Talk-Polywell.org

This is more about a theory of EMC2's Polywell development plan than about Polywell theory per se, but we don't stand on protocol around here...

Referring to the picture on EMC2 FDC's web site,

which has a file name for WB-8, but a picture caption for WB-7.1...

Could it be that WB-7.1 will be tested in pulsed mode and WB-8(.0) will be tested in continuous mode? Hardware generally being the same, except the cooling pipe flanges will actually be put to use for WB-8? Apologies if this has already been thrashed about elsewhere.

I don't think so, but to add fuel to the fire, they are moving the vessel and lab, that could be to support a higher density power feed in addition to floor space than your typical sub-urban commercial rental space.

Maybe we be able to tell something better if we get a "new address" to google map.

I think WB8 is a pulsed machine. A derivative of this machine may be configured for steady state operation. I think this is referred to as WB8.1.
I suppose a P-B11 version would be WB8.2

Moving again? Anything might be read into that. It may be that they are more confident that they can produce fusion levels that will result in neutron radiation levels (even if well shielded) that make their neighbors and/or city planners nervous. Or it may have to do with utility power issues, space issues,security, bugets, desire to be closer to other scientific resources, etc. etc. Moving to Los Alimos Natl lab, where Nebel has links, or China Lake, which is more secure(?) and is close to Navy research facilities might be two possible destinations.
It could be a simple lateral shift for some of the above reasons, or it may be a move in anticipation of more intense efforts. That combined with what was said at a Navel confrence may hint that such is the case, or not

viewtopic.php?t=2454

Dan Tibbets

Dan,
8.1 is the improved drive version for PB&J.
I really do think they are going pulsed, maybe longer pulses, but it would seem that to get adequaute cooling for steady state would take a longer effort then the timelines conctracted. But again, unless we go in the lab and look, we don't really know.
To be honest, all the flanges on the WB8 graphic do make one wonder...it does seem they could support some form of cooling. it would also not be the first time they have run a cooled magrid device.

Could it be that WB-7.1 will be tested in pulsed mode and WB-8(.0) will be tested in continuous mode?

Yes to the former, I doubt the latter very much. The physics of concern are fast enough for anwers from pulses. Continuous operation cooling would be expensive and create design complications.

DeltaV wrote:
which has a file name for WB-8, but a picture caption for WB-7.1....

When EMC2FDC's web page was last updated, they had a statement under that picture regarding WB8 that made it sound like the WB8 work had already been successfully completed. I pointed that out to Dr. N. and he changed the verbiage to what it is now, but left the picture as it is. Oh well.

Then again, I guess I'd rather have him doing research than polishing the website.

Previously fuel was puffed across the chamber, like a smoker exhaling smoke across a room. As soon as you "gun up" the runs can be exponentially longer. As was said before cooling then becomes the limiting factor. OTOH diagnostic data just got a whole lot better than a puff of fuel floating across the chamber.

True, the gas puffer associated with the small size was the time limiting element of the experiments. I believe in WB 6 the magnets could be run for several seconds before the heat build up became too much. The escaping neutral and subsequent arcing limited tests to ~ 0.25 ms.
This was long enough for most of the physics. But, if the test could be run for ~ 20-100 ms, I'm guessing that simple neutron counts could show if thermalization compromised performance within a time frame expected to be longer than the fusion lifetime of the ions.
If the neutron counts started high, but dropped, this would probably indicate that significant thermalization times were longer than 0.25 seconds, but shorter than the length of the test.
With ion guns, WB 7.1 may have run for long enough to provide this data.
Even a tenth of a second would be extremely reassuring.

Dan Tibbets

@ Dan, yes, just gunning up should have yielded that data. I'm assuming that because WB-8 exists the WB-7.1. Gunned runs were productive. were those runs 10 times longer? Just enough to move a decimal pt?

Nebel did say they were getting "time domain information" on WB-7.

But I don't know if this means that they have the controls to actually operate at beta = 1 for any amount of time, rather than sweeping through it, or whether they just have better diagnostics. I strongly suspect it's the latter.

There also seems to be some expectation that the results are going to come out either “Woo-Hoo!” or “This thing stinks”. It would surprise me if we get either of those answers. It will probably be more nuanced than that. Among other things, we are going to have time domain information. That wasn’t available on the WB-6. There will probably be some surprises in there. In plasma physics, theory very seldom predicts things accurately ahead of time. It’s usually an explanation that comes after an observation. Elms in tokamaks are an example of this. Right now we don’t know these answers (so please don’t read more into this statement than is here).

I'm a little fuzzy on this aspect of operation, but my understanding is that they turn on the e-guns and the electrons build up through beta = 1 (i.e. n*kBolt*Te = B**2/(2*mu0) ), and that for continous operation (even milliseconds) they would need a control to limit the current when it reached beta=1 so it didn't blow out the field (i.e. they want to avoid n*kBolt*Te > B**2/(2*mu0) ). I guess I'd be surprised if they had a control sophisticated enough to detect beta=1 and respond in the necessary sub-millisecond time frame. Maybe MSimon can offer some guidance on how difficult/expensive that might be.

I know in puffed-gas operation the ions are pulled in; I'm not sure whether they regulate the ion guns' flow such that they maintain the desired 1/1E+6 difference or what goes on there exactly. Maybe someone else can clarify.

TallDave wrote:Nebel did say they were getting "time domain information" on WB-7.

But I don't know if this means that they have the controls to actually operate at beta = 1 for any amount of time, rather than sweeping through it, or whether they just have better diagnostics. I strongly suspect it's the latter....

The Beta=1 tests were done by sweeping the magnet current up so that at some point Bets= one was passed. and this was measured with a PMT (light output maximum). I think this was one of the measurements that Nebel was uncomfortable with. They apparently used a different modality and/or diagnostic which he was more confident of and apparently it confirmed the earlier findings/ intrepretation.

For the fusion tests, I'm uncertain what procedure was used, except that the magnets were turned on first. I have assumed the magnet current was fairly stable, but I could be wrong. The WB 6 report did say that the electron gun current stayed at ~ 40 amps well past the point where arcing started. The rest of the run away current must have been coming from the 12,000 volt charged magrid that was fed by capacitors. They (WB6) may have had fairly stable magnetic field and E- gun current conditions. What was uncontrollable was the flooding of neutrals throughout the system which led to the magrid casing (?) arcing to ground (structures). If this was the case, then ion guns would presumably have allowed significantly longer runs in WB7.1.

Dan Tibbets

TallDave wrote:Nebel did say they were getting "time domain information" on WB-7.

But I don't know if this means that they have the controls to actually operate at beta = 1 for any amount of time, rather than sweeping through it, or whether they just have better diagnostics. I strongly suspect it's the latter.

In one statement, I don't recall where, Dr. N. said their new instrumentation had microsecond sampling capability. This would allow 250ish samples in the time domain with a run time equal to WB6.

I hope WB7 could run longer, but I don't recall statements to make me think it would. Anyone?

That makes sense Kite.

Dan,

I'm skeptical WB-8 or any machine is capable of operating more than a few milliseconds without the aforementioned controls (Bussard mentions the need for "large controllable power supplies"), and adding cooling for runs of more than a few seconds seems fraught with so many challenges you might as well make a new machine.

I'm fairly confident the magnet current is constant, since it doesn't do any work. But if the e-gun current is constant, then how do they stop the well from blowing out? My understanding is the e-gun current should drop off as the WB forms (with peak confinement at beta=1) because the e-gun current is, essentially, the e-losses. If you're pumping in more electrons than you're losing, you're increasing the left side of the equation. So I'm thinking they're just blowing the well out every time.

I suppose you could run a smaller current to get a longer run... ah, but there must be a limiting factor at low beta -- you'd need to exceed the cusp losses pre-WB in order to form the WB. I wonder if there's a way to calculate the minimum current?
-------
The Valencia emitter current graph is confusing in this regard. It has a spike at 10.222, then it goes negative. The rest of the time it seems to be near zero. The fusion clearly happens between 10.221 and 10.222, based on neutron counts.

I'm not 100% sure what the emitter numbers mean. My guess is they turn it on at 10.221, it reaches 40A at 10.222, then the well almost instantly blows out, then it arcs, all before 10.224.

Wow. I really wonder what this looks like at microsecond scales. That would be really interesting to see. It looks like even for the emitters we only get 2 readings per millisecond for WB-6.

The arcs were from feedthrough
leads into the main vacuum tank and the tank walls, and had
nothing to do with the machine or its containing cage/shell.
This took place over 0.5-2 msec after puff-gas actuation, so
little time was available for true Polywell operation. The cap
drive current to the test system then ran away to over 4000 A
to this external feedthrough arcing, as the Polywell formed
and fusions occurred. This destroyed the well depth (due to
drop in drive voltage). However the system did run at
emitter currents (to the machine) of 40 A for about 0.3-0.4
msec, proving the basic concept.

I'm still not sure this gives us longer runs of "true Polywell operation" with ion guns. It's not clear to me whether the arcing ends the run before the WB blows out.

TallDave wrote:That makes sense Kite.

Dan,

I'm skeptical WB-8 or any machine is capable of operating more than a few milliseconds without the aforementioned controls....


I'm fairly confident the magnet current is constant, since it doesn't do any work. But if the e-gun current is constant, then how do they stop the well from blowing out? My understanding is the e-gun current should drop off as the WB forms (with peak confinement at beta=1) because the e-gun current is, essentially, the e-losses. If you're pumping in more electrons than you're losing, you're increasing the left side of the equation. So I'm thinking they're just blowing the well out every time. ...

I'm still not sure this gives us longer runs of "true Polywell operation" with ion guns. It's not clear to me whether the arcing ends the run before the WB blows out.

I'm speculating, but if the cusps open back up gracefully once Beta exceeds one, and especially if they open up faster than they closed as Beta= one was approached, then some constant electron current that provides more electrons than leak out at the beginning willl build towards Beta=1 and the best containment, then if there are excess electrons entering, Beta will exceed one and the containment will fall proportionately. IE: the Beta=1 condition is regulated . What would change would be the effective containment on the high side of Beta=one that would dispose of the excess electron current. Within limits, I think this might allow a constant electron current (in excess of what is needed to maintain simple cusp confinement) which would maintain the Wiffleball without fine tuning. Knobs to adjust the electron current would be most important for maximizing efficiency in a production machine. For quick and dirty research, such would not be mandatory.

I think this would be consistent with what was seen in the Beta=1 tests. The current was maintained, while the B field was swept up. The light emission maxed at Beta= one and dropped off on either side as the WB was compromised and the density dependent glow discharge intensity fell off on either side of Beta= one.
It would be sort of like an air tank with a small hole in it. Air flowing into the tank at a certain rate would build up pressure in the tank till the gas escaping through the small hole equaled the input. Any additional air input (within limits) would not increase the pressure, only increase the rate of air escaping through the hole. The dynamics are different, but I wonder if the results are ~ the same.

Dan Tibbets

I'm speculating, but if the cusps open back up gracefully once Beta exceeds one

I don't think it works that way Dan. Its more like overfilling a balloon. Think of beta =1 being the point where the balloon pops. The wiffle ball will have to operate at beta almost equal to but < 1. For beta < 1, the magnetic forces are greater than the plasma forces leaving the plasma no where to go but to stay confined. For beta > 1 the plasma forces are greater than the magnetic forces so the plasma can escape to the magrid.