Wednesday, February 4, 2009

So I'm Discussing With Art Carlson

So I'm discussing with Art Carlson whether it is worth it to look deeper into the Polywell Fusion Reactor design and do some experiments with superconducting coils and continuous operation of a test reactor. Said experiments to cost about $10 million. Well Art is sure that the explanation that Physicist Robert Bussard gave for how the device works can't possibly be true and it is all just a bunch of believers. Cultists if you will. He did not hold back when expressing his views.
StevePoling wrote:
Can anyone articulate an experiment that would falsify either proposition? I mean something cheaper than building a fully-operational Wiffleball-N?
I spent some more time pondering this. I was thinking in the direction of leaving out the cusp itself and just investigating a pencil of plasma propagating along a field through hoops of various potential. Then I realized this is pointless because whiffle-ball theory is not falsifiable.

I mean, suppose I set up an experiment involving cusp physics and electric fields and I showed that it all worked as I expected. What would the polywell proselytes say? That the real polywell has (unspecified) non-Maxwellian effects that my setup didn't take into account. That is basically the last answer I got from Rick Nebel. Of course I can't refute that because nobody has ever said what those effects might be in detail. Maybe if I worked real hard for a year or so (Are there any volunteers to pay my salary?), I could prove a fairly general theorem that would rule out a large class of options. (My shining example for this type of calculation is Todd Rider.)

Basically, there is no whiffle ball theory, only some handwaving with manifest inconsistencies. On the experimental side, there is no published, robust evidence that anything unusual is happening at all. What are we doing here?
But it is falsifiable at least ultimately in an engineering way. Either you get more power out than you put in or you don't.

For $10 million we build the Super Conducting coil job and that should tell us if a power producer is possible. It should also be possible to measure the wiffle-ball. Lasers. Microwaves. Field probes. Whatever.

Or we might go with a lower cost liquid nitrogen cooled copper magnet coil version. It would have a much lower magnetic field than a superconducting coil. But you can build it faster. Vary the current through the magnet coils and see how the losses scale.

I must say though that I'm starting to feel like a tokamak guy: "there are problems that can only be worked out at the next larger level". It must be a plasma physics disease.

5 comments:

Tom Cuddihy said...

From the discussions I've seen the "wiffle ball theory" itself is not actually necessary for a polywell to hit breakeven, although it would make it easier.

Is that correct or have I missed some key discussions?

I'm not sure why it matters if the wiffle ball theory works or if that formulation is an oversimplification of the processes at work.

What matters is --will effects observed on WB-7 scale up or not? that's the question.

M. Simon said...

The wiffle ball is supposed to answer the question: why are the losses so much smaller than expected?

Roj said...

I occasionally am unreasoningly prone to wildly thinking we need more than one device being built or operated.

When undergoing these bouts I believe we need an Ln2 cooled device, a Whopper WB-100, and a dodec.

What do I know?

Loren said...

The wiffle ball, as I understand it, is a response by the magnetic field to the negative charge in the center. The electrons push back on the magnets, stretching them out and shrinking the cusps.

Provided you can look at the shape of the magnetic field, coudln't you simply make a small polywell(perhaps even use some of the earlier machines), put a negative grid in the middle like in a Farnsworth(or is it the other one?), and measure the field at varying levels of charge in the grid?

The grid wouldn't really matter in this case, since you're not running ions in this machine, you're simply looking at how the magnetic fields respond to the electrical charge in the center.

M. Simon said...

Loren,

That experiment (characterization of the Wiffle Ball) is currently being done on WB-7.1.