New patent application, similar to polywell.

[D Tibbets]

OK, chrismb, I think we are on the same page on this. What remains foggy is the crossection / MFP relationship to the average temperature. At ~ 80 KeV the Coulomb collision crossection is only ~ 1-3 orders of magnitude greater than the fusion crossection, depending on whether you are talking about D-T of D-D. Based on your discription you could use an arbitrarily different MFP definition based on 1degree deflections but the increased frequency is balanced against the decreased energy loss (or scatter)per collision. . What is important is the temperature. Granted the tendency of an initial monoenergetic plasma species to thermalize will result in both increased energy and decreased energy particles, and my reading of the graphs suggest the average temp is below the mean, and these cooler ions will have increasing unfaverable effects on the energy equation as the Coulomb crossection increases rapidily in these cool ions.

The real question then is 1) what will slow this thermalization process to time frames comparable or longer than the expected lifetime of the ion (before loss to confinement failure or fusion).
And 2) What mechanisms are in place to recover some of the energy, hopefully most of the energy.
In the Polywell the expected ion lifetimes are only perhaps 10-20 ms, and annealing servers to slow the thermalization process. From the electrons perspective the lifetimes are perhaps 0.1 to 0.5 ms and recirculation helps with energy recovery and also thermalization- at least on the high side.

In a Tokamak, with it's relatively huge-containment times your analysis seems reasonable, but the thermalization timescales and modifications are paramount in machines which claim non thermalized conditions, such as these do. You can argue about the input losses but it has to be in the context of the thermal spread present (and energy recycling). Arguments about the thermalized performance is a different beast.
.
Non thermalized assumptions is the basis of these machines claims for advanced fuel capabilities. If thermalization cannot be controlled, it is accepted that these machine will probably be limited to D-T fusion (at best) because of the large fusion gain, and the higher fusion crossection relative to the average coulomb collision crossection obtainable in a thermalized machine.

Then you can start arguing about Bremsstrulung, ignition considerations, etc...

Dan Tibbets

[D Tibbets]

A somewhat different issue that incorporates these various crossection rates versus temperature is x-ray losses.. In one of Bussard's papers, it was pointed out that Bremsstrulung not only rears it's ugly head at high temperatures, but also at lower temperatures. D-T Bremsstrulung exceeds DT fusion at temperatures below ~ 5 KeV. The Bremsstrulung increases at a steady rate of ~ 3/4 (or was that the 2/3rds?) power of the temperature. As the fusion crosection is quite steep in this region the Bremsstrulung losses will dominated till the fusion croscection exceeds ~ 0.01 Barns. For D-T this is ~ 5 KeV, I assume this is based on a monoenergetic energy, a thermalized plasma would need to be pushed a little further. The cutoff was ~ 15 KeV for D-He3. The value for D-D wasn't given, but would probably be ~ 7-9 KeV. As the fusion crossection levels off at higher temperatures, the Bremsstrulung losses again catch up, so there is a well defined window where fusion output can exceed input losses, even assuming perfect confinement. With monoenergetic plasma these cutoffs are well defined, with thermalized plasmas the picture is foggier, and it is even more foggy when you add in tricks that the Polywell claims. Energy recovery/ conversion plays a role again. At ~ 25% thermal cycle recovery the numbers are modified some. 80- 90% recovery of the x-ray energy and/ or the fusion energy modifies this even more.

Dan Tibbets

[chrismb]

The real question then is 1) what will slow this thermalization process to time frames comparable or longer than the expected lifetime of the ion (before loss to confinement failure or fusion).
And 2) What mechanisms are in place to recover some of the energy, hopefully most of the energy.
In the Polywell the expected ion lifetimes are only perhaps 10-20 ms, and annealing servers to slow the thermalization process. From the electrons perspective the lifetimes are perhaps 0.1 to 0.5 ms and recirculation helps with energy recovery and also thermalization- at least on the high side.

Yes. I agree these are major questions. I was hoping Alex would engage and discuss those questions, wrt his device.

As you know, I do not hold much sway with the idea of 'annealing', but the prospect that there is any chance whatever of this actually happening in polywell is the only thing keeping me watching.

Incidentally, *this* is the data I would want to see from a 'working' experiment. Any experiment that cannot state the energy confinement and confinement time I would judge as simply not measuring the basics. Two simple numbers. A value in Joules, and a value in seconds. These are key. Counting neutrons is hopeless as they can pop up from many difference processes and mechanisms.

Alex, if you have those two figures duly measured, I'd sure be greatful if you can let us know them.

[Giorgio]

I was hoping too, but it looks like his main interest now is in looking for financial help to keep his research going.
Can't blame him, I will do the same in his situation.

Maybe he will drop back to discuss some results sooner or later.

[AlexK]

Hi,

I've been away and also unfortunately don't have the time to get involved in the back and forth detailed discussions on this website. Our new device is only just running and we don't have energy and particle confinement times. It is also a tiny prototype with voltages not exceeding 5kV, just to prove the principle - so we are not making neutrons. People here should know however that Linear Electrostatic Traps can contain keV ions for minutes (!), as of 2010 (but of course at low density), for example look here (Rev Sci Instrum. 2010 May;81(5):055105.) or just google it. There are some free papers available.

It is true that Coulomb Scattering is the limiting factor in IEC devices, at least the ion-accelerating kind (Polywell is something else, more like a thermal plasma confinment scheme). It is also true that the energy loss (ions getting lost from scattering) and fusion gain are close rivals, i.e. a Q of 1000 will never be possible. A Q > 10 is, however, and not just for DT. Bremsstrahlung is not a concern in our device since electrons are cold/warm, and Todd Rider makes several erroneous assumptions in his famed dissertation.

At the moment we are trying to stay clear of making bold pronouncements regarding net fusion power production, and getting entangled in the political and sensationalist aspects of this research...

Cheers,

Alex K.

[chrismb]

People here should know however that Linear Electrostatic Traps can contain keV ions for minutes (!), as of 2010 (but of course at low density), for example look here (Rev Sci Instrum. 2010 May;81(5):055105.) or just google it. There are some free papers available.

Low density indeed! In fact; non-collisional, no less!

Whereas a fusion system relies of collisionality to get reactions. So this is an impossible dichotomy - you want as many collisions as possible in a fusion reaction, but not at a rate of lost net particle power(wrt input energy) where you end up with a population of ions that never reach a high enough energy to improve their fusion probabillity.

So starting with an ion trap is a great start, but the question then moves immediately on to how many collisions/s are there, and what does this translate into for viable energy outputs?

I think you are perfectly correct to stay clear of making any excess claims, which of course you are in no position to do yet anyway without data, but actually I float around these sites on the basis that I don't have the time to spend not getting involved in discussions and wasting time later on flogging a bad experiment that someone else can spot an obvious weakness in that I have missed.



...



One observation of running at 5kV - if a device has any chance of operating as a viable energy-producing device, you should be able to detect DD neutrons from such a thing at 2kV. It's just a matter of scaling - if you can generate 1 watt of neutrons at DD peak, then you should be able to detect a few hundred, or thousand, neutrons/s at 2kV because the DD reactivity between those energies is around 9 orders of magnitude.

If you can't generate detectable neutrons at 2kV then you'll never be able to generate even a few watts of neutrons at DD peak, based on the cross-section scaling alone.

Detecting DD fusion neutrons at 2kV is, as it were, a condition for showing that your device can scale up drive energy.

[D Tibbets]

Um... Im not sure Chrismb's assertion in the above post is entirely true. At 2 KeV the fusion crossection ratio certainly applies. But, so do fusion crossection vs Coulomb crossection ratios. If the anticipated fusion rates come from beam-beam collisions, then these will be much rarer at these energies. Bean- background, or purely thermalized collisions would be much greater. The scaling would probably need to add a few more orders of magnitude to the difference. Of course this assumes the higher energy regime is beam- beam (say at 80KeV). Also, the density is critical. Even if the plasma was thermal in the PZL1-X (the copper block machine)tested by EMC2, they reported detectable D-D fusion, even at ~ 0.3 KeV.- presumably due to the Wiffleball effect with a 3.5 T B field resulting in density increases by ~ 1000X or more with resultant fusion rates > 1,000,000 times higher.

I am also, not sure why AlexK feels that the Polywell is not a typical electrostatic accelerated machine. Perhaps he bases his position on a disbelief in Annealing.

Dan Tibbets

[chrismb]

Um... Im not sure Chrismb's assertion in the above post is entirely true. At 2 KeV the fusion crossection ratio certainly applies. But, so do fusion crossection vs Coulomb crossection ratios.

Well, you are right in that comment, but wrong to relate it to whether or not you'd measure neutrons.

Energy lost through the higher Coulomb collision rate would, indeed, lead to a bigger loss of energy from the system, but that should simply mean you will end up pumping in more energy to get enough ion population to a fusible energy. The neutron rate should still be detectable at the same rate, but your comment relates [only] to the efficiency of converting input energy to rate of collisions. [I wil agree there are a few caveats on that, but I'm stated the general principle.]

[Giorgio]

Bremsstrahlung is not a concern in our device since electrons are cold/warm, and Todd Rider makes several erroneous assumptions in his famed dissertation. .

Now, this is something I would like to discuss and have more info from you.

[AlexK]

well, ok, I'll bite: I'm just saying that electrons at 1 keV don't radiate terribly much, but are warm enough to not provide undue collisional drag on ion beams passing through. Pretty straight forward, I think.

[choff]

I've often wondered if given the original secrecy of the Emc2 concept that a key component might not have been disclosed in Dr. Bussard's Google talk or anything since. If that were the case then the Polywell might be even more similar to FP's Marble.