasing the expected changeouts to a monthly cycle.

Any alternative solutions to cap wear? There’s always that big coil… 😉

… I know, I know…

… but this really does seem like the money is being steered to the least likely candidates… again…

[Ivy Matt]

Tulse wrote: If I had $20 million, instead of putting it all into one highly speculative basket, I’d give $5 million each to those three different approaches, with a promise of additional $5 milllion to the first group to show breakeven. So much of this basic research is done on a shoestring, and given that all these approaches require far less capital than tokomaks or ICF in order to produce useful information, I’d think spreading around a bit of seed money to lots of approaches would be a more efficient way to produce the desired goal.

I don’t have any inside knowledge or anything, but it seems to me that Tri-Alpha doesn’t have much trouble raising funding and EMC2, although it doesn’t get a whole lot of funding, seems to be receiving a sufficient amount so far. As long as the Navy is getting the results it wants, I imagine they’ll have a safety net.

AaronB wrote: Wow. Maybe they’re closer to achieving the fusion dream than we’d thought. 😉 LPP would gladly take $5M in seed money, and do a lot of good things with it. Forget $5M, we’d take $1M and see how close we could come to breakeven!

I don’t know, maybe you’re going about this the wrong way. $1 million is a paltry sum. Maybe you should be telling people you need to raise $200 million to develop the commercial demo device, and investors will be quicker to open their pocketbooks. 😉

zapkitty wrote: Tri-alpha is secretive but I’m sure having the name Jeff Bezos would result in a red-carpet guided tour complete w/ plushy CBR souvenir.

Tri-Alpha has Paul Allen, though.

zapkitty wrote: FF is an open book.

I hate to say it, but that might be a problem. I don’t know why it should be, though. At least LPP has a patent. It’s my understanding that General Fusion hasn’t been granted any patents yet.

zapkitty wrote: Musing aloud, the only outstanding issues with FF that wouldn’t be solved with net power would be capacitor wear and, to a lesser extent, electrode wear.

Some people would also argue that both methods of energy capture are unproven, and the actual amount of energy captured will be much less than is supposed. I really don’t know enough to opine on that, though.

zapkitty wrote: but this really does seem like the money is being steered to the least likely candidates… again…

The way I’d put it is that the approaches that seem to me most likely to bear fruit soonest are receiving the least amount of funding. Then again, from what I’ve read, General Fusion is aiming for a demonstration of net power in 2013. Assuming they actually achieve it, the only ones I know of who I think could beat them are LPP, NIF, maybe EMC2, and possibly (who knows?) Tri-Alpha.

[AaronB]

We can play the secretive and expensive cards too if it will help. It makes it that much easier when things don’t go according to plan, and besides, the mysterious dark horse or black box project is alluring to some people.

[Ivy Matt]

Well, I hope that won’t be necessary—at least the secretive part. General Fusion is actually not all that secretive—certainly not compared to Tri Alpha, or even to EMC2 these days. In fact, the thing I’ve noticed about them is that, for whatever reason, they seem to get their share of publicity. See here for a recent example. I have, however, noticed that even the more in-depth articles don’t seem to delve into what exactly they’re working on at the moment. Whether that’s a conscious policy by General Fusion or simply the reluctance of reporters to delve too deeply into trivia, I would be interested to know.

By the way, regarding General Fusion’s funding, apparently they need $50 million to demonstrate net gain. Of that $13.9 million (Canadian) is publicly funded through Sustainable Development Technology Canada. According to Next Big Future, they will need a further $300-$500 million to go into commercial production. The same Next Big Future article also has a comparison of funding raised for Tri Alpha, EMC2, and LPP.

I would like to amend my statement in my previous post to: “the approaches that seem to me most likely to bear fruit soonest with the least amount of money are receiving the least amount of funding—but less than they need”.

[Tulse]

General Fusion’s approach seems absurd to me — not physically impossible, just absurd in a Rube Goldberg, “I can’t believe how silly that thing sounds” way. And I can’t imagine that they will have breakeven by 2013, largely because I don’t think they can even physically build the ridiculously complicated reactor by then, with its 200 pneumatic pistons firing simultaneously to send shockwaves through a sphere of spinning liquid metal to compress a plasma toroid, all in a device several stories tall, and requiring absurdly precise timing. There is no way that anyone could build such a large, mechanically complex device requiring precise tolerances and have it researched and characterized and debugged and working in two years. Absurd.

And, out of all this, they’re still just heating up water to produce steam to run a turbine to run a generator, like we have for about a century.

This is fusion done by orks.

I really do wish that LPP could find its own angel investor from among the rich digital entrepreneurs. Tri Alpha has Microsoft’s Allen, and now GF has Bezos — perhaps LPP could interest Elon Musk in a potential power source for the next stage of SpaceX vehicles and Tesla cars.

[AaronB]

I like the General Fusion guys, but I think their approach will run into some difficulties if I understand it correctly. The spinning liquid metal will have two open “whirlpools” at the ends, one on the top and one on the bottom. Because of gravity applying pressure on the spinning liquid, the top whirlpool will be larger than the lower one. When the hammers strike the outside of the vessel, the shock waves won’t go straight in. They’ll radiate inward in all directions, with the idea that the centrally-directed waves will reach the middle at the same time and compress the plasmoids that are shot into the middle (which isn’t the exact middle) through the middle of the whirlpools. However, the shock waves will hit the surface of the whirlpools before they get to the middle. The shock waves will be so powerful that they will cause the surface of the whirlpools to virtually explode in a mist of metal droplets, which will interfere with the plasmoids as they come rushing in. Also, shock waves travel at different speeds through the liquid at the top of the device compared with the bottom, because of the pressure differences caused by gravity. Turbulence will also cause the shock waves to get distorted, and there’s going to be a whole lot of turbulence from the liquid being pumped in at certain points, like in a jetted tub.

I’m not saying it’s impossible, but it would be extremely difficult to get the shock waves to combine at the center of a perfectly still pool, let alone a spinning, pumping, pounded, asymmetrical double whirlpool that has metallic mist flying out of both ends. Imagine getting a small plastic swimming pool, filling it with water, then inviting all of your friends over to kick the sides at exactly the same time, in hopes that the center of the water will fly up so suddenly that it could launch a floating ping-pong ball into orbit. It is reasonable to guess that you and your buddies would get soaked while the ping-pong ball remained fairly well earth-bound. It would be fun to watch, record, and post to YouTube.

Assuming they work out these minor problems, then they still have to deal with the hard stuff, like X-ray and other plasma cooling by the liquid metal, dealing with the fusion products, steam cycle inefficiencies, etc. They do get bonus points for creativity and thinking big. Who knows? They might pull it off.

[zapkitty]

AaronB wrote: I like the General Fusion guys, but I think their approach will run into some difficulties…

Some “Well, buts…”

The basic concept has been known since Trinity and obviously they are going to try to use precision control of the pistons to compensate for any asymmetries of the central vortex. Indeed, their whole shtick relies on that level of control.

As for the extent of those asymmetries? The amount of the widening at each end of the central vortex, and thus the cumulative lensing effects would be determined by how close to full the chamber is at the start and the fluid rotation rate… but that again falls into the realm of calculation and piston control.

They are dancing closer to the edge of bomb code than any other fusion contender except NIF but there should still be plenty of daylight between what they want to do and the verboten zone. (Better be… the last thing the world needs is the oligarch’s favored global reservoir of their poison declaring that fusion is ITAR’d even if Canada and everyone else has it…)

But I don’t think the mist should be as big of a problem as you surmise because the plasma will already be in the middle of the central vortex by the time the wavefronts converge.

[Tulse]

There is just so much that has to go right, and on a mechanical level, in order for their approach to work. It has tons of moving parts that have to work repeatedly and with very precise timing, whacking a couple of tons of spinning molten lead. It’s nuts — it’s by far the most mechanically complex approach to fusion out there, and when you’ve got so much precision mechanics involved, you’ve got a whole boatload of variables. FF and Polywell are elegant — the devices are fairly simple, with almost no moving parts, and the domain of relevant variables is much more straightforward to characterize and manipulate.

I wish the General Fusion folks all the best, and I’d be quite delighted if we get fusion from such a steampunk-y device. But I’m not holding my breath, and I really wish Bezos would have spent his money on more likely candidates who can test their approach much more cheaply.

[Alex Pollard]

Agree with everything above about the sheer implausibility of General Fusion’s timetable, and quite frankly, the whole concept.

Bezos has a lot of cash but that says nothing about his investment skills.

It is ironic that someone who is blowing a huge stock bubble

http://truthingold.blogspot.com/2011/04/lets-talk-about-real-bubble-amzn.html

is unable to recognise when an outfit he invests in is blowing smoke.

It is really is tragic that our financial system allocates so much capital to fraudsters, who then squander it.

[zapkitty]

Finally went through the GF website, and picked up a couple of mitigating points they left lying around, namely that the “hang time” for the plasmoids in the central vortex is in the range of “hundreds of microseconds”, and the pulse rate for their default 100MW setup is… one Hz.

(Indeed, the whole setup reminds me of Gally’s Hertza Haeon…)

Let’s pile on the penalties and omit the wave going supersonic… sound v in STP lead is ~1.3 km/sec, how much less for liquid lead? And the lithium added to the lead?… and then the vortex…

But even given an WAG penalty of 300-400 m/s the travel time for the wave to cross 1.5 meters to the center of the sphere would still be an order(s) of magnitude shorter than the plasmoid hang time. And thats even when handicapping the wave v big time.

Add a whole second for the vortex to recollect any “mist” and this would seem to make both timing and the mist a non-issue.

[delt0r]

*shock* waves travel faster than the speed of sound. Often many time faster. The general Idea of GF is magnetic target fusion. The idea has many merits and sits somewhere between magnetic confinement and internal confinement. If i was going to bet on a dark horse to win the fusion race, it would be GF. There are some papers around too if you look around. Don’t have my references handy.

[Rezwan]

delt0r wrote: If i was going to bet on a dark horse to win the fusion race, it would be GF. There are some papers around too if you look around. Don’t have my references handy.

Can you get those references handy and draft an article for our “contenders” section? That would be great!

[Author]

Posts