Fusortron?

Yours and mine. This is where you can gas on about how you think the universe works. To a point, after that we'll expect you to actually test your stuff and report.

Fusortron?

Postby Joe Jarski » Mon Nov 08, 2010 11:48 pm

I don't like to just throw out wild ideas, but this developed out of another post regarding Chris's fusor and it'd take more time and resources than I'll be able to muster anytime soon. Besides, I'm heading off in a different, more conventional direction with my first fusor anyway... Fusor2?? Maybe someone will find it useful in the mean time or maybe it's been explored already or it just plain isn't gonna work.

The basic idea is a combination of a cylindrical fusor, cyclotron and ion gun. Ions would be injected into the cyclotron and accelerated to a fixed velocity creating a ring of D ions orbiting around the central grid. Working in pulse mode the grid would switch to a negative potential at the same time the cyclotron AC is stopped. D ions head for the grid in some sort of spiral de-orbiting maneuver where some fuse and some scatter where they are recollected and accelerated by the cyclotron AC field. And then to steal Doug's words... wash, rinse, repeat.

This isn't well thought out yet - I may be shooting it down myself before too long!
User avatar
Joe Jarski
 
Posts: 231
Joined: Thu Sep 16, 2010 8:37 pm
Location: SouthEast Michigan

Re: Fusortron?

Postby chrismb » Tue Nov 09, 2010 2:19 am

This has been patented and, looking at the patent, actually tried out. Look up Rolf Stenbaka on USPTO.

I have to rush to work right now, will post it up later unless someone beats me to it.
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: Fusortron?

Postby Joe Jarski » Tue Nov 09, 2010 7:26 am

That's good to know - thanks Chris. The more I think about it there are a lot of problems, but I'm curious how/if they managed to get around them.
User avatar
Joe Jarski
 
Posts: 231
Joined: Thu Sep 16, 2010 8:37 pm
Location: SouthEast Michigan

Re: Fusortron?

Postby Doug Coulter » Tue Nov 09, 2010 9:06 am

Interesting, doesn't sound like a completely dumb idea, either. I'm assuming the H field is along the grid axis? This would have a couple of effects. One would be the prevention of electrons getting out to the tank walls -- a major loss source. I am wondering if it wouldn't be good to change the field coincident with the grid pulse here. If you could make it suddenly huge, you'd have a pinch type device, or if it went away right then the ions wouldn't see it as a force tending to drive them away from a straight in trajectory. Of course, they'd still have the circular motion from the cyclotron and spiral, but I don't see that as being a problem.

I've been meaning to try an axial field on mine, and may get to it soon, mainly for the electron trapping function. If I put a permanent magnet set in there, of course it wouldn't live too long, but it would show if it has the desired effect and could be designed later to prevent magnet heating. In that case, you'd use a fairly weak field, just enough so the electrons couldn't be driven straight out to the walls which is the main energy loss in my fusor now (which is, btw, doing real well, but...the usual few micro-watts of fusion for much more power input). A field strong enough to make the ions go circular, with, I assume the electrodes arranged like one of the cyclotron designs, hadn't occurred to me.

That's one reason I was ragging on RichardH to get a door for his new system -- things like this are worth looking at IMO, but if you make your system hard or expensive to get inside of, you'll never wring enough changes to see the possibilities. Something as "complex" as this is bound to show various classes of "emergent behavior" -- always clear in hindsight, but difficult to predict well enough up front.

I am also exploring the idea of cutting excess secondary electron emission from a grid by having the outside insulated so ions don't hit the outside of the vanes/wires. Secondary electrons generated there see the full field to the tank walls. You'd still have hits on the inside of the grid, but electrons released there wouldn't see nearly as much field from the applied field, and might linger because of the additional charge of the ions at focus -- kind of the same idea Bussard was trying to get to (but I think, failed). At any rate, trying a few of those things would be fun and informative, so I'm all for it, and have a couple on the list for this rebuild about to get underway.

No point not trying the easy stuff! No question we learn something no matter what, and fairly quick. And say we get lucky and get some improvement. A small improvement of a few X would still be of great value for the people using these as neutron sources. A complete failure would be instructive at worst, and who knows, maybe there's more than a few X in one of these, or it would show you something that would make the next step obvious (that last has happened here and is a frequent outcome of "just try it").
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Fusortron?

Postby chrismb » Tue Nov 09, 2010 2:41 pm

Best you take a look at the diagrams in the patent first;

stenbacka_pat4853173.pdf
(660.69 KiB) Downloaded 511 times


So to summarise this, you generate fast ions from a gun, fire them into a circular beam in which there is an outward electric field force and inward magnetic force on the ions.

The reason this one came to my attention is that I went looking for all such inventions with this field configuration, because this is the field config that mine uses. An ion performs cyclotron motion in a magnetic field but its inward radial acceleration [from the magnetic field] is opposed by an electric field force pointing outwards, thus enlarging the radius. This might seem a little counter intuitive that it is self-stabilising, until you look at the force-radius curve, but this configuration is actually the only way to keep ions in a stable orbit. This is the 'Penning trap' field topology. If you put a magnetic field force outwards and an electric force inwards, or both inwards, then the beam is unstable. The only stable configuration is a magnetic field force inwards with an electric field force outwards.

It is unclear to me if Ralf Stenbacka actually fully understood this, because it doesn't appear to be the principal reason for him to choose this field topology. His idea is that the cyclotron acceleration inwards is opposed by the electric field. This e-field can then be switched off at any given moment. As the electric field is switched off, so the ions in the circular beam get pulled [more strongly] towards the centre, taking a smaller radius, and they end up all falling into the centre-point of the ion ring simultaneously, thus the idea is to get all the fast ions in a tiny space all at once.

I will award 10 house points to the first person to spot the fatal flaw in this fusion energy idea!......
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: Fusortron?

Postby Doug Coulter » Tue Nov 09, 2010 3:19 pm

Well, without more time and effort than I have at the moment to decode the not-very-clear writing in that patent (nothing special there, they all stink that way, it's deliberate for later legal wiggle room), and depending on Chris' observations, I'll say this. (and BTW Chris, it was your idea I thought I had a minor improvement on, not this one I'd never heard of)

The H field doesn't inwardly direct any charged particles. It simply bends them if they are moving across field lines, right hand rule -- if they were the other charge, or going the other way across the lines, it would bend them away. Cutting off a central E field that was holding them farther out than the usual cyclotron orbit means they'd only collapse to their normal cyclotron radius, which might make them denser than they were...but they are still all going the same way, and still all repelling one another via space charge, and therefore spreading out...so, no collisions to speak of. Therefore, no fusion. They spread out due to space charge, in both time and space (though if the RF is still there, it will attempt to keep them bunched in time). So I can't see how that works at all. It takes relative ion motion to get collisions...relative center-of-mass momentum, and I see no way this produces that.

I could have missed something in the 3 minutes I looked at the patent, and probably did. Most of the rest is very standard beam storage ring kind of stuff with the sensor signal taking a shortcut across the storage ring to a kicker and so forth.

So, what did I miss?
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Fusortron?

Postby chrismb » Tue Nov 09, 2010 4:58 pm

The magnetic field is only local to the vicinity of the ion storage ring. So as the efield is cut, the ion path tightens up and leaves the storage ring (as the mag field is directed such that the radial force is towards the centre) but as the ions emerge from the field then they head straight to the centre. The idea is that if they miss then they simply find themselves back over on the other side of the storage ring and the efield is switched back on so that they recover back into the storage ring for another go. See fig 10, which is the most descriptive on the ion path (during an efield switch-off-on)

As there are [intended to be] ions everywhere in the ring, the proposition is that ALL the ions in the ring at any one time will then head straight for the centre.

It's just a Joe is speculating - cyclotron orbits on the outside, then fusor-like in the middle.
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: Fusortron?

Postby Joe Jarski » Tue Nov 09, 2010 6:59 pm

I'm still trying to make sense of the patent. There are several things that either I don't understand yet, or it just doesn't look like it would work - unless there are some important details left out, but a few problems that I see are...
1. Pulling the ions out of the storage ring to the center
2. Getting the ions to go to the center where they would collide
3. Getting them back into the storage ring without hitting the chamber walls
I'm very much a beginner at this stuff, but is it even possible for the ring magnet to cause enough of a bend to send the ions to the center?

I'll have to get a sketch posted of what I had in mind.
User avatar
Joe Jarski
 
Posts: 231
Joined: Thu Sep 16, 2010 8:37 pm
Location: SouthEast Michigan

Re: Fusortron?

Postby Doug Coulter » Tue Nov 09, 2010 9:35 pm

(Joe, I have most of the same issues with it -- but yes, you can bend *slow* ions easy without much of a magnet, of the members here who are actual beamline experts, only JohnF has been tickled into posting so far -- and see pix of his lab for the magnets it takes to bend fast ions...not tiny ones.)

Ok, so if you have fields the right magnitudes and all (which I'd have to work out but something smells rotten here...) AND you had more than one bunch going, so (at least) two bunches could pass through the center at a time doing this trick -- assuming near 180 degree, OK, some would hit. The instant they are out of the field, space charge causes each bunch to blow up unless they are going so fast they just don't have time to defocus much due to that.

Also, unless you're going really fast, they never meet in the middle -- the huge charges needed to have any luminosity take a lot of input energy to make it to collision, rather than just spray out randomly before they get close enough -- one of the reasons you don't see fusors running on 5kv. But the approach doesn't work unless the ions are going slow enough that the field can bend them that sharply toward the center on the way out of the outer ring. I'd have to look at the focusing, but it looks to me like the field that would tend to plane focus them on the way "in" from the ring, would then defocus them coming back, for starters.

Also, isn't the center electrode kind of in the way here? Oh I see now -- cute trick. But all this is so poorly written and hard to get through, doesn't look like it's worth it to even take it any farther. The fact that like most bogus patents he's adding things on how to collect the energy he hasn't manage to make and obviously hasn't run the numbers on tells me a lot. Yeah, 6Li will give off some more energy if a slow neutron hits it (uranium might be a better multiplier,,,,duh) but thinking that enough of the neutrons can be slowed and intercepted that way as to think you're going to get in the range of 1::1 captures per neutron produced -- priceless ignorance. An open message to everyone (not just here) -- if you get gain, just call the power plant guys and have them do the rest. It's a done deal at decent efficiency. Just tell 'em you got this thing that stays yellow hot no matter how you try to cool it, and you'll get a knock on the door! If you need a phone number, I have a few of those. Lerner and his un-makeable energy capture scheme comes to mind. Patent away Eric, no one will ever need or want that silly crap even if you could make it.

I'll give up on winning the meaningless points. If nothing came of this since '89 we already have an (non) existence proof -- it doesn't work, and the patent's expired too -- so free for anyone to play with as well, so if anyone else thought it could work, they'd be doing it. I'd rather get Chris' thing up here someplace and not waste time on this. Unless he's changed it a lot it looked to me like it could work (only had semi-minor engineering issues mainly), and that makes me wonder why he's being so reluctant to expose it to those who could tell the difference and maybe even contribute to some success.

Wouldn't it be a lot easier to just make two cyclotrons inside the same H field such that the beams in them would rub like gear teeth of two gears turning the same direction would at their edges? Or three? All you'd need is more appropriately shaped "dee" sections in there and the right RF drives. Or just two intersecting beam storage devices...with the usual re-bunching and re-focusing fields to recover most of the ones that merely scattered a little bit? Of course, the closer you get to collision, the bigger the scatter angles, and no scheme handles that one well.

That is what got me interested in the cone trap beam-ring papers -- being pure electrostatic, one could have counter-rotating beams in one of them, and by adding the usual re-bunching and re accelerating cavities to the stuff in the papers, along with whatever refocusing quads needed -- you're there. Heck, I even bought the toolage to make those and got going, but I'm doing so much better on the fusor with every new thing I try these days, it's beginning to convince me it's not quite the dead end most think of it as. In other words, not an utter waste of time, just a waste of time re getting to gain that way alone. As a super improved neutron source, maybe not a waste of time at all, it's getting so fierce that way we are running less and shorter runs for our own safety. Doesn't take that many microwatt-hours of fusion to make the whole lab start being radioactive -- and my kitchen isn't that far away.

The real limit to colliding beams is (the lack of) space-time bunching, and the fact that you can't maintain either at high luminosities for long times or distances without space charge blowing it all apart. (see klystron or TWT design -- same problems) So you need very short focal lengths and decent energies to get momentary luminosity right at the interaction, but diffuse elsewhere so it's controllable at all. Which is why so far beam on target devices that even give up the nice center mass conservation law still beat fusors by a fat margin -- at least, the target is dense and harder to miss entirely than a particle in a diffuse beam. And that's without smart target tech, which I'm working on as a side issue; for the moment, if it looks like paying off, it will be the main issue very quick -- the advantage of being the owner of one's own lab is being able to follow your nose and no BS from bean counters, managers, and office politics.

There is a reason they need to use 30kv in a CRT TV to put maybe 10-40 watts on screen phosphor (a ma or so) in small dots (good focus, but still the beam is nearly all empty space at that current density -- a few fruit flies in the cathedral) with a decently long focal length with mere electrons, which have sqrt mass ratio issues at the same energies (voltages) compared to ions, here. The factor in this case is about 60 for D -- eg for the same space charge defocus effects, the ions need to have 60x the energy for the same current and FL to have the same beam-spread due to self repulsion...and that would be at the same luminosity as the E beam in a CRT -- low, no where near even the density of a gas, much less a solid. Fairly depressing when you run the numbers.

This is very much born out by my data here, where as I reduce the gas pressures and current, I get both more fusion and more Q along with better visual focus. Reduced scattering off orphan neutrals doesn't explain all of that, though it can't be hurting. I'm still working up enough data to make pretty parametric plots in post-able form, but my notes tell all. Trouble is, down there, the thing isn't self ionizing, so you have to provide the ions with another setup, and with the plain DC fusor, recirculation is actually pretty much nil. Which is why I put up that two grid approach, one an ion trap to gather and bunch ions, then fire them all at once through the center to be recovered after the firing for another try -- driven, not depending on luck or magic.

Until I did the low pressure/current thing, the differences between sloppy made and designed grids and accurate ones that could make a good focus didn't show up -- other things made more difference. Now, that has become important.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Fusortron?

Postby Joe Jarski » Tue Nov 09, 2010 10:04 pm

Here's a very rough sketch of what I had in mind. Blue being the magnet poles, tan being the chamber, red & green are the D's and the white buttons would form a virtual grid. Ions would be injected tangentially and accelerated in the gap like a traditional cyclotron, however, at a certain energy level they'd get to the outer edge of the D's and find a steady state orbit just beyond since they no longer cross the gap, but are still well away from the chamber walls. Then the accelerating field stops and the grid switches negative. After scattering there'd be a random mess that would eventually cross the gap and accelerate again.

ftrn1.jpg

OK, I think I'm done with this one, move along now.
User avatar
Joe Jarski
 
Posts: 231
Joined: Thu Sep 16, 2010 8:37 pm
Location: SouthEast Michigan

Next

Return to Theories and speculations

Who is online

Users browsing this forum: No registered users and 2 guests