Page 1 of 1

Beams extending beyond outer grid in multigrid systems

Posted: Wed May 30, 2012 11:49 am
by Richard Hull
Frank Sanns has noted beam streamlines extending beyond the outer grid in a multi grid system. A good deal of discussion centered around his positing that the reason was due to increasing energy of the ions with recirculation. A number of counter arguments against this explanation were offered at

viewtopic.php?f=18&t=7899#p55858


The discussion is hereby moved to continue here.

The system Frank devised is an odd one and without definitive operation at full, useful fusion potentials with good neutron measurements, any advantage over a simple fusor remains totally moot, regardless of all our speculations.

From my point of view, this extension of visible beamiing can readily be explained due to a number of additive factors and causitive agents.

The visible beams, (star rays), are due to recombination in natural streamlines of recirculating ions and electrons in electrostatically formed beams centered on inner grid, (cathode), openings. This is easily seen with the eye in viewing the normal fusor interior.

We run at a relative high vacuum pressure such that we ionize easily and visibly at the ragged edge of glow discharge and a runaway arc or townsend discharge modality. Thus, the mean free path for ions is rather short, on the order of 2 to 3 times the diameter of the fusor chamber, at best.

Ions being decelerated as they head towards the outside grid or even shell of a fixed fusor stand a good chance of being neutralized on their way. (Causing the visible light we see.) Some fraction of these fast outer grid or shell bound neutrals can and will certainly continue on past the outer grid, (extra-outer grid volume), and become reionized in gas collisions on the other side of the outer grid and recombine again along this path, thus causing a ray or visible beam to extend beyond the outer grid. Those positive ions formed in this extra-grid zone will be repelled by this similarly charged grid and be lost to the system only to recombine in the extra-outer grid volume.

These are system losses and in a single grid system wind up as implanted neutrals in the shell wall where the streamline touches it. In the dual grid system they are still lost to the recirculatory system being outside of the elecrostatic system in which they were originally formed.

I am wondering if Frank's system, as seen, is not just some sort of lensing arrangment with no recirculation occuring at all. I note that the extra grid beam seems to be emitted from an effective conductive point at the base of the secondary grid. I guess I just don't see anything interesting or novel here.

Richard Hull

Re: Beams extending beyond outer grid in multigrid systems

Posted: Wed May 30, 2012 4:15 pm
by Chris Bradley
Frank S. wrote:
> The mean free path is inversely proportional to the square of the size of the atom, molecule, or electron. ... This says that the vast majority of ionization in the fusor must be due to ions colliding with neutrals to form more ions.
This assumes there is nothing substantially bigger than an ion, does it not? But would fast neutrals not be substantially bigger than bare deuterons? If so, then would your own logic not suggest the majority of ionisation is from fast neutrals?



> 2. If ions are formed from other ions, then the location of the light must be the location of ions.
This is assuming that ions are the majority ionisers, so it's dependent on your point 1 being correct.



> 3. ...Clearly some mechanism is different when polarity of the inner grid is switched. This difference must be a mechanism of recirculation where ions can gain additional energy on additional passes.
Sorry, I don't follow that leap. If the central grid is positive, then it seems typically more successful at generating ions. It seems to me that a fusor is somewhat reliant on there *not* being too many ions, because if the volume gets filled up with conductive species then you can't hold high voltages. I'd respond, then, by saying that if you reverse the current it is actually a 'better' ioniser of material, it's just that that 'better' in one thing (generating more plasma) is worse and another (generating high energy collisions). It just depends on your point of view. If someone was trying to make a plasma light, looks like they'd not be picking the -ve central grid config!



> 4. Increasing current while maintaing constant voltage (typically in the 5kv to 15 kv region for ease of control) will quite dramatically change the look of the star (recirculation). This should not be as the electrical field has not changed, there is just more field lines of it.
Could you clarify, please? If there is more current then, surely, the assumption would be that there will be more ions in these recirculating beams, and the space charge these beams can reach is a function along with the beam width, such that:- more current = more charge = higher space charge = beam widths wider = increasing opportunity for deflections away from optimum recirculation paths?



I'm not saying there is anything notably wrong with your observations, just that it doesn't seem to be as conclusive as you are suggesting. There appear to be other, possibly simpler, explanations.

Re: Beams extending beyond outer grid in multigrid systems

Posted: Thu May 31, 2012 2:18 pm
by Frank Sanns
They were pictures. Just pictures of some operating conditions that I was set up to run and share with those that are interested. I have no agenda here but to share what I saw and observed. I was asked and gave my opinion of what seemed unexplained in the photos. Sorry if running at 15 kv and 6 ma is not sufficiently into neutron territory for this experiment to have any value. I refuse to report neutron numbers in a fusor that is not conditioned after many days of running to stabilize it. The numbers would be meaningless no matter what they would have been. Get what you want out of the pictures and draw you own conclusions including their value or lack there of.

Frank Sanns

Re: Beams extending beyond outer grid in multigrid systems

Posted: Thu May 31, 2012 3:02 pm
by Chris Bradley
Likely why Richard moved it, Frank, so as to not detract from the experimental photos themselves. I think your experiment serves a valuable purpose of opening a discussion of a very important and fundamental aspect of fusors that >10 years of amateur experiments still haven't got to the bottom of ... what *are* the 'beams' and their cause, are they essential to see before fusion? So, thanks for sharing your experiments with us. These are great questions, and if experiments can be constructed that give answers, they may (or may not) unlock significant future design steps or new experiments.

Re: Beams extending beyond outer grid in multigrid systems

Posted: Sat Jun 02, 2012 11:52 pm
by Dan Tibbets
My two cents worth (again).

Under ideal conditions the ion falls down the potential well to the grid. At that point the ion no longer sees an electric field from the grid. From that point on their path and direction are determined by the space charge from other free ions and electrons and the ions momentum. The grid has no effect due to Gauss's Law. This is of course dependent on the grids being close enough together, depending on the plasma frequency. Simple consideration is that the simple wire grid satisfies this condition. The shape does not even need to be too symetrical from this standpoint. The ions will deflect from the center somewhat due to some central virtual anode formation, and pass to the outside of the grid, at which point they will see the negative charged grid again and begin decelerating. They do not see the positive charge (or ground) on the shell,unless they hit it due to up scattering. Again due to Gauss Law. Even if the ion had zero angular velocity at the start, it would pick up some on the first pass, not due to the grid, but due to the central virtual anode (ignoring collisions), so the first orbit will have some elliptical characteristics. But, assuming the grid is symmetrical another aspect of Gauss Law will cause the ions to be accelerated towards the center, even when the ion is very close (and still outside) to a particular wire.

And, I am skeptical that fast neutrals dominates ionization. The cascading secondary electrons probable dominates. The first electrons from the cathode grid starts the process, and the secondary electrons knocked off of neutrals leads to a logarithmic cascade of ionization. A fast neutral could knockoff an electron or even attach an extra electron (negative ion) but it is my impression that this effect is minor compared to the cascading secondary ionization driven by free electrons. The initial input electrons has plenty of energy to do this, and importantly, the cross section for ionization by a free electron peaks at ~ 100 eV. The secondary electrons will thus have ~ 100 eV of energy and this is plenty to knock off several more electrons before the net eV energies drops to significantly lower ionization cross section energies. Each (eg) 10 KeV primary electron may ionize 10-20 neutrals, and the resultant free secondary electrons with an average energy of ~ 100eV can cause a further 5-10 ionizations, then each primary electron may cause ~100 ionizations. I have seen this ~ number in print. This assumes the MFP for the primary electrons is ~ 1/10 the machine radius. The mean free path of the 'cool' secondary electrons is perhaps many thousands of times shorter, so they would not have any problem in propagating mayn ionization collisions.
This brings up a point about hot neutrals, say at 10KeV. They also would have the longer MFP. So even if they did initiate an ionization and a free secondary electron before hitting the wall, these secondary electrons would again be the direct cause of most of the subsequent ionizations.

Granted that if the neutrals dominate the population of ions and atoms/ molecules by an order of magnitude or more, the collision chances go up, but any neutrals accelerated by collisions or charge exchange travel to the wall without any containment. Compare this to the ion containment/ recirculation that may result in as many as ~ 20 passes. The electrons also fly straight to the Wall in a simple fusor, but they are being constantly replaced by the input current. One fast electron may lead to hundreds of secondary ionizations. Finally; considering the MFP, machine diameter, quality of the grid, spalling reactions, ion to neutral ratios, etc. complicates things tremendously.

Dan Tibbets



Dan Tibbets

Re: Beams extending beyond outer grid in multigrid systems

Posted: Mon Jun 04, 2012 11:31 am
by Richard Hull
Almost all the electrons emitted at the cathode hit the wall or shell. The MFP of an electron even at our high operating pressure far exceeds any ion MFP by a factor of 10-100. This might be why we are so good at doing fusion. The wall hits pop off lower energy electrons and ionize the gas near the shell where the deuterons can readily accelerate towards the inner grid.

That there is an electron cascade is obvious or this system would not work at all. We are operating a glow lamp here, for the most part, albeit one assembled and carefully controlled to do fusion. The cascade can run away as many here who actually operate fusors are well aware of. This is why careful shepherding of the system by adroit hands is a must.

The ideal place to ionize the gas is at the shell and this does seem to be happening when operation is at its peak performance level, (sweet spot), or we would not have accelerated deuterons to fuse.

That we have fusion at all is due to artifice of the operator in creating the special environment where the physics allows fusion to occur. A near perfect fusion of art involving the brain and experience-guided hand in the true artisitic sense and of pure science unable to do other than fuse when certain conditions it specifies are met.

Richard Hull

Re: Beams extending beyond outer grid in multigrid systems

Posted: Mon Jun 04, 2012 4:53 pm
by Dan Tibbets
I was waiting for corrections to my broad statements. High energy electrons hitting the wall and sputtering off low energy electrons (closer to the ionization cross section peak of ~ 100 eV) and these electrons promoting cascading secondary ionizations because of this, and the significantly lower MFP of these wall scattered electrons causing peripheral ionization makes sense.

Also, my comments about the Gauss Law effects may be a stretch with a wire grid. It would be more consistent with a wire grid with many loops and cross wire loops, or especially an otherwise solid sphere that has small holes cut in it. This would better focus the ions towards the center, but at the cost of less recirculation. Where the best compromise would be is an open question.

Dan Tibbets

Re: Beams extending beyond outer grid in multigrid systems

Posted: Mon Jun 04, 2012 5:30 pm
by Steven Sesselmann
Chris and others,

Quite a lot of work has beem done on the subject of ion velocities in fusors, at the University of Sydney. Dr Joe Khachan and his students have produced a number of papers (thesis) on the subject. Optical spectrometry looking at the doppler shifted spectral lines was used. I think they managed to get pretty accurate measurements. If anyone is interested in reading some of these papers, I might have a copy somewhere.

Personally I don't find it strange at all, that some ions have a higher energy than the outer grid potential. In a plasma with a Maxwellian distribution, there are always going to be some slower and some faster ions, so you would expect a bell curve if you plotted the energies on a chart. All it takes is a few ions with abnormally high energies, to start off the ionization process in the higher energy region of the fusor.

Steven

Re: Beams extending beyond outer grid in multigrid systems

Posted: Tue Jun 05, 2012 12:17 pm
by Richard Hull
Yes, it is the lower energy wall "knock-off" electrons that do the ionizing. It is the cathode that initially supplies them. A good read of "Ionized Gases" by Von Engel will show that beyond a couple of hundred electron volts, the MFP of the electron will traverse the gas environment without much ionization taking place.

I would dare say that there is a space charge of lower energy electrons just off the anode shell in a "sweet spot" functioning fusor. This is a prime deuteron creation zone that helps to warrant a decent population of fast deuterons within a properly operated device.

Again, all of this is idealized with the full and certain knowledge that a whole lot of other deleterious as well as aiding possible situations are present in the device as well. This one thing mentioned in the previous paragraph does probably explain where the bulk of our fusion energy deuterons are coming from, though. These high energy deuterons interactions with each other via recirculation and impacting chance based fast neutrals are the aiding factors in making the fusor function for us as a fusion device.

Richard Hull