The classic derivation of the Debye Length assumes that the potential energy of the particles in the applied field is much smaller in magnitude than the kinetic energy of the particles.
Is this always true in a polywell?
Debye Length
-
alexjrgreen
- Posts: 815
- Joined: Thu Nov 13, 2008 4:03 pm
- Location: UK
Debye Length
Ars artis est celare artem.
What a clueless response...93143 wrote:If the gradient of the potential energy is small enough that the characteristic length of a 'substantial' change in kinetic energy is much larger than the Debye length, I don't see how it would matter...
If a Polywell were operating in the regime Art Carlson envisions (Debye screening), the characteristic length for the potential well is the Debye length.
But wait - the potential well is about 80-90% of the plasma temperature...
...yeah, it breaks the definition.
This is actually why I think it won't work like Art says; the fact that you've got a bunch of electrons all blazing in one direction, slowing down, increasing their density greatly, and attracting ions from the high-density edge, which then go roaring past them, decreasing their density as they accelerate, and producing (or trying to produce) a net positive charge below the electrons... then what? I think it will form a series of concentric alternating multiple wells, like the layers of an onion - a "polywell", if you will - that averages out to what EMC2's graphs show. Overall the plasma would be quasi-neutral, but locally it would be anything but. The ions are fast where the electrons are slow, and vice versa; the particle density differences between these regions (due to residence time differences) will produce a large charge separation, which will then dictate the maximum layer thickness, since the induced potential difference can't be greater than the particle energy.
I have been attempting to come up with a simulation that demonstrates this, but it's harder than I thought, and I've been busy with other things, so progress has been slow...
[quote="93143"]...
But wait - the potential well is about 80-90% of the plasma temperature...
...yeah, it breaks the definition.
This is actually why I think it won't work like Art says; the fact that you've got a bunch of electrons all blazing in one direction, slowing down, increasing their density greatly, and attracting ions from the high-density edge, which then go roaring past them, decreasing their density as they accelerate, and producing (or trying to produce) a net positive charge below the electrons... then what? I think it will form a series of concentric alternating multiple wells, like the layers of an onion - a "polywell", if you will - that averages out to what EMC2's graphs show. Overall the plasma would be quasi-neutral, but locally it would be anything but. The ions are fast where the electrons are slow, and vice versa; the particle density differences between these regions (due to residence time differences) will produce a large charge separation, which will then dictate the maximum layer thickness, since the induced potential difference can't be greater than the particle energy.
A couple of questions:
I was under the impression that the potential well represented the plasma temperature, Other than a paragraph in one of Bussard's paperers I've not seen much describing what causes the potential well depth shortfall compared to the injection energy of the electrons. Is this a 10-20% energy loss? Or is it transferred to the magrid somehow?
The densities certainly would increase/ decrease as the partical speeds decreased/ increased in a constant volume system- like moving from one end to the other end of a cyclinder. But in a sphere with at least some convergence this gradient in density would be moderated. And, if confluence/ central focus is great enough the ions will be densest in the central region where thay are fastest, or perhaps I should say densest and fastest in a shell near the center due to the central ion slowing due to a vertual anode formation. The electrons would also be draged inward some by the vertual anode, so their density would also increase near the center to a degree greater than you would expect based only on their momentum. These interplays between the ions- ions, ion-electrons, and electrons- electrons would further complicate the picture. How this would effect occilations, or how purposely induced occilations would effect this is beyond me.
Dan Tibbets
But wait - the potential well is about 80-90% of the plasma temperature...
...yeah, it breaks the definition.
This is actually why I think it won't work like Art says; the fact that you've got a bunch of electrons all blazing in one direction, slowing down, increasing their density greatly, and attracting ions from the high-density edge, which then go roaring past them, decreasing their density as they accelerate, and producing (or trying to produce) a net positive charge below the electrons... then what? I think it will form a series of concentric alternating multiple wells, like the layers of an onion - a "polywell", if you will - that averages out to what EMC2's graphs show. Overall the plasma would be quasi-neutral, but locally it would be anything but. The ions are fast where the electrons are slow, and vice versa; the particle density differences between these regions (due to residence time differences) will produce a large charge separation, which will then dictate the maximum layer thickness, since the induced potential difference can't be greater than the particle energy.
A couple of questions:
I was under the impression that the potential well represented the plasma temperature, Other than a paragraph in one of Bussard's paperers I've not seen much describing what causes the potential well depth shortfall compared to the injection energy of the electrons. Is this a 10-20% energy loss? Or is it transferred to the magrid somehow?
The densities certainly would increase/ decrease as the partical speeds decreased/ increased in a constant volume system- like moving from one end to the other end of a cyclinder. But in a sphere with at least some convergence this gradient in density would be moderated. And, if confluence/ central focus is great enough the ions will be densest in the central region where thay are fastest, or perhaps I should say densest and fastest in a shell near the center due to the central ion slowing due to a vertual anode formation. The electrons would also be draged inward some by the vertual anode, so their density would also increase near the center to a degree greater than you would expect based only on their momentum. These interplays between the ions- ions, ion-electrons, and electrons- electrons would further complicate the picture. How this would effect occilations, or how purposely induced occilations would effect this is beyond me.
Dan Tibbets
To error is human... and I'm very human.
It depends on how you define the plasma temperature. The electrons come shooting in at a certain energy, which enables them to climb, at maximum, a certain potential difference. Due to the way the charge distribution behaves in the real system, with virtual anode effects and whatnot, they never quite have to do this; the well ends up at maybe 80% of the drive potential, and the electrons make it all the way through the core and back up the other side. So their maximum energy is higher than the well depth.D Tibbets wrote: I was under the impression that the potential well represented the plasma temperature, Other than a paragraph in one of Bussard's paperers I've not seen much describing what causes the potential well depth shortfall compared to the injection energy of the electrons. Is this a 10-20% energy loss? Or is it transferred to the magrid somehow?
The ions are accelerated by the potential well from a very low speed. Their maximum energy is essentially equal to the well depth.
...is that a question?The densities certainly would increase/ decrease as the partical speeds decreased/ increased in a constant volume system- like moving from one end to the other end of a cyclinder. But in a sphere with at least some convergence this gradient in density would be moderated. And, if confluence/ central focus is great enough the ions will be densest in the central region where thay are fastest, or perhaps I should say densest and fastest in a shell near the center due to the central ion slowing due to a vertual anode formation. The electrons would also be draged inward some by the vertual anode, so their density would also increase near the center to a degree greater than you would expect based only on their momentum. These interplays between the ions- ions, ion-electrons, and electrons- electrons would further complicate the picture. How this would effect occilations, or how purposely induced occilations would effect this is beyond me.
I'm thinking superposition, rather than "moderation". But I won't know at least until I get a sim done, and probably not even then...
Debye Screening Length
Hello,
Just want to point out. Thomas Ligon said:
"The debye screening length is about the order of the dense plasma core. This means that every particle and electron can see every other particle and electron."
Everything can see everything else. That means, that if you were going to make a model of WB-6. You would need to model factorial[the number of things in the polywell] interactions for every timestep.
Now there are lots of ways to simplify that mess. For example, Bussard ran a 1.5 dimensional code to model the core.
Just want to point out. Thomas Ligon said:
"The debye screening length is about the order of the dense plasma core. This means that every particle and electron can see every other particle and electron."
Everything can see everything else. That means, that if you were going to make a model of WB-6. You would need to model factorial[the number of things in the polywell] interactions for every timestep.
Now there are lots of ways to simplify that mess. For example, Bussard ran a 1.5 dimensional code to model the core.
Have you contacted Dr. Mike? Send me a PM if you need his e-mail.93143 wrote:I have been attempting to come up with a simulation that demonstrates this, but it's harder than I thought, and I've been busy with other things, so progress has been slow...93143 wrote:If the gradient of the potential energy is small enough that the characteristic length of a 'substantial' change in kinetic energy is much larger than the Debye length, I don't see how it would matter...
Engineering is the art of making what you want from what you can get at a profit.
TallDave:
Maybe we should change the topic title from 'Theory' to 'Speculations'?
Or start up another topic called 'Theoretical Speculations'.
You wanna put numbers into equations on this bold statement or is it all just 'gut feel'?I wasn't sure layering was plausible outside of cosmic plasmas, but of the conceptual pictures I've seen posted here that does seem most consistent with physics and known or inferred results.
Maybe we should change the topic title from 'Theory' to 'Speculations'?
Or start up another topic called 'Theoretical Speculations'.
-
kunkmiester
- Posts: 892
- Joined: Thu Mar 12, 2009 3:51 pm
- Contact:
Icarus,
Heh, well that was seven months ago.
Anyways, equations probably aren't good enough to model overall behavior of the system, as Luis Chacon's and Joel Rogers' simulations show in their contradictions of Rider and Carlson's conclusions (remember Rick's comment about square vs parabolic wells, and sensitivity to assumptions?). Given my limited resources, I'll settle for "most consistent with physics and known or inferred results" of the models I've seen proposed. Not as pretty as a full simulation of what every particle is doing at every usec in a perfect mockup of WB-D, but useful at least in the inductive reasoning sense.
Maybe if we're lucky 93143 will give us something more elaborate.
You might check out Joel's results, he looked at "Debye slices" in his simulation.
Matt,
I think Tom might have been referring only to smaller devices. Large devices probably have some Debye screening.
Heh, well that was seven months ago.
Anyways, equations probably aren't good enough to model overall behavior of the system, as Luis Chacon's and Joel Rogers' simulations show in their contradictions of Rider and Carlson's conclusions (remember Rick's comment about square vs parabolic wells, and sensitivity to assumptions?). Given my limited resources, I'll settle for "most consistent with physics and known or inferred results" of the models I've seen proposed. Not as pretty as a full simulation of what every particle is doing at every usec in a perfect mockup of WB-D, but useful at least in the inductive reasoning sense.
Maybe if we're lucky 93143 will give us something more elaborate.
You might check out Joel's results, he looked at "Debye slices" in his simulation.
Matt,
I think Tom might have been referring only to smaller devices. Large devices probably have some Debye screening.
n*kBolt*Te = B**2/(2*mu0) and B^.25 loss scaling? Or not so much? Hopefully we'll know soon...
TallDave
Physics without equations is pretty limited, but engineering is impossible.
If this thing works, a set of simple first order equations describing overall behavior of it will be arrived at in short order by the huge mathematical/physics community. As it stands, there are a few guys tinkering around in isolation. It will die on the vine without a broad spectrum of theoretical discussion, an essentially free input also. An incomprehensibly short-sighted, insular bunker-mentality holds sway.
You are plain wrong. Do you truly believe there are no equations to model this system, as in, there is some new phenomena going on that has no basis in current physics?Anyways, equations probably aren't good enough to model overall behavior of the system,
Physics without equations is pretty limited, but engineering is impossible.
If this thing works, a set of simple first order equations describing overall behavior of it will be arrived at in short order by the huge mathematical/physics community. As it stands, there are a few guys tinkering around in isolation. It will die on the vine without a broad spectrum of theoretical discussion, an essentially free input also. An incomprehensibly short-sighted, insular bunker-mentality holds sway.
Engineering can operate well enough with rules of thumb. i.e 0.5 order equations. And except for the loss equations we have the 0.5 order equations:icarus wrote:TallDave
You are plain wrong. Do you truly believe there are no equations to model this system, as in, there is some new phenomena going on that has no basis in current physics?Anyways, equations probably aren't good enough to model overall behavior of the system,
Physics without equations is pretty limited, but engineering is impossible.
If this thing works, a set of simple first order equations describing overall behavior of it will be arrived at in short order by the huge mathematical/physics community. As it stands, there are a few guys tinkering around in isolation. It will die on the vine without a broad spectrum of theoretical discussion, an essentially free input also. An incomprehensibly short-sighted, insular bunker-mentality holds sway.
P = k R^3
P = K B^4
For SC magnets P = 1/R (derived from above and constant field in the SC wires)
Engineering is the art of making what you want from what you can get at a profit.
minor detail those loss equations lol .... hey why not just go ahead and build a plethora of them if its as easy as you are suggesting?And except for the loss equations
When there's huge holes in your theory and no evident attempt to plug them, or refute challenges, it smells like scam and failure ... just saying, no accusations.