Focusing the beam by using differential velocity of the ions
Focusing the beam by using differential velocity of the ions
If we put ion guns symmetrically around the outer shell, pointing at the same central spot. The starter beam could be focussed by profiling the velocity of the ions. As time progresses we increase the velocity. In this way the later emitted ions would be at a higher speed and could be arranged to hit the central spot at the same time as the earlier ones, hence concentrating the beam and increasing the confinement time.
Re: Focusing the beam by using differential velocity of the ions
Alternatively you could use a multi-MHz beam buncher to do the same thing without having to try to modulate the voltage right. Problem with these strategies is that it requires either a long drift space, or large energy differences in the ions for them to converge in the center if you want reasonable sized bunches. I spose if you could modulate the voltage right in a pulse it could be worth it...
Re: Focusing the beam by using differential velocity of the ions
Does anyone use the pulse technique in synchronism with the ion bunches?
You could get a double whammy. Pulsed power and (if you profiled the voltage pules) bunching of ions.
Shouldn't be too difficult with today's fast op amps.
(I'm guessing you can detect the ions as they pass through the grid)
You could get a double whammy. Pulsed power and (if you profiled the voltage pules) bunching of ions.
Shouldn't be too difficult with today's fast op amps.
(I'm guessing you can detect the ions as they pass through the grid)
-
Adam Szendrey
- Posts: 1333
- Joined: Fri Mar 29, 2002 5:36 pm
- Real name: Adam Szendrey
- Location: Budapest, Hungary
Re: Focusing the beam by using differential velocity of the ions
Op amps? For what purpose? Low power signal generation maybe?
What do you mean by "detecting ions as they pass through the grid"?
One by one (impossible) ? Or the stream?
Anyway the majority of losses comes from electron losses not ion losses (if i'm correct).
About your other post on sectioning the grid (i suppose in both cases you mean the inner cathode grid).
I don't quite see how would that help.
Focusing seems to be the best with a good symmetrical grid having the same voltage all over...if it's done right it will result in a bright spot in the center.
The inner grid is usually small..sectioning it and leading several wires to it sounds ultra painfull. I strongly doubt it would worth it.
Ofcourse i still might be wrong.
Adam
What do you mean by "detecting ions as they pass through the grid"?
One by one (impossible) ? Or the stream?
Anyway the majority of losses comes from electron losses not ion losses (if i'm correct).
About your other post on sectioning the grid (i suppose in both cases you mean the inner cathode grid).
I don't quite see how would that help.
Focusing seems to be the best with a good symmetrical grid having the same voltage all over...if it's done right it will result in a bright spot in the center.
The inner grid is usually small..sectioning it and leading several wires to it sounds ultra painfull. I strongly doubt it would worth it.
Ofcourse i still might be wrong.
Adam
Re: Focusing the beam by using differential velocity of the ions
To maximise efficiency you need to maximise output (focus the ions so they have more time to interact) and reduce losses (minimise electron emission).
With pulsed operation, I envisage bunches of ions oscillating between the centre of the inner grid and its outside. You can detect the bunch passing through the grid by an increase in voltage due to the passing charge. (hence the op amp). The Hv can then be turned on to reverse the velocity of the ions and also profile them so they all converge on the centre of the grid.
Makes sense or a mad idea?
Paul
With pulsed operation, I envisage bunches of ions oscillating between the centre of the inner grid and its outside. You can detect the bunch passing through the grid by an increase in voltage due to the passing charge. (hence the op amp). The Hv can then be turned on to reverse the velocity of the ions and also profile them so they all converge on the centre of the grid.
Makes sense or a mad idea?
Paul
-
Richard Hull
- Moderator
- Posts: 15023
- Joined: Fri Jun 15, 2001 9:44 am
- Real name: Richard Hull
Re: Focusing the beam by using differential velocity of the ions
Sounds like Farnsworth multipaction to me. Bunching of electrons and ions is old news 30's 40's 50's. Farnsworth failed with this idea and went to the concept of recirculation as in the fusor when operated at low pressures.
Richard Hull
Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Re: Focusing the beam by using differential velocity of the ions
Did they have the technology then to profile the velocity?
Any references to multipaction so I can investigate?
Any references to multipaction so I can investigate?
-
Richard Hull
- Moderator
- Posts: 15023
- Joined: Fri Jun 15, 2001 9:44 am
- Real name: Richard Hull
Re: Focusing the beam by using differential velocity of the ions
I am sure the technology was there in the world, but not the resources at the Pontiac Street location.
The Fort Wayne facility was an ITT backwater. The real money there was made by the Federal tube division. (special military stuff). Within this backwater, the fusion team was, itself, a separate and isolated backwater that was misunderstood both within the plant in Fort Wayne and by ITT Corporate management in New York.
The largest yearly budget to pay for all salaries (7) and all equipment, building space and R&D was $3 million in 1965. Most of this was gobbled up that year paying for several outshopped, contracted academic studies of the fusor process in grants to universities. This was to see if the eggheads thought the fusor process made sense. So, money was always very tight and a lot of key pieces of gear were lashed up "in house". The main computation engines were two large monroe-matic scientific mechanical calculators which cost thousands at that time.
Richard Hull
The Fort Wayne facility was an ITT backwater. The real money there was made by the Federal tube division. (special military stuff). Within this backwater, the fusion team was, itself, a separate and isolated backwater that was misunderstood both within the plant in Fort Wayne and by ITT Corporate management in New York.
The largest yearly budget to pay for all salaries (7) and all equipment, building space and R&D was $3 million in 1965. Most of this was gobbled up that year paying for several outshopped, contracted academic studies of the fusor process in grants to universities. This was to see if the eggheads thought the fusor process made sense. So, money was always very tight and a lot of key pieces of gear were lashed up "in house". The main computation engines were two large monroe-matic scientific mechanical calculators which cost thousands at that time.
Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Re: Focusing the beam by using differential velocity of the ions
RE: Focusing ions so they do not strike the Grid (or cage).
I am not an expert on ion optics, but have done quite a bit of electron optics work, lately, for an unrelated project. It is possible, in principle, to devise electrostatic lenses, that do not intercept any charged particles... or at so few as to be neglible.
Generally speaking, tubular lenses, such as those found in the three tube "Einzel" lens arrangements, can be operated so they do not have any impacts from the ions. Aperture lenses on the other hand, may intercept few or most of the ions being focussed. The grid cage structure is a special case, of a multi-aperture lens.
That having been said, I am not absolutely sure that the cage normally does much focussing of the ions in a spherical fusor. It must do some, but with large "open-ness" of the cage, the fields around the small diameter cage wires fall away as 1/r from the wires ( the r here is the radius of the cage wires, not the radius of the cage itself. In the far field, the cage and the fusor shell form a sort of spherical capacitor, which will have a voltage gradient that goes approximately as 1/R , also, but this R is the cage radius. Consequently, the fields around the cage wires are rather complicated.
If we follow a single ion inward from wherever it originates, we can see that general path will be approximately along a radius of the fusor, leading the ion past the gird cage wires, and on into the center of the fusor, and (if it does not collide with anything interesting like another deuteron) straight on out the other side.
But... as it passes through the cage, and depending how close it passes to a grid cage wire, the ion will be deflected toward the wire and if it does not bend into the wire and strike it, will continue on missing the center by a margin that gets larger, as the original ion path gets closer to the grid wires.
With the grid cage at some 10kV or more, if the ion originates far enough out to gain most of the 10 kV potential energy and convert it to kinetic energy, it can pass rather close to the grid wires without actually hitting them.
However, like the interplanetary space craft that swings past a massive planet to get a gravitational slingshot effect, these ions skimming on by, are strongly deflected and may completely miss the central region of the fusor. This is one argument for making the center cage small.
As we contemplate how these ions travel, it needs to be realized that they will be originating in many different places, at completely unsynchronized times, and thus being soooo small, have an almost perfect chance of never colliding with another accelerated ion. If the gas density is high enough, as Richard has pointed out in the thread on Efficiency, below, then the mean free path of an ion is short, at best only a few fusor diameters long. Thus what the ion WILL collide with, is (1) the fusor wall; (2) the grid cage; and (3) other neutral gas molecules. From simple geometric considerations, the fusor shell and the neutrals seem likely to receive the majority of the collisions. And as the internal pressure is increased, fewer ions will reach the wall and the neutral -ion collisions will dominate.
There needs to be a balance here of ion energy, and gas pressure. The majority of ions must absorb fusing energy in the form of velocity, (U ~ 20 - 30 KeV ), and the neutral deuterium gas density must be high enough so as to stop all ions by collision before they get too far out the other side of the grid cage, .(thus losing energy as they are decelerated by the field around the cage.)
I don't know where this all balances, or even if. At first glance, the requirements seem contradictory. High gas density could preclude a mean free path long enough to gain the necessary energy. But without sufficient gas density, the ion travels to the wall and deposits its remaining energy as heat, unless its trajectory is curved sufficiently to allow it to loop back for another pass.
As many have observed, the simplicity of design is deceiving.
Dave Cooper
I am not an expert on ion optics, but have done quite a bit of electron optics work, lately, for an unrelated project. It is possible, in principle, to devise electrostatic lenses, that do not intercept any charged particles... or at so few as to be neglible.
Generally speaking, tubular lenses, such as those found in the three tube "Einzel" lens arrangements, can be operated so they do not have any impacts from the ions. Aperture lenses on the other hand, may intercept few or most of the ions being focussed. The grid cage structure is a special case, of a multi-aperture lens.
That having been said, I am not absolutely sure that the cage normally does much focussing of the ions in a spherical fusor. It must do some, but with large "open-ness" of the cage, the fields around the small diameter cage wires fall away as 1/r from the wires ( the r here is the radius of the cage wires, not the radius of the cage itself. In the far field, the cage and the fusor shell form a sort of spherical capacitor, which will have a voltage gradient that goes approximately as 1/R , also, but this R is the cage radius. Consequently, the fields around the cage wires are rather complicated.
If we follow a single ion inward from wherever it originates, we can see that general path will be approximately along a radius of the fusor, leading the ion past the gird cage wires, and on into the center of the fusor, and (if it does not collide with anything interesting like another deuteron) straight on out the other side.
But... as it passes through the cage, and depending how close it passes to a grid cage wire, the ion will be deflected toward the wire and if it does not bend into the wire and strike it, will continue on missing the center by a margin that gets larger, as the original ion path gets closer to the grid wires.
With the grid cage at some 10kV or more, if the ion originates far enough out to gain most of the 10 kV potential energy and convert it to kinetic energy, it can pass rather close to the grid wires without actually hitting them.
However, like the interplanetary space craft that swings past a massive planet to get a gravitational slingshot effect, these ions skimming on by, are strongly deflected and may completely miss the central region of the fusor. This is one argument for making the center cage small.
As we contemplate how these ions travel, it needs to be realized that they will be originating in many different places, at completely unsynchronized times, and thus being soooo small, have an almost perfect chance of never colliding with another accelerated ion. If the gas density is high enough, as Richard has pointed out in the thread on Efficiency, below, then the mean free path of an ion is short, at best only a few fusor diameters long. Thus what the ion WILL collide with, is (1) the fusor wall; (2) the grid cage; and (3) other neutral gas molecules. From simple geometric considerations, the fusor shell and the neutrals seem likely to receive the majority of the collisions. And as the internal pressure is increased, fewer ions will reach the wall and the neutral -ion collisions will dominate.
There needs to be a balance here of ion energy, and gas pressure. The majority of ions must absorb fusing energy in the form of velocity, (U ~ 20 - 30 KeV ), and the neutral deuterium gas density must be high enough so as to stop all ions by collision before they get too far out the other side of the grid cage, .(thus losing energy as they are decelerated by the field around the cage.)
I don't know where this all balances, or even if. At first glance, the requirements seem contradictory. High gas density could preclude a mean free path long enough to gain the necessary energy. But without sufficient gas density, the ion travels to the wall and deposits its remaining energy as heat, unless its trajectory is curved sufficiently to allow it to loop back for another pass.
As many have observed, the simplicity of design is deceiving.
Dave Cooper
Re: Focusing the beam by using differential velocity of the ions
Dave,
you response interests me greatly.
I read a patent by Buzzard, ref
http://www.orbit6.com/fusor/BussardPatent2.htm
which proposes changing the simple grid design, for a "quasi-hexagonal-conical " one to generate ion acoustic waves.
I think there may be a better way to do this by using a Linac type construction. Your tubular lenses sound very similar to what I had in mind, except mine are hexaganol.
Can you expand on how your electron lens prevents collision with the apperatus? (or give reference)
BTW here is another Patent that people may find interesting.
http://www.geocities.com/researchtriang ... us/8632/#C
you response interests me greatly.
I read a patent by Buzzard, ref
http://www.orbit6.com/fusor/BussardPatent2.htm
which proposes changing the simple grid design, for a "quasi-hexagonal-conical " one to generate ion acoustic waves.
I think there may be a better way to do this by using a Linac type construction. Your tubular lenses sound very similar to what I had in mind, except mine are hexaganol.
Can you expand on how your electron lens prevents collision with the apperatus? (or give reference)
BTW here is another Patent that people may find interesting.
http://www.geocities.com/researchtriang ... us/8632/#C
Re: Focusing the beam by using differential velocity of the ions
On tubular lenses... there are numerous references. Building Scientific Apparatus by Moore, Davis and Coplan... one of our vintage references, has a nice chapter on electron optics principles.
However I ran many hundreds of trajectories using Simion 7 with various electron sources..and tubular electrode arrangements to eventually conclude it is possible to design lenses to avoid any collisions between the ion and the electrodes.
For the aperture lenses it is also possible to get neglible numbers of colliding electrons (ions) . The voltage and spacings are different for tubular and aperture lonses.
So far, I have not examined the spherical cage assembly in detail to see all possible trajectories in 3D. However we can say that a ring like electrode will have more or less the same focussing behavior whether it is a hexagon or a circle.
But one needs to keep in mind... that unless the ions are being produced in exactly the same place, (on a sub atomic resolution) and accelerated precisely the same..( i.e.: travelling through the exact field, or more correctly the same spatial voltage gradient, they will not be synchronized with any other ion, and collisions will be a matter of chance.. and happen with very low probabiliites.
To my mind, the inability to synchronize the ions with subatomic resolution, is the problem with the RF acceleration mode, mentioned above and on the other threads here. Converting these tiny dimensions into a time/phase equivalent give results that are far beyond any electronic timing that I know about.
Not sure if this elaboration is helping... but it does suggest the parameters that need to be controlled.
Dave Cooper
Paul Riley wrote:
> Dave,
> you response interests me greatly.
> I read a patent by Buzzard, ref
> http://www.orbit6.com/fusor/BussardPatent2.htm
> which proposes changing the simple grid design, for a "quasi-hexagonal-conical " one to generate ion acoustic waves.
> I think there may be a better way to do this by using a Linac type construction. Your tubular lenses sound very similar to what I had in mind, except mine are hexaganol.
>
> Can you expand on how your electron lens prevents collision with the apperatus? (or give reference)
>
> BTW here is another Patent that people may find interesting.
>
> http://www.geocities.com/researchtriang ... us/8632/#C
However I ran many hundreds of trajectories using Simion 7 with various electron sources..and tubular electrode arrangements to eventually conclude it is possible to design lenses to avoid any collisions between the ion and the electrodes.
For the aperture lenses it is also possible to get neglible numbers of colliding electrons (ions) . The voltage and spacings are different for tubular and aperture lonses.
So far, I have not examined the spherical cage assembly in detail to see all possible trajectories in 3D. However we can say that a ring like electrode will have more or less the same focussing behavior whether it is a hexagon or a circle.
But one needs to keep in mind... that unless the ions are being produced in exactly the same place, (on a sub atomic resolution) and accelerated precisely the same..( i.e.: travelling through the exact field, or more correctly the same spatial voltage gradient, they will not be synchronized with any other ion, and collisions will be a matter of chance.. and happen with very low probabiliites.
To my mind, the inability to synchronize the ions with subatomic resolution, is the problem with the RF acceleration mode, mentioned above and on the other threads here. Converting these tiny dimensions into a time/phase equivalent give results that are far beyond any electronic timing that I know about.
Not sure if this elaboration is helping... but it does suggest the parameters that need to be controlled.
Dave Cooper
Paul Riley wrote:
> Dave,
> you response interests me greatly.
> I read a patent by Buzzard, ref
> http://www.orbit6.com/fusor/BussardPatent2.htm
> which proposes changing the simple grid design, for a "quasi-hexagonal-conical " one to generate ion acoustic waves.
> I think there may be a better way to do this by using a Linac type construction. Your tubular lenses sound very similar to what I had in mind, except mine are hexaganol.
>
> Can you expand on how your electron lens prevents collision with the apperatus? (or give reference)
>
> BTW here is another Patent that people may find interesting.
>
> http://www.geocities.com/researchtriang ... us/8632/#C
Calibration of the system
Lets assume we can build the unit to 0.25mm tolerance.
We can trim out the x,y tolerance by differential voltage on the last drift tube. (BTW I am assuming that we have a microprocessor controlling this lot).
If we assume 2kV for each section of the linac, I can design a circuit that can control the differential voltage to at least 2v accuracy, with a resolution of 0.2v (biggest problem will be noise) i.e. 0.1 %. Now the question. Is this sufficient to calibrate out the x,y tolerance. (I need help here).
As far as timing is concerned, (i.e z direction) We could expect to electronically control the phase to +-1 degree. Again is this sufficient to calibrate out the physical tolerance?
I see calibration as not too difficult (but time consuming) as we can maximise neutron count for a given input voltage for each pair of tubes.
To break even we need to be able to get the right density for the right time. Because the tubes are in a spherical arrangement, the centre will have a maxwellian distribution, so we can use normal fusion theory to determine the outcome.
Because the later ions travel faster, a good approximation to the confinement time is the time that the front edge of the bunch arrive at the centre compared with the time the back edge of the bunch arrives. (I think) This is a less onerous condition than just having two guns facing each other because for a short while the ions at the centre will bounce off each other.
As the arrangement is spherical we get a compression ratio (CR) (pressure at the outer edge/ pressure at the centre) that is proportional to r^3 (assuming the tubes themselves take up a small volume) Realistically we could build a device with a 1m case, and a 0.1mm (or less) target sphere. This is a CR of 10^12 !! so not only is the temperature high the effective pressure is high as well.
Would anyone like to try doing the sums with realistic pressures and see what yield we could get. (my maths on this is a bit rusty and I do not have your experience in what is achieveable or realistic in terms of initial effective ion pressure)
We can trim out the x,y tolerance by differential voltage on the last drift tube. (BTW I am assuming that we have a microprocessor controlling this lot).
If we assume 2kV for each section of the linac, I can design a circuit that can control the differential voltage to at least 2v accuracy, with a resolution of 0.2v (biggest problem will be noise) i.e. 0.1 %. Now the question. Is this sufficient to calibrate out the x,y tolerance. (I need help here).
As far as timing is concerned, (i.e z direction) We could expect to electronically control the phase to +-1 degree. Again is this sufficient to calibrate out the physical tolerance?
I see calibration as not too difficult (but time consuming) as we can maximise neutron count for a given input voltage for each pair of tubes.
To break even we need to be able to get the right density for the right time. Because the tubes are in a spherical arrangement, the centre will have a maxwellian distribution, so we can use normal fusion theory to determine the outcome.
Because the later ions travel faster, a good approximation to the confinement time is the time that the front edge of the bunch arrive at the centre compared with the time the back edge of the bunch arrives. (I think) This is a less onerous condition than just having two guns facing each other because for a short while the ions at the centre will bounce off each other.
As the arrangement is spherical we get a compression ratio (CR) (pressure at the outer edge/ pressure at the centre) that is proportional to r^3 (assuming the tubes themselves take up a small volume) Realistically we could build a device with a 1m case, and a 0.1mm (or less) target sphere. This is a CR of 10^12 !! so not only is the temperature high the effective pressure is high as well.
Would anyone like to try doing the sums with realistic pressures and see what yield we could get. (my maths on this is a bit rusty and I do not have your experience in what is achieveable or realistic in terms of initial effective ion pressure)