Virtual Polywell
I think I got electron_fluid.c to work finally. I had to reduce the radial limits to 25 steps to get reasonable time - eventually I have to loop 1000's of times over this integral to get 1 millisecond of simulation. Waiting a day for that might make sense, but waiting a week doesn't.
I need to change the way I plot the data for checking, so I think that will be the next step. I also have to add the correct dimensionless parameters so the numbers scale right. But the core of the integral is now behaving correctly, so it looks like I can figure out initial electron density. Then the fun begins and I can see if it explodes like I expect it should!
I need to change the way I plot the data for checking, so I think that will be the next step. I also have to add the correct dimensionless parameters so the numbers scale right. But the core of the integral is now behaving correctly, so it looks like I can figure out initial electron density. Then the fun begins and I can see if it explodes like I expect it should!
Howdy Simon, great work with the LC POPS ! 5 points for thinking about it, 95 for getting it down on paper and out to us 
Systems that work with the intrinsic characteristics of the polywell are going to be the difference between it making and breaking power. We must be on the same frequency(whatever the hell it may be) I was brainstorming a few weeks back and i swear the thought of LC system entered my mind. I kinda canned the idea after the reading some of the pops literature.
First Problem: Density
So above we have the current pops equations for frequency. In the top equation the R^2well Mion product may be talking about density, and in the bottom equation density is mentioned precisely. I thought this was the "dealbreaker" because accurately quantizing plasma density, especially in dynamic polywell systems could be a big challenge. As i remember Langmuir probes certainly weren't up to the task and they calculated WB6's electron plasma density by back calculating using the B=1 condition
So this is all well and good for electron plasma, but this is without ions in the system. Bummer.
Second problem: Frequency
Depending on the stability of drive voltage, well radius and fuel injection/reaction products/vacuum system changing the density, this could vary the pops frequency quiet significantly. This is a very bad thing when you are talking about a harmonic oscillator. An LC circuit's frequency can be tuned within limits. Is pops going to fluctuate beyond these limits ?
This will all translate to a minimum performance requirement for control systems hoping to operate in pops enabled polywells. Simon your the man, are they up to the task ?
Im sure your LC system would sit along side my Ring Cavity Ionization Scheme as definite possibility in enhancing future polywells once we have some hard data and can narrow the numbers down. At the moment as you said a power amplifier driven system with octave band output filters looks the go.
Btw i felt with the wiki page on klystons was lacking. I have one tiny 2.5 second animation that explains more than that whole page. Dammit my links broken !
Systems that work with the intrinsic characteristics of the polywell are going to be the difference between it making and breaking power. We must be on the same frequency(whatever the hell it may be) I was brainstorming a few weeks back and i swear the thought of LC system entered my mind. I kinda canned the idea after the reading some of the pops literature.
First Problem: Density
So above we have the current pops equations for frequency. In the top equation the R^2well Mion product may be talking about density, and in the bottom equation density is mentioned precisely. I thought this was the "dealbreaker" because accurately quantizing plasma density, especially in dynamic polywell systems could be a big challenge. As i remember Langmuir probes certainly weren't up to the task and they calculated WB6's electron plasma density by back calculating using the B=1 condition
WB6 Lab Notes wrote:Beta = (8pi)(ne)(Ei)/B^2. If run at beta=one,
with known drive voltage/energy Ei and known B field strength, the density will be uniquely
determined. The beta=one condition can be measured by PMT data, which will always peak when
the electron density reaches its maximum, which occurs only when this is achieved
So this is all well and good for electron plasma, but this is without ions in the system. Bummer.
Second problem: Frequency
Depending on the stability of drive voltage, well radius and fuel injection/reaction products/vacuum system changing the density, this could vary the pops frequency quiet significantly. This is a very bad thing when you are talking about a harmonic oscillator. An LC circuit's frequency can be tuned within limits. Is pops going to fluctuate beyond these limits ?
This will all translate to a minimum performance requirement for control systems hoping to operate in pops enabled polywells. Simon your the man, are they up to the task ?
Im sure your LC system would sit along side my Ring Cavity Ionization Scheme as definite possibility in enhancing future polywells once we have some hard data and can narrow the numbers down. At the moment as you said a power amplifier driven system with octave band output filters looks the go.
Btw i felt with the wiki page on klystons was lacking. I have one tiny 2.5 second animation that explains more than that whole page. Dammit my links broken !
Last edited by Keegan on Tue Feb 05, 2008 12:35 pm, edited 1 time in total.
Purity is Power
Hey Dr Mike thats awesome, how do you make your plasma ? Grapes ? Lighting Match Sticks ?drmike wrote: My kids loved the plasma balls in the microwave.
Quarter Wave Graphite Microwave Resonatorsare the f@(%ing shit ......
*hints* This might be a nice way to get boron into the reactor core.
-k
Purity is Power
Candles and match sticks! I prefer the older microwaves with fans, the balls follow the fans around. But the new ones are still fun too!Keegan wrote: Hey Dr Mike thats awesome, how do you make your plasma ? Grapes ? Lighting Match Sticks ?
I don't protect the microwave either, I let the balls go where they like.
It is an honor to be of service.Keegan wrote:Howdy Simon, great work with the LC POPS ! 5 points for thinking about it, 95 for getting it down on paper and out to us :)
Leaving out the constant I like (sqrt(Vwell/Mioin))/Rwell as more explanatory. The sqrt term is due to the way velocity changes with energy and mass. Divided by the radius of the reaction volume. i.e. frequency goes down directly with size. Or in other words the longer a particle has to travel.First Problem: Density
So above we have the current pops equations for frequency. In the top equation the R^2well Mion product may be talking about density, and in the bottom equation density is mentioned precisely. I thought this was the "dealbreaker" because accurately quantizing plasma density, especially in dynamic polywell systems could be a big challenge. As i remember Langmuir probes certainly weren't up to the task and they calculated WB6's electron plasma density by back calculating using the B=1 condition
No density involved, except in figuring out the Debye length - which is a small factor.
WB6 Lab Notes wrote:Beta = (8pi)(ne)(Ei)/B^2. If run at beta=one,
with known drive voltage/energy Ei and known B field strength, the density will be uniquely determined. The beta=one condition can be measured by PMT data, which will always peak when
the electron density reaches its maximum, which occurs only when this is achieved
So this is all well and good for electron plasma, but this is without ions in the system. Bummer.
We can ignore this to a certain extent except as it determines optimum drive voltage.
Second problem: Frequency
Depending on the stability of drive voltage, well radius and fuel injection/reaction products/vacuum system changing the density, this could vary the pops frequency quiet significantly. This is a very bad thing when you are talking about a harmonic oscillator. An LC circuit's frequency can be tuned within limits. Is pops going to fluctuate beyond these limits ?
This will all translate to a minimum performance requirement for control systems hoping to operate in pops enabled polywells. Simon your the man, are they up to the task ?
Yes. They are up to the task. One thing that can be done is lower the Q for broader bandwidth. That will lower the gain, it may not matter.
Over at Fusor.net they have measured the frequency in the small fusors they typically make and came up with something in the 4 MHz range. Roughly what I would expect with D-D.
Im sure your LC system would sit along side my Ring Cavity Ionization Scheme as definite possibility in enhancing future polywells once we have some hard data and can narrow the numbers down. At the moment as you said a power amplifier driven system with octave band output filters looks the go.
I have sketched up some schematics (spent all night thinking about it). I'm a little slow with my dwg programs so I haven't turned it into a proper diagram but I will try to get something out later today.
Engineering is the art of making what you want from what you can get at a profit.
I downloaded electron_fluid.c and it runs fine here. I'd like to try visualizing the data as well.drmike wrote:I think I got electron_fluid.c to work finally.
I have a super crude display code which lets me look at the output via gnuplot. It sucks because there's no way to tell how radius affects the potential.
My todo list is
1) Compute normalization including correct potentials - the first gsl integral was supposed to do that but it won't be right once I start iterating.
2) Add above normalization and correct physics coefficients so I get true dimensionless potential solutions
3) Set up iterations and make sure I get self consistent solution.
The first integral can also be used to compute the electron density distribution, so I'll set that up as a subroutine so it can be used in several ways. Ways to visualize the particle density and potential will be very useful, especially when I get more advanced with some time step codes. But that's biting off a lot - small baby steps and steady progress makes each step feel like I'm getting somewhere.
But mostly it's fun, and the physics is cool. It'll be great to see if we can add a PLL that can work with a Tesla coil!! I never thought of combining a DSP with a Tesla coil before, but any excuse for a good time is a good one!!
My todo list is
1) Compute normalization including correct potentials - the first gsl integral was supposed to do that but it won't be right once I start iterating.
2) Add above normalization and correct physics coefficients so I get true dimensionless potential solutions
3) Set up iterations and make sure I get self consistent solution.
The first integral can also be used to compute the electron density distribution, so I'll set that up as a subroutine so it can be used in several ways. Ways to visualize the particle density and potential will be very useful, especially when I get more advanced with some time step codes. But that's biting off a lot - small baby steps and steady progress makes each step feel like I'm getting somewhere.
But mostly it's fun, and the physics is cool. It'll be great to see if we can add a PLL that can work with a Tesla coil!! I never thought of combining a DSP with a Tesla coil before, but any excuse for a good time is a good one!!
We may have to use an arbitrary waveform generator to get the right voltage profile for enhanced compression.
Even more fun.
Phase lock it to the natural reactor frequency and it gets even better.
Still. I would be most happy if all that was needed was an LC tuned circuit.
Even more fun.
Phase lock it to the natural reactor frequency and it gets even better.
Still. I would be most happy if all that was needed was an LC tuned circuit.
Engineering is the art of making what you want from what you can get at a profit.
I guess the whole point of my original rant on this page was to simply show that the pops frequency is tied to reactor conditions that are variable.
This to me is an extremely bad thing.
To make superhetrodyne recievers and tesla coils you need a stable frequency source. I cant see how pops is any different. A tiny drift in frequency may sound insignificant. But one must realise errors add up, compound and destroy time. The driving source could end up completely out of phase with the oscillator.
Varying reactor conditions could change the pops frequency limiting Q. This could hurt LC POPS systems. All the marvells of digital synthesizers and DSP feeding high power RF power amps keeping one step ahead of POPS dynamics could be the saving grace in getting a real POPS system to achieve high Q.
POPS appears to me to be 2 things.
1. A driven oscillator, periodically compressing the ion cloud to greatly enhance fusion Q
2. Synchronization. A way to get (parts of) the ion cloud and electron cloud at different regions of the core at different times.
How could this be useful ? Think of a four stroke engine cycle. Intake, Compression, Ignition, Exhaust.
2 possible clouds(ion, electron) x 2 possible states (compressed, expanded) or (left,right) = 4 strokes and we have an engine.
Once all that has soaked in then consider Spin. If the whole core was spinning at a fixed rpm and we could somehow using POPS modulate the core on and off, shit, we might actually be able direct the bulk of helium alphas to only special HV Direct Conversion Radians of the reactor, while dodging the coils. I see this as an excellent opportunity to solve sputtering/contamination issues and HV Direct Conversion E field Problems.
This to me is an extremely bad thing.
POPS is naturally a harmonic Oscillator. We want to exploit pops by driving it to increase fusion Q. That makes it a driven oscillator. It will "ring up" over time. Many of you tesla coilers should be familiar how the primary and secondary ring up and down over timePARK.PDF wrote: This concept
requires uniform-electron injection into
the central region of a spherical device to
produce harmonic oscillator potential. An
ion cloud (referred to as the Periodically
Oscillating Plasma Sphere, or POPS)
in such an environment will undergo
harmonic oscillation with an oscillation
frequency independent of amplitude.
To make superhetrodyne recievers and tesla coils you need a stable frequency source. I cant see how pops is any different. A tiny drift in frequency may sound insignificant. But one must realise errors add up, compound and destroy time. The driving source could end up completely out of phase with the oscillator.
Varying reactor conditions could change the pops frequency limiting Q. This could hurt LC POPS systems. All the marvells of digital synthesizers and DSP feeding high power RF power amps keeping one step ahead of POPS dynamics could be the saving grace in getting a real POPS system to achieve high Q.
POPS appears to me to be 2 things.
1. A driven oscillator, periodically compressing the ion cloud to greatly enhance fusion Q
2. Synchronization. A way to get (parts of) the ion cloud and electron cloud at different regions of the core at different times.
How could this be useful ? Think of a four stroke engine cycle. Intake, Compression, Ignition, Exhaust.
2 possible clouds(ion, electron) x 2 possible states (compressed, expanded) or (left,right) = 4 strokes and we have an engine.
Once all that has soaked in then consider Spin. If the whole core was spinning at a fixed rpm and we could somehow using POPS modulate the core on and off, shit, we might actually be able direct the bulk of helium alphas to only special HV Direct Conversion Radians of the reactor, while dodging the coils. I see this as an excellent opportunity to solve sputtering/contamination issues and HV Direct Conversion E field Problems.
Purity is Power
Well, I'm a controls engineer. To me it is an extremely good thing. ;-)Keegan wrote:I guess the whole point of my original rant on this page was to simply show that the pops frequency is tied to reactor conditions that are variable.
This to me is an extremely bad thing.
Seriously. I don't see any unsolvable problems. There are any number of ways to solve the problem.
1. Monitor the SWR on the HV line for POPS frequencies.
2. Monitor ion frequencies in the reactor.
3. Monitor neutron output (if we ever get enough)
4. Modulate the HV at a pilot frequency (far from POPS) and use the phase of the return to control the HV supply voltage. (which is a very neat trick - if I have to use it I will explain it)
Keeping the HV sufficiently stable. I think the HV DC supplies can be kept within .1% without excessive difficulties.
=====
What is difficult is that the requirements are getting out of range of the typical amateur. Unfortunate. Not a show stopper.
Engineering is the art of making what you want from what you can get at a profit.
Ain't no amateurs on this bus!
When the goin' gets weird, the weird turn pro
I still want to look at TM waves. POPS is electrostatic. I think we can get directed beams and stable operation with RF antennas rather than oscillations on the grid. We don't even know what all the problems are yet, but if there's a way it can be done theoretically there's a way it can be done period!
When the goin' gets weird, the weird turn pro
I still want to look at TM waves. POPS is electrostatic. I think we can get directed beams and stable operation with RF antennas rather than oscillations on the grid. We don't even know what all the problems are yet, but if there's a way it can be done theoretically there's a way it can be done period!
I'd like to see you get a 6 MHz Yagi into one of these jobs. Or a 6 MHz waveguide.drmike wrote:Ain't no amateurs on this bus!
When the goin' gets weird, the weird turn pro 8)
I still want to look at TM waves. POPS is electrostatic. I think we can get directed beams and stable operation with RF antennas rather than oscillations on the grid.
Yep. However, let us try to avoid unnecessary complications.We don't even know what all the problems are yet, but if there's a way it can be done theoretically there's a way it can be done period!
“Physicists dream of Nobel prizes, engineers dream of mishaps.” Hendrik Tennekes
Engineering is the art of making what you want from what you can get at a profit.