Need for superconducting magnets?

Discuss how polywell fusion works; share theoretical questions and answers.

Moderators: tonybarry, MSimon

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

Need for superconducting magnets?

Post by JohnP »

Pardon the naive question... In Bussdard's video he comments that the magnets used were fairly weak, and the goal was to contain the electron cloud, not the plasma. So why the talk of superconducting magnets for WB8 etc? Will they be needed if electrons are at a higher energy?

Zixinus
Posts: 200
Joined: Thu Jun 28, 2007 4:39 pm

Post by Zixinus »

From my understanding, superconducters (that because they have electricity flowing trough them, are electromagnets) are ultimately unavoidable in the long run, however for the first demo plant and experiments, liquid nitrogen cooled copper wires would be enough.

bcglorf
Posts: 436
Joined: Mon Jul 23, 2007 2:58 pm

not for B strength

Post by bcglorf »

From what I've read, superconducters were suggested to reduce the overall power usage of the magnets. Copper magnets can generate strong enough magnetic fields, but superconducting magnets, at a certain size, should use less power to produce the same field. Thus, it's about getting better efficiency as you scale up. I believe MSimon's run some numbers on the cost/benefit analysis on using superconducting magnets. They use less current, but you've also gotta keep them cool which may end up costing more than you save switching to them.
In a sentence, it's for efficiency, not necessity.

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

Post by JohnP »

superconducters were suggested to reduce the overall power usage of the magnets.

I think once you've run current into the superconductor you don't need to add more. As long as the temperature is low the steady-state current consumption should be zero. Yes, that would cut electricity consumption but now you're paying for liquid helium.

MSimon
Posts: 14331
Joined: Mon Jul 16, 2007 7:37 pm
Location: Rockford, Illinois
Contact:

Post by MSimon »

JohnP wrote:superconducters were suggested to reduce the overall power usage of the magnets.

I think once you've run current into the superconductor you don't need to add more. As long as the temperature is low the steady-state current consumption should be zero. Yes, that would cut electricity consumption but now you're paying for liquid helium.
An MRI uses about 4 liters of LHe every week or two. The LN2 requirements are about 10X that. Not really significant. The coils we would use would be of the same rough volume and current.

We will be using truck loads of LN2 at the end of our testing regime with WB7x. The reason for using LN2 cooled Cu coils is that you can do D-D fusion without giving too much thought to neutron bombardment of superconductors.

Our problem in an operating reactor is that we need to get rid of 10 to 20 MW of alpha impacts within inches of a magnet coil at 4degK. I think it can be done. It is what we call in engineering a "very interesting problem".

tonybarry
Posts: 219
Joined: Sun Jul 08, 2007 4:32 am
Location: Sydney, Australia
Contact:

Post by tonybarry »

Hello MSimon,
From your post I infer that you are in the process of constructing the next of Dr. Bussard's prototype polywells. May I ask is this your own private effort, or is it part of a larger group? If a larger group, is this academia based (i.e. through a university) or as part of defence or government funded research?
I am very pleased that Dr. Bussard's work is continuing.
Regards,
Tony Barry

MSimon
Posts: 14331
Joined: Mon Jul 16, 2007 7:37 pm
Location: Rockford, Illinois
Contact:

Post by MSimon »

tonybarry wrote:Hello MSimon,
From your post I infer that you are in the process of constructing the next of Dr. Bussard's prototype polywells. May I ask is this your own private effort, or is it part of a larger group? If a larger group, is this academia based (i.e. through a university) or as part of defence or government funded research?
I am very pleased that Dr. Bussard's work is continuing.
Regards,
Tony Barry
This is a public effort

http://iecfusiontech.blogspot.com/

Any one with an idea or criticism is free to join in or design your own.

At this point all that exists is a preliminary design. No funding yet.

It just seemed like some one should design a test reactor. So about a month or two ago I decided to give it a go.

I welcome criticism, advice, and anything else you care to throw my way.

esecallum
Posts: 13
Joined: Mon Aug 27, 2007 11:14 pm

Post by esecallum »

MSimon wrote:
tonybarry wrote:Hello MSimon,
From your post I infer that you are in the process of constructing the next of Dr. Bussard's prototype polywells. May I ask is this your own private effort, or is it part of a larger group? If a larger group, is this academia based (i.e. through a university) or as part of defence or government funded research?
I am very pleased that Dr. Bussard's work is continuing.
Regards,
Tony Barry
This is a public effort

http://iecfusiontech.blogspot.com/

Any one with an idea or criticism is free to join in or design your own.

At this point all that exists is a preliminary design. No funding yet.

It just seemed like some one should design a test reactor. So about a month or two ago I decided to give it a go.

I welcome criticism, advice, and anything else you care to throw my way.
Can you please,please explain why we never use electrostatic feilds to confine charged particles and always stick with magnetic fields with their heat generation and large currents to maintain the feild...


electrostatic fields are much more efficient and easier to use and lightweight compared to huge copper wires and cooling pipes...

an electric field is simply parallel plate electrodes at some voltage and this field can be maintained for a long time with little energy expenditure.



any thoughts?

Tom Ligon
Posts: 1871
Joined: Wed Aug 22, 2007 1:23 am
Location: Northern Virginia
Contact:

Post by Tom Ligon »

esecallum wrote:
MSimon wrote:
tonybarry wrote:Hello MSimon,
I welcome criticism, advice, and anything else you care to throw my way.
Can you please,please explain why we never use electrostatic feilds to confine charged particles and always stick with magnetic fields with their heat generation and large currents to maintain the feild...


electrostatic fields are much more efficient and easier to use and lightweight compared to huge copper wires and cooling pipes...

an electric field is simply parallel plate electrodes at some voltage and this field can be maintained for a long time with little energy expenditure.



any thoughts?
escallum, you've hit the nail on the head regarding what IEC machines are basically attempting to do. What Dr. Bussard realized going in to this thing was that electrostatic forces are the only sensible way to handle the ions.

Has anyone here mentioned the Elmore, Tuck, and Watson machine? Back in the late 1950's, they mentioned the possibility of making a simple gridded sphere machine which was an electrostatic accelerator for electrons. Converge electrons on the center of a sphere, create ions inside the inner electron accelerating grid, and the ions would oscillate thru the center of the machine. They showed that, in theory, some fusion would be possible.

The only catch was that the electrons would have a short life before the electrons intercepted a grid wire, and any conceivable transparency would be insufficient to make a net power machine, by a huge margin.

Dr. Bussard's magrid machines (WB6, etc) are Elmore-Tuck-Watson machines in which the electron accelerating grid is magnetically insulated. This is not a bulk trapping concept like a tokamak, and the ions are not very much affected by the magnetic field. The main goal is just to make the grid very hard to hit by the electrons.

esecallum
Posts: 13
Joined: Mon Aug 27, 2007 11:14 pm

Post by esecallum »

Tom Ligon wrote:
esecallum wrote:
MSimon wrote: I welcome criticism, advice, and anything else you care to throw my way.
Can you please,please explain why we never use electrostatic feilds to confine charged particles and always stick with magnetic fields with their heat generation and large currents to maintain the feild...


electrostatic fields are much more efficient and easier to use and lightweight compared to huge copper wires and cooling pipes...

an electric field is simply parallel plate electrodes at some voltage and this field can be maintained for a long time with little energy expenditure.



any thoughts?
escallum, you've hit the nail on the head regarding what IEC machines are basically attempting to do. What Dr. Bussard realized going in to this thing was that electrostatic forces are the only sensible way to handle the ions.

Has anyone here mentioned the Elmore, Tuck, and Watson machine? Back in the late 1950's, they mentioned the possibility of making a simple gridded sphere machine which was an electrostatic accelerator for electrons. Converge electrons on the center of a sphere, create ions inside the inner electron accelerating grid, and the ions would oscillate thru the center of the machine. They showed that, in theory, some fusion would be possible.

The only catch was that the electrons would have a short life before the electrons intercepted a grid wire, and any conceivable transparency would be insufficient to make a net power machine, by a huge margin.

Dr. Bussard's magrid machines (WB6, etc) are Elmore-Tuck-Watson machines in which the electron accelerating grid is magnetically insulated. This is not a bulk trapping concept like a tokamak, and the ions are not very much affected by the magnetic field. The main goal is just to make the grid very hard to hit by the electrons.


thank you.

I understand,but what I am saying could the outer magnetic field coils in the WB6,WB7 and WB8 be replaced by electrostatic confinement plates instead?

This would make it lighter,cheaper and solve the cooling problen and i squared*r heating losses in the field coils.

These again would be convex fields but as very little energy is required to main an electric field the advantages should be clear.


any answers?

tonybarry
Posts: 219
Joined: Sun Jul 08, 2007 4:32 am
Location: Sydney, Australia
Contact:

Post by tonybarry »

Hello esecallum,
I am not a nuke engineer, I have a background in EE but mainly in control (digital, simple analog) rather than power. So please read with caution ...
As far as I understand it, an electrostatic grid may be viewed as "holes" and "conductors", and the ratio of hole to conductor determines how often a charged species will hit it. The numbers seem to indicate that using egrids will doom the reactor to never producing break-even power due to electron losses to the grid "conductor". One cannot make a grid so fine (holy?) that the electron loss is sufficiently low to offer break-even power.
Bussard's polywell uses magnetic fields. The advantages over egrids are that instead of outright repulsion, the magnetic field offers the incident electron a "right handed pass". The electron's trajectory is (presumably chaotic) but confined to the centre region. As the electron pressure increases (as more electrons are dumped into the centre region) the electrons push back against the magnetic field, causing it to contract (similar to the bow shock front of the solar wind against the earth's magnetosphere). This compression of the magnetic field tends to close off the cusps between each magnetic pole (the "wiffleball" effect) and when the right conditions are reached, the cusps become "watertight" (electron - tight?) and do not allow electrons the opportunity to escape the central region.
This is a Good Thing, to quote Joe Strout. When a "hot" proton is dumped into the centre region (by means I am still unclear on), it is a hard thing to control. It is hot (moving quickly) and is a relatively heavy species (1800 times the mass of an electron) so is hard to retain in the core zone.
The enormous electrostatic attraction of the contained electrons for the hot protons keeps them in the centre. And the electrons are kept in the centre by magnetic fields. The electrons do not crash into the magnetic coils because the coils are shielded by the field.
I hope this explanation is a) correct and b) enlightening. If I have missed any salient point I would be grateful if the knowledgeable people about could enlighten me.
Regards,
Tony Barry

esecallum
Posts: 13
Joined: Mon Aug 27, 2007 11:14 pm

Post by esecallum »

tonybarry wrote:Hello esecallum,
I am not a nuke engineer, I have a background in EE but mainly in control (digital, simple analog) rather than power. So please read with caution ...
As far as I understand it, an electrostatic grid may be viewed as "holes" and "conductors", and the ratio of hole to conductor determines how often a charged species will hit it. The numbers seem to indicate that using egrids will doom the reactor to never producing break-even power due to electron losses to the grid "conductor". One cannot make a grid so fine (holy?) that the electron loss is sufficiently low to offer break-even power.
Bussard's polywell uses magnetic fields. The advantages over egrids are that instead of outright repulsion, the magnetic field offers the incident electron a "right handed pass". The electron's trajectory is (presumably chaotic) but confined to the centre region. As the electron pressure increases (as more electrons are dumped into the centre region) the electrons push back against the magnetic field, causing it to contract (similar to the bow shock front of the solar wind against the earth's magnetosphere). This compression of the magnetic field tends to close off the cusps between each magnetic pole (the "wiffleball" effect) and when the right conditions are reached, the cusps become "watertight" (electron - tight?) and do not allow electrons the opportunity to escape the central region.
This is a Good Thing, to quote Joe Strout. When a "hot" proton is dumped into the centre region (by means I am still unclear on), it is a hard thing to control. It is hot (moving quickly) and is a relatively heavy species (1800 times the mass of an electron) so is hard to retain in the core zone.
The enormous electrostatic attraction of the contained electrons for the hot protons keeps them in the centre. And the electrons are kept in the centre by magnetic fields. The electrons do not crash into the magnetic coils because the coils are shielded by the field.
I hope this explanation is a) correct and b) enlightening. If I have missed any salient point I would be grateful if the knowledgeable people about could enlighten me.
Regards,
Tony Barry
yes,i understand what you are saying.

thanks.

but what i am saying is to have plate electrodes just outside the device and charge those plates to a -ve potential and thus they will repel the electrons inside the device from hitting the sides.

the separate electodes outside the device would be in a polyhedral shape approximating a sphere.


no grids inside the device involved.

think of a football and 6 sided patches making it up..

thus you would have an approximate sphere comprised of polyhedral plates and would basically confine the electrons to the centre.


increasing the -ve charge would determine the tightness of the confinement.


what do you say to this idea?

bcglorf
Posts: 436
Joined: Mon Jul 23, 2007 2:58 pm

electron losses are why

Post by bcglorf »

In many ways the polywell is a functional variation of your idea. Instead of using negatively charged plates it uses the positive charge on the magrid to attract the electrons. The magnetic field only works to get a higher density in the centre versus the outside and to prevent the electrons crashing into the positively charged grid. The benefit is that it's easier to inject your ions off the magrid. With negatively charged plates(if electron losses could be fixed) your gonna have to find a way to drop in the ions closer to the well than the plates.

Tom Ligon
Posts: 1871
Joined: Wed Aug 22, 2007 1:23 am
Location: Northern Virginia
Contact:

Post by Tom Ligon »

esecallum wrote:[
yes,i understand what you are saying.

thanks.

but what i am saying is to have plate electrodes just outside the device and charge those plates to a -ve potential and thus they will repel the electrons inside the device from hitting the sides.

the separate electodes outside the device would be in a polyhedral shape approximating a sphere.


no grids inside the device involved.

think of a football and 6 sided patches making it up..

thus you would have an approximate sphere comprised of polyhedral plates and would basically confine the electrons to the centre.


increasing the -ve charge would determine the tightness of the confinement.


what do you say to this idea?
If you want to force the electrons in by repelling them with a negative charge, consider this will be an irresistable target for the ions!

Dr. Bussard tried repeller plates on the corners of WB5, a closed-box machine of the type of HEPS and PXL1, as a means of preventing electron leaks from the corners. Didn't work. Almost anything you try to do along these lines that is good for electrons will be bad for ions, and vice-versa.

The one great assymetry between the two is that electrons are thousands of times less massive, thus can be manipulated relatively easily with a magnetic field.

There might be one out, and that is the use of careful timing of the charges applied to the accelerating elements. Some of the Hirsch-Farnsworth IEC guys have been working on that. George Miley's C-machine is an example, and I think POPS might be along these lines. M. Simon is excited about that last one, and the guys who came up with it appear to be associated with Dr. Bussard's latest effort.

rexxam62
Posts: 60
Joined: Sun Jul 29, 2007 10:13 pm

Post by rexxam62 »

I thought the Polywell IEC Fusion machine did not need magnetism or heat to drive the device??? I thought it was all done electrostatic.

Post Reply