LANL Helps Polywell

Point out news stories, on the net or in mainstream media, related to polywell fusion.

Moderators: tonybarry, MSimon

MSimon
Posts: 14334
Joined: Mon Jul 16, 2007 7:37 pm
Location: Rockford, Illinois
Contact:

LANL Helps Polywell

Post by MSimon »

*

http://iecfusiontech.blogspot.com/2009/ ... ywell.html

*

Rick Nebel has some kind words about LANL. Also some tit bits (I love Brit English) about the experiments.

The full article was published in a New Mexico Paper and in Business Week. I give the Business Week link at IEC F. It has some othe LANL stories which I left out from the above.
Engineering is the art of making what you want from what you can get at a profit.

Aero
Posts: 1200
Joined: Mon Jun 30, 2008 4:36 am
Location: 92111

Post by Aero »

Interesting tid bits.
Aero

JohnP
Posts: 296
Joined: Mon Jul 09, 2007 3:29 am
Location: Chicago

Post by JohnP »

Interesting, their use of cameras. I'm not up on plasma instrumentation, but I woulda figured some kind of electronic probe would be better for nanosecond timescale data gathering.

MSimon
Posts: 14334
Joined: Mon Jul 16, 2007 7:37 pm
Location: Rockford, Illinois
Contact:

Post by MSimon »

JohnP wrote:Interesting, their use of cameras. I'm not up on plasma instrumentation, but I woulda figured some kind of electronic probe would be better for nanosecond timescale data gathering.
Maybe. But also maybe seeing what the fusor guys call "star formation" is helpful. And given the pulsed nature of the experiments that star is only going to be in evidence for a millisecond (more or less).
Engineering is the art of making what you want from what you can get at a profit.

Jccarlton
Posts: 1747
Joined: Thu Jun 28, 2007 6:14 pm
Location: Southern Ct

Post by Jccarlton »

There might be some other issues not related to the plasma. For instance at my company we were having arcing problems in vacuum. Fortunately we run for long enough to see the arc without needing a special camera, but running ion pulses probably requires a high speed camera to event see the event. Tracking down wild arcs and be a frying pain.

chrismb
Posts: 3161
Joined: Sat Dec 13, 2008 6:00 pm

Post by chrismb »

But why does WB7 run such transient events? What theory of waffleball formation covers such transience?

You need to have an idea what you're looking for before you look for it - is this just a hunting exercise, saying "blimey, we haven't got a slightest clue what's going on with this bang-shoot, but we've spent millions over decades, so we now need to throw the fastest analysis we can at it to give us a clue what the hell is happening!"?

Surely after 20 years of running these things based on a theory that it will be a continuous process that there should be some sort of observable 'continuous process'!?? This doesn't sound like progress to me....

KitemanSA
Posts: 6179
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Post by KitemanSA »

Remember folks, part of their work this time around is about high temperature nubs. Could they be trying to observe the heat-up pattern of the nubs?

93143
Posts: 1142
Joined: Fri Oct 19, 2007 7:51 pm

Post by 93143 »

chrismb wrote:Surely after 20 years of running these things based on a theory that it will be a continuous process that there should be some sort of observable 'continuous process'!?? This doesn't sound like progress to me....
a) WB-7 isn't actively cooled and can't run continuously at fusion conditions. Hey, at least it handles pulses better than WB-6 did...

b) Even with sub-millisecond pulses, the wiffleball (not waffleball) should essentially reach steady-state. Bussard stated this a long time ago. The physics are very fast.

c) I was under the impression WB-6 and WB-7 were both capable of running for longer times at low field, say with a helium test plasma like the one shown on the EMC2 website.

Jccarlton
Posts: 1747
Joined: Thu Jun 28, 2007 6:14 pm
Location: Southern Ct

Post by Jccarlton »

chrismb wrote:But why does WB7 run such transient events? What theory of waffleball formation covers such transience?

You need to have an idea what you're looking for before you look for it - is this just a hunting exercise, saying "blimey, we haven't got a slightest clue what's going on with this bang-shoot, but we've spent millions over decades, so we now need to throw the fastest analysis we can at it to give us a clue what the hell is happening!"?

Surely after 20 years of running these things based on a theory that it will be a continuous process that there should be some sort of observable 'continuous process'!?? This doesn't sound like progress to me....
The wiffleball is run for short periods to prevent the power supplies and magnets from burning themselves in cinders. The WB6 magnets were nothing more than wire coils inside stainless steel toroids and I believe the WB7 magnets are essentially similar. As they are they can't handle the loads required for more than a very short time. Haven't you read any of the papers and WB 6 reports. Because the program has been run on such a shoestring for all these years Neither Dr. Bussard or Dr. Nebel has had the kind of resources to design and build power supplies and magnets that could handle the full load of continuous operation, not a small design process in of itself. Anyway arcing only lasts a very short time and might not be visible without a high speed camera and once the arc occurs its over anyway.

Art Carlson
Posts: 794
Joined: Tue Jun 24, 2008 7:56 am
Location: Munich, Germany

Post by Art Carlson »

JohnP wrote:Interesting, their use of cameras. I'm not up on plasma instrumentation, but I woulda figured some kind of electronic probe would be better for nanosecond timescale data gathering.
Despite being qualitative and ambiguous, cameras are incredibly useful. The most common mode of operation is probably "framing", which simply takes several snapshots at predetermined times, but "streak" mode is also very useful. This uses a slit to image a line on the camera and then sweeps the line, giving you 1-d spatial information, but time-resolved. You can also combine the cameras with other diagnostics, to give you time-resolved spectrometry or interferometry.

chrismb
Posts: 3161
Joined: Sat Dec 13, 2008 6:00 pm

Post by chrismb »

Jccarlton wrote:The wiffleball is run for short periods to prevent the power supplies and magnets from burning themselves in cinders.
You've both missed my point because the thread has wandered. The question here that I have posed/answered is how to make progress with *an* experiment when the magnets cost so much. My suggestion is to run with permanent magnets because they can't burn up any power suppiles, nor short circuit, nor...&c., &c.. Now the downside of that is a) can't reach the field levels supposedly required for MWs, b) has cusps that terminate at the magnet, c) cannot run continuously, in part because of b).

Sure, I know these things. But if you wanted to run a cost-affordable experiment which can perform in a pulsed operation, then why not permanent magnets? You *should* get to know/prove if a wiffleball will form, if it ever could.
93143 wrote: b) Even with sub-millisecond pulses, the wiffleball (not waffleball) should essentially reach steady-state. Bussard stated this a long time ago. The physics are very fast.
I just flat-out reject this, and await experimental evidence otherwise. To me this seems to be the same basic type of misunderstanding that Bussard had over the Farnsworth fusor when he thought the ions in a fusor collided at the centre.

If you ever watch a fusor [without ion gun injection], it takes several seconds to 'start up'. The glow cranks up, the ions run around, and, nice-and-gently, up lights a central plasmoid. It is clear that these devices have multi-seconds physics mechanisms going on.

As far as I can see, the argument that it reaches steady state in ms is based on the time of flight of an ion back and forth. But it is such a hopelessly simplistic understanding of how space charge and ion flows get established within a large volume that I am quite sure no such 'steady state' is realisable in ms. In any case, I would beg to suggest that the notion of 'steady state' on a ms timescale is somewhat of an oxymoron. It's abit like claiming tokamaks can already reach steady state operations [at some stage during their pulse]!

KitemanSA
Posts: 6179
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Post by KitemanSA »

Chris,
Might you be assuming that some developing loss mechaniosm is part of the operating condition? I recall a statement that said that the FUSION activity is not visible and that any glows had to do with neutrals charging and discharging (I may have gotten that last part wrong). None-the-less, the implication is that any glow has nothing to do with the physics of a Polywell operating per-se; only with loss mechnanisms that would have to be taken care of in a full scale well.

If this is the case, then the wiffle ball MAY in fact reach steady-state way before the loss mechanisms build up. I.E steady state for one physics does not imply steady state for all.

Just a thought.

chrismb
Posts: 3161
Joined: Sat Dec 13, 2008 6:00 pm

Post by chrismb »

KitemanSA wrote:Chris,
Might you be assuming that some developing loss mechaniosm is part of the operating condition? I recall a statement that said that the FUSION activity is not visible and that any glows had to do with neutrals charging and discharging (I may have gotten that last part wrong). None-the-less, the implication is that any glow has nothing to do with the physics of a Polywell operating per-se; only with loss mechnanisms that would have to be taken care of in a full scale well.

If this is the case, then the wiffle ball MAY in fact reach steady-state way before the loss mechanisms build up. I.E steady state for one physics does not imply steady state for all.

Just a thought.
Yes. You're right. I am, indeed, commenting on the observations of lossy mechanisms. So in consideration, yes, you may be right that the wiffleball is up and running. Though - if there is a build-up of other 'stuff' going on in the process that chokes it, then the wiffleball may be 'steady state', but the device itself isn't. It can only be understood by experiment - and I am waiting for that (but not holding my breath!).

Aero
Posts: 1200
Joined: Mon Jun 30, 2008 4:36 am
Location: 92111

Post by Aero »

If you ever watch a fusor [without ion gun injection], it takes several seconds to 'start up'. The glow cranks up, the ions run around, and, nice-and-gently, up lights a central plasmoid. It is clear that these devices have multi-seconds physics mechanisms going on.
A fusor and a Polywell are much different machines. I've never watched a fusor, but I have watched a light bulb which I expect warms up much more quickly than a fusor. It was in a Failure Analysis course in college, a slow motion replay of an incandescent light bulb filament (coil wound filament) as the bulb was switched on. At first a cold coil, nothing, then the electricity hit the filament which jerked to the left, then the right, then the left, etc. until after four or five cycles, the filament started to glow. After about 8 to 12 cycles the picture was washed out by the brightly glowing filament. The point of the film was to illustrate the mechanical stresses on the filament, but it also showed that (at 60 Hz AC), it takes about 200 ms for the light bulb to produce significant light. That is already much longer than the required time claimed to achieve steady state operation for a Polywell fusor.
As far as I can see, the argument that it reaches steady state in ms is based on the time of flight of an ion back and forth. But it is such a hopelessly simplistic understanding of how space charge and ion flows get established within a large volume that I am quite sure no such 'steady state' is realisable in ms.
So, is steady state in 20 ms reasonable? We can do some simple reality checks. IIRC, it is claimed that an average ion must transition the Polywell about 10,000 times before it fuses. This is steady state operation. Though it is not to chrismb's point of start up transition to steady state taking seconds of time, it's a simple calculation to show what the ion velocity must be in order to transition a 30 cm Polywell 10,000 times in 20 ms. I get an average velocity of about 15,000 meters per second for WB-6, five percent of the speed of light, quite reasonable (really?). But for WB-100 it comes out to half the speed of light which is not so believable. If one were to include decelerating, stopping and reversing directions, and accelerating at the end of each transit, the calculation would result in higher velocities. As I recall, it is claimed that the ions spend most of their time outside the Wiffle ball, annealing as they accomplish this "turn-around" maneuver. The above numbers are average velocity. What must their real transit (peak) velocity be?

The above is inconclusive so we should do another reality check. What was the well depth for WB-6 and hence the ion peak velocity? The ions start with near zero velocity, 15 cm from the center of the Wiffle ball, accelerate like mad until they reach the electron sheath of the Wiffle ball, cruise through at steady velocity, then decelerate like mad, stopping about 15 cm from the center only to reverse directions and repeat. What is the diameter of the Wiffle ball? I would guess about 5 cm, from the picture on the EMC2 web site. That gives the ions about 5 cm to accelerate and decelerate. But now I'm in over my head because I don't recall the well depth of the Wiffle ball, the ion mass, or the force exerted on the ion mass by the electrostatic Wiffle ball and probably couldn't do the math even if I did.

I hope someone will respond to this argument, because the assumption that a Polywell reactor achieves steady state operation (WB-6) within 20 ms is looking a little shaky to me.
Aero

chrismb
Posts: 3161
Joined: Sat Dec 13, 2008 6:00 pm

Post by chrismb »

Aero wrote:IIRC, it is claimed that an average ion must transition the Polywell about 10,000 times before it fuses.
200,000, according to my [as yet unchallenged] calculation of viewtopic.php?p=13258&#13258 , using the stated numbers in the paper detailing a full-sized reactor. But that's for a 'full-speed' fuel proton (I guess) which'd be 10Mm/s, so that's an *average* flight time of 40ms to fusion.

For these lower voltage experiments, throw a few numbers at me and we can work it out; pick a drive voltage, a core particle density, reaction core size, wiffleball size and fuel type, and we can run the numbers together....

Post Reply