Hacker News Reader

Comment by mcv | Hacker News Reader

There's that famous rule of thumb: if it sounds too good to be true, then maybe it is. Lockheed Martin's cheap, compact fusion reactor does sound too good to be true. Then again, if it works, that's enormous. And the massive, expensive ITER can't possibly have the same kind of impact on our energy. I want Lockheed Martin to be correct.

But are they? If this approach has been tried before without success, and no fusion scientist considers this likely to work, then I wonder why Lockheed is even trying this. On the one hand I think Lockheed must be on to something. On the other hand, if they don't have any good data to share (or are unwilling to share it), then why are they going public with this at all? If it makes sense to keep it secret because it's a secret, commercial, profit driven project, then why not just keep it secret?

Why seek publicity when they don't want to share anything real? Is the share price down? Do they need extra funding?

View on HN

All reasonable points, and there is plenty of good reason to be skeptical. On the other hand, Lockheed Martin is one of the largest defense and aerospace contractors out there, with a solid track record for bringing to market sophisticated aerospace solutions. They also made ~$3B in profit last year, so they are probably ok on the money front. At the least they deserve a bit more benefit of the doubt than, say that Italian guy running around claiming cold fusion.

On the other hand, they are the makers of F-35, no? I think a fair amount of skepticism is needed.

On the other other hand, they are the makers of the SR-71 Blackbird, so I think a little benefit of the doubt is needed. ;)

That was designed in the '60s by a prolific aircraft designer. It was quite a different time.

But they continued that with e.g. the F-117 stealth fighter(-bomber), after Kelly Johnson was no longer running the show. The story of the prototype's development is stunning, see e.g. http://www.amazon.com/gp/product/0316743003/

As for the F-35, the TFX like requirements are its biggest problem (https://en.wikipedia.org/wiki/General_Dynamics_F-111_Aardvar... see also The Strategy of Technology if you want the whole picture).

The F-35's issues have been ascribed to the Marine Corps' demand for VTOL.

https://medium.com/war-is-boring/fd-how-the-u-s-and-its-alli...

The best thing you can say about the F-35 is that it's a perfect example of design-by-committee.

Absolutely. Had it been anyone else, it'd probably have been written off right away. But it's Lockheed Martin, so everybody assumes there's something to it.

Still, why go public on this without any data? Not publishing at all because they want their competitive advantage I'd understand. Publishing data I'd also understand. But a spectacular hype story with nothing to back it up? That suggests it'd about something other than just the science or the product.

I still hope it's real, though.

Maybe it is a glimpse into internal politicking and budget wars. It's harder to cut funding to a project like this if people are beating down your door asking about it. But they'll only do that if they've been given some tantalising details about it.

It's almost certainly a portable black hole for funding applications.

Last time this was discussed here the most common problem cited was the emission of fast neutrons, which move fast and break things, especially including the superconductors which are supposed to be part of the magnetic containment.

Regardless of the success of this particular experiment, you've got to hand it to the US when it comes to fusion research.

US is :

Funding ITER (founding member of ITER)

Funding Z-Pinch

Funding Polywell

Funding steam-based fusion (yes, really)

Funding laser fusion

Funding this new approach by Lockheed Martin Skunkworks

Funding 3 other inertial containment fusion approaches

Over 30 small scale experiments that don't appear to currently justify bigger investment.

The EU (which has ~20% more GDP than the US):

Funds ITER ...

Funds 2 other tokamaks (exactly the same approach as ITER, but smaller scale, and they're being defunded, most people claim because of ITER)

... that's it

Japan

Also supports a number of fusion research paths.

China

Funded a single Tokamak that is reportedly decommissioned by now. One might add that the organization that built it is suspected of forging their results. The end.

Most fusion approaches, including ITER, can only work if our current physics theories are flawed. Polywell, to a lesser extent Z-Pinch and several others, for example depend on the ability to keep a gas non-thermalized for a relatively long time (meaning molecules in a single volume of gas having different temperatures for longer than it takes them to move from one side of the gas volume to the other). Violates thermodynamics as it currently exists. If ITER generates the Neutron pressure it is predicted to generate, it will collapse in a matter of hours at best (which is were most sources got the "fast neutrons destroy matter" criticism. And yes, it's of course true for all fusion approaches. BUT it only matters if it destroys matter fast enough. For example, cosmic radiation destroys any kind of matter a lot better than even the worst fusion reactor. And you're exposed right now ! Nothing can stop them ! Run for the hills, except that even hills don't stop all of them. However, the rate of damage is so low, it would take millions of years to destabilize the head of a needle. The same is true for small fusion reactors).

The reason I mention this is that this is the criticism against a lot of the fusion approaches. Z-pinch cannot work because it violates ... Polywells can't work because gases thermalize whether or not you have big electrical fields fooling around ... of course, similar (and "equally" valid, in other words, with an unknown level of validity since we haven't tried) criticism exists against ITER. ITER is simply the bet that what you might call "the scientific community of physics" feels most comfortable with.

When I was taking philosophy, one remark stayed with me. On many occasions in history the scientific community has rallied around certain theories. Generally "the next theory" (e.g. relativity versus Newtonian physics) was available for decades before it became popular. Every single time the scientific community rallied around one of the theories and not once, not one single time, did they rally around the right theory (why ? Because almost every single time they rallied around the old theory, even in the presence of known flaws like the black body paradox. Scientists are no different from the rest of us : if possible, they'd like everything to neatly stay just the way it already is).

Given that, the US and to a lesser extent Japanese approach to fusion appears to be the right one. Maybe they should do more, but not by increasing ITER's funding. Frankly, if there is a good way to do fusion, the US deserves to find it, and frankly, the EU does not.

> Most fusion approaches, including ITER, can only work if our current physics theories are flawed.

Yeah this isn't true at all. Moreover, the US isn't funding fusion well at all - it has repeatedly threatened to pull out of ITER, was late to come on board, and has thrown only token amounts of cash at other projects. The Z-Pinch also is not a fusion device, and NIF was built to study atomic weaponry implosion physics, with inertial confinement thrown in for good publicity more then anything else.

Moreover, ITER will work perfectly well, and Q > 1 fusion has been demonstrated in previous, smaller tokamaks (it is just not a practical power source). Engineering problems are very different to "fundamental physics", and there's no particular reason to think that we can't build a suitable neutron absorber for a tokamak, just that it's tricky.

I try to remind myself that the CFR vs ITER, isn't much different in scale of leap than what Craig Venter pulled vs the government's genome project (in the sense of it being rather breathtaking as a leap forward compared to the backwards prior approach).

That is to say, such leaps over what people commonly think is the required approach, are in fact possible. It reminds me to leave my mind open to something like the CFR being possible.

This is a wildly inappropriate comparison. For one thing, Craig Venter didn't necessarily produce a "good" genome, and his approach wasn't objectively better at the time.

For another, it was hardly a problem of engineering at that point, the basic chemistry was well known and well understood. It was very much a problem of data-processing and logistics to get it all done though.

They are going public and not giving any specifics, because they want someone else to foot the bill.

They probably didn't make much progress in the last couple of years, and costs and risks are still to high.

>There's that famous rule of thumb:

No. It says "If it sounds too good to be true then it probably is." I am very hopeful though.

Reply on news.ycombinator.com