Group Ion gun design - ideas only

[Richard Hull]

I would think that a group designed gun, while being of a specific internal design dimensions to satisfy ion optics design criteria sould also be adapatable to a number of mechanical configurations.

This would take the form of various mounting methods and high voltage or extractor voltage entry and exit arrangements.

Mounting flanges:

Certainly any one here who is contemplating using ion guns is no longer in the pee-wee class operator regime. As such, I would think that only conflat flanges should be considered on our approved design. None the less, it should contain a body barrel of a diameter that will allow it to be attached with a KF-40 or KF-50 type fitting as well.

The conflat of choice would be the ubiquitous 2.75" conflat which will allow up to a 1.5" ion gun body diameter.

This 1.5" diameter might be the one we settle on just to keep standard parts and tubing which is readily available as the goal for the entire project. In theory, we should find all parts in standard vacuum fittings and material supplier catalogs, (Lesker, Duniway, etc.), and not send anyone off into a custom assembly area. (Though they could certainly do that should they desire). This is not to say that some special tool work might not be demanded, but that at every turn, we should look to a stock solution. in the mechanical design of the device.

This has been just an intial pass at first thoughts should this effort blossum.

Richard Hull

[Brian McDermott]

Are the Farnsworth designs too out of date by today's standards? If not, they may provide a good starting base for an amateur design. Do any drawings (other than those found in the patents and on the Farnsworth Chronicles webpage) exist in the public domain?

Also, the technique described in this post looks interesting: viewtopic.php?f=12&t=4966#p32274 .

[Starfire]

The problem is that ion-gun design and production, is a separate ball game. It will take a separate approach to the construction of a suitable chamber and detract from any ongoing spherical fusor project. Though it may work as an add-on to existing chambers. That said, it is food for thought and worthy of a design and costing exercise. I, for one, welcome any ideas and had started a gun construction - Phil Fostini's posting ;- viewtopic.php?f=12&t=4926#p32234 shows a good design approach.

[DaveC]

A 2.75" conflat flange mounted ion gun should be quite adequate, bearing a superficial resemblance to the "nude" ion gage.

A tungsten ribbon directly heated cathode can put out a few mA with ease, and the ionizing voltage needs only to be 100 volts or less... to make ions.

What to do with the ions, thus created, is an important consideration here. Do we shoot beams of them at another similar gun across the chamber from the first ion gun? Do we let them reflect off a repeller electrode at the far end of a cylndrical .tube? (This was the idea of the reflex fusor, we discussed some time ago.) Do we aim into the voloume of the standard spherical fusor? This needs to be thought through.

I would urge the developing of a very simple electrode design to allow as many as possible build ion guns.

With reasonable care in construction, an ion beam should be able to be focused to 10 microns or smaller spot, At accelerating voltages of 10-30 kV, a pair of colloiding ion beams would have a much higher ion density and thus a greater collision probability than we normally get in the spherical fusors.

That said, imagine the improvement in neutron production if a large number of small ion guns were located at regular intervals on the periphery of the fusor shell, with their ion beams aim accurately at the center.

Been mulling this over for a while, but as yet do not have a complete ion gun design for this type. Again basic simplicity seems very important.

The objective, I would think, is to eliminate the inner grid cage... as it intercepts a large amount of the input energy, from gas molecule collisions.

I will try to set up something of a "workbench" (that's what SimIon 7 call the test bed) si that ion gun design and placement can be modelled fairly quickly.

Dave Cooper

[3l]

Hi folks:

I am using Phil's original ion gun.
It has performed well in operational service.
I have used it for 100 hrs now.
It is a critical part of the gun target experiments.

Happy Fusoring!
Larry Leins
Fusor Tech

[Richard Hull]

Thanks for all the inputs.

Dave,

I think the collisional beam idea is good as I touted this before, but still with a loop or ring to urge their impact and accelerate towards. A simple duo-plasmatron should suffice and simple components and simple design is very improtant, as you note.

It is off to do some scrounging now to see what is avaialble prefab. If anyone runs into a stock of usable surplus pieces, they should be purchased in a quantitiy where they might be shared amoung would-be ion gun folks.

Richard Hull

[Andrew Seltzman]

Here is what I have so far for a stock ion gun construction. CAD grid is 1/8".

The entire assembly will fit in a 2.75" CF half nipple. The main gun assembly is constructed within a 0.75" OD, 0.5" ID by 2.5" long high alumina tube.

The extractor cone is SS and is friction fitted into the alumina bore (cool down and insert). The extractor cone connects to the outer grid by contact to a SS wire. Outer grid is grounded or held at negative several hundred volts. A NdFeB ring magnet is positioned externally around the tube.

A 2 conductor HV feed through runs +5KV bias to the filament via a microwave transformer with the diode reversed. This runs the filament and provides the electric field required to accelerate deuterium ions. Deuterium should ionize near the surface of the filament. Additionally the extractor cone will emit electrons that will be accelerated towards the filament ionizing deuterium along the way. This would create a rather wide energy spread, however it is simple.

Filament connection wires will be feed through a SS disk and insulated via thin ceramic tubes. A SS tube (not shown) brazed to the plate will be bent into an elbow and crimped to a 1/8 OD ceramic tube at an angle normal to the gun body.

The ceramic tube will exit through the side of a 2.75 CF collar with 1/8” NPT port. A 1/8” bored through swagelok fitting with Teflon ferrules will seal the ceramic tube to a vacuum. The ceramic tube will be adapted to a 1/4” SS tube immediately after exiting the swagelok.

This is probably no where near perfect, but it’s a start. Any ideas?

Andrew Seltzman
www.rtftechnologies.org

[3l]

Hi Andew::

The design looks good except for the teflon part and the magnet.
Under vacuum and heat teflon will degrade and out gas.
The ion source will get hot as blazes,even oil cooled and your Nd Iron magnet outside or inside will demagnetize above 400 degrees.
I quit water cooling because you can get steam bubbles that pop in the cooling tank and scatter water in the shop.
The gun Phil designed all vacuum is pumped in front of the extractor tube...D2 is admited right behaind the filament.
I bias the gun at 6 kv bias but that is only because I had a xformer on hand that made 6kv.

Happy Fusoring!
Larry Leins
Fusor Tech

[Andrew Seltzman]

Thanks for the insight. I have several specific questions about the gun design.

These guns would be brazed to the hemispheres of a fusor, vacuum would be supplied on the extractor cone side.

I was previously worrying about the magnet demagnetizing, now it seams in this design that is certain. If the magnet ring was moved from in contact to the ceramic tube to in contact with the conflate half nipple, would it remain cool enough? This is more of a thermal dissipation question. I was thinking that the filament would generate no more heat then say a 50W light bulb. Of course there is still the plasma generating heat though. While the high alumina will get very hot, it will be positioned on standoffs so that it is not in contact with the wall. In that case could the wall be cooled convectionaly, either passively or by forced air cooling?

I have had no problems with Teflon in vacuum systems when kept cool. If the wall is kept sufficiently cool would this still work?

Secondly, would my design of grounding the extractor cone and biasing the filament at +5kV work? I don’t believe the filament has to be at a negative potential w.r.t. ground as long as the electric fields are the same inside the ion gun. Can you confirm this?

Andrew Seltzman
www.rtftechnologies.org

[3l]

Hi Andrew:

My ion gun setup ala Phostini stays at 800 degrees F at all times.
you have to think through even the wire connections.
12V @ 10amps is 120 Watts..... is the needed current
to get the filament into the proper emission range for 40 ma.
The only change I made to Phil's design was to remove the original filament and replace it with a heated cathode, thoriate the oxide cathode to increase the electron yield.
On cooling if you have a radiator and blew large cfm of air over it ,you might keep a small ion source happy.
The direct dc approach will work but I just use ac for 1/2 wave acceleration of the deuterons. My ion source only puts out deuterons on the negative cycle.
Filaments are fragile and get brittle as they thin down in use.
You might want to design in a heated oxide cathode as filaments deposit tungsten every where .. an oxide cathode prevents this common shorting malady.
I got mine on Ebay by buying a large russian tube and carefully busting it and removing it's cathode.
Joe Zambelli did the same for his ionizer grid for his fusor.
Rf has no heating problem and no filaments to burn out even at very high rates..


Happy Fusoring!
Larry Leins
Fusor Tech

[Adam Szendrey]

Actually N35 grade (most popular) NIB magnets loose their magnetic force far below 400 degrees celsius. They start to loose strength at temperatures as low as 80 degrees.
There are special NIB magnets for high temperature use (expensive), and samarium cobalt magnets have much higher temperature limits (their curie temeprature is around 800-900 degress celsius, ofcourse they start loosing strength far below that, if i recall correclty they can be used reliably up to 300 degrees, a guy at the magnet shop told me that).
Though they are not as strong as NIB magnets, it's the second strongest magnet type (i have no info on cerium based rare earth magnets).
Cheramic magnets (BaFe, SrFe) can be used up to 400 degrees, they are cheap ,but weak (the so called isotrope types are stronger).
I thought i share this info , maybe it's useful.

Adam

[MARK-HARRISS]

Hi Larry, I was planning some experiments to recreate early tube filaments:
Marconi originally used lime on a thick platinum wire as the lime is stable in air and fairly rugged. The rare earth filaments of strontium and barium oxide I understand are applied as a carbonate (barium meal anyone?) which degrades to an oxide under a vacuum with heating. That chemical company you mentioned earlier has several likely candidates listed as fireworks ingredients.

Mark H

[walter_b_marvin]

would it be possible for the design to incorporate a simple solid target
and detector. This would facilitate the poor amoung us to have some hardware to play with while we save for better stuff. It would also let the theorically chalanged among us to get their feet wet, so to speak

[DaveC]

Just a couple general comments here.

1. In order to make Deuterons you need no more than 100 ev. Actually you need less than 30ev. What you need far more than voltage is amperage.. since that is the measure of ionizing particles... namely the electrons. So for a high current ionizer, you really need.. lots of cathode surface area.

2. For those thinking of the low temperature oxide cathodes... these are definitely worth a long careful look. The most commonly used cathode coatings are the Tri-Carbonates consisting of Barium, Strontinum and Calcium carbonates. According to Rosebury's Handbook of. Electron Tube and Vacuum Techniques ( good section on oxide cathodes begins on pg 96 of the 1993 2nd edition) these are best used over a nickel cathode base. There is a several step bake-out and activation process to get a triple carbonate cathode converted to a triple oxide cathode. Word of caution, however. THESE ARE INTENDED FOR USE AT PRESSURES BELOW 10^-6 Torr. Most of Fusor work will be far above these low pressures., and positive ion bombardment will quickly poison and ruin oxide cathodes.
Use this for comparison, but I would suggest Tungsten heaters here.

3. I think those planning to use magnets (a good idea to boost collision probability) should take the time to calculate the radius of curvature versus magnetic.field strength, for various anode voltages. Unless you insist on ionizing at multi-kV energies, you should be able to use commonly available ferrite ring magnets like are used in Magnetrons and... hold on here.... loudspeakers. You typically need a 500 hundred gauss for kV energies to bend into a few cm radius paths. At lower energies, you need quite modest B fields. The formulas you need are:

Radius (meters) = (mass x Velocity)/ (charge x Mag Field)
R = mv/qB
(kg, meters, coulombs and Tesla. and seconds.)

The second, for converting the velocity to voltage.:

Energy = qV = 1/2 mv^2

where V is volts, q is the electronic charge 1.6x10^-19 coul, m is deuteron mass in kg and v is velocity m/sec. Both formulae use the same units.

Probably everybody knows this, but in case it's only a dim memory.

4. Ion gun materials need to be suitable for high temperatures > 1000 C. Tantalum or tungsten wire is good, nickel seems to be okay too. Stainless steel is reaching the upper limit of its healthy existence.

5. For tungsten filaments/cathodes - you may need to light the filament in a low pressure atmosphere of Hydrogen to remove the Tungsten Oxide. Without the oxide removal it will not emit worth a hoot. If you can,.... use the recommended chemical cleaning bathfor Tungsten.: Boiling 5% Sodium Hypochlorite solution (aka Chlorox or common bleach) This works very well. Its best done in a beaker, a simple, stovetop process.
Takes about 20 mins or so.

Hope this helps a little.

Dave Cooper

[Richard Hull]

Dave just gave a lot of good info. I printed it out and it now goes in my ion gun project file pile.

Heavy coiled tungsten is the best emitter at our pressures. Real ion guns producing real ion currents in the multimilliamp range have filaments that cook at 20 amps or more.....much more!

Extractors are usually tantalum or tungsten and even these will erode over time. One thing you can go to the bank with is that your actual realized ion current will be far, far less than the ionizer current.

If you can actually get 10ma deuteron current, you have a plus ultra gun indeed!

Richard Hull

[DaveC]

Some more good points Richard.. thanks.

I have been doing a bit of electron trajectory studies for an egun project, and discovered that too strong an accelerating voltage on the extractor, leads to excessive collisions woth the extractor electrode. There seems to be an optimal set of voltages for the electrodes, which can allow nearly all the ions ( true at least for electrons ) to pass through . If implementable with the higher pressures for the deuterons, should (may?) allow much lower grid operating temps.

Morre as I learn it.

Dave Cooper

[MARK-HARRISS]

Hi Dave, yep you have seen the whole picture: I have been focussing on one aspect as usual without looking at everything interacting. I should have my vac chamber finished by the end of the year and one of the first things i want to do is try a series of materials as triode cathodes and see what performs best: I had planned to just use vacuum but hydrogen would be better, I could test an electrolysis hydrogen generator at the same time too I suppose.

Two setups I have hopes for are:
1. Thoriated TIG electrodes used as filaments with maybe some carburising beforehand.

2.Fine diamond dust rolled into a metal surface that has a good electron work function (doped diamond has a good "work function" or whatever it's called) so I may be able to luck out with natural random doping like I did with my homemade blue SiC LEDS. I would also heat the diamond surface indirectly with a filament.

Hydrogen will attack carbon such as graphite but hopefully not as much with diamond.

One of the reasons for going to this much trouble would be to make some vacuum/gas devices as a sideline such as triodes and nixies, also I would like to experiment with activating thoriated tungsten filaments.

regards
Mark H

[Richard Hull]

I have thought about Th activated tig electrodes. They have made them in .010 sizes and I have a few. I have also heard that they are discontinuing them. My local shop can't get them from any supplier. This would leave .020 as the smallest available. With the .020, of course, you are talking about 40-100 amp filament currents. No shortage of electrons there.

Richard Hull

[walter_b_marvin]

I have some simple questions maybe someone might like to try to ponder over

1) Is this ion gun a standalone beaste or does one need a fusor with vacuum chamber

2) What are the vacuum reqs

3) Does this guy allow a solid target

4) what kind of power supply is required

5) What kind of detector is best with this guy, and what would it detect?

6) Any special machined parts reqd?

[DaveC]

Mark - you only need the H2 firing.. to clean the Tungsten filament surface in order to get good electron emissivity. Then you can operate nicely in high vacuum. I think tungsten needs to operate below 10^-6 Torr for a good lifetime and useful output.

If the filament gets contaminated, you just repeat the H2 firing, and or clean it again.. and you should be ready to go again.

I think the diamond will not do too well at the really high temps, but.... doped diamond was being developed as a field emission material, but I think some of the shine is off the CRT applications, because of the hydrocarbon contamination - heater decompose the HC's which allow gettering to occur in sealed systems. With a high pressure continuously pumped system, the issues probably will center around decomposition by the D2 itself.

What you have in mind sounds very interesting. Please keep us up to date with your progress.

Dave Cooper

[Adam Szendrey]

Essentially an ion gunned fusor does the same as any fusor...it accelerates and collides ions. Either by using a target, or two or more guns.It does not need a fusor system to operate it can be standalone. You will need a vacuum chamber (tubes, and a collision chamber) to do it, but a large chamber is not required..well this depends on how many guns you are using , etc..The vacuum needed is about the same as for a fusor (microns range, Larry had a vacuum of 10 microns in his thick target experiment).
As Larry's great desing shows, a solid target is absolutely possible, and succesfull.
You will need a filament supply (low voltage, high amp), and a HV supply to create and accelerate ions.
The same detection gear is required to measure the output (x-rays, neutrons).
You might need some machined parts for a gun, it depends on the design.
Actually you don't really need to much extra stuff to build an ion gunned system, once you have an operational fusor.

Adam

[Richard Hull]

To answer numbered questions.

1. An ion gun is quite capable of working by itself and needs nothing else. It is a complete assembly.

2. The device does require a vacuum pumping mechanism to facilitate ion production and distribution of the created ions beyond the gun.

3. Ion guns don't need a target they are separate from other systems and are add ons. They just squirt ions. Where you squirt 'em and what you do with them is up to you.

4. In general the ion gun can requires as many as three separate supplies. Two can be used in some models.

You will need a source of low voltage high current for the filament or heater to create electrons. You will need and ionizer voltage of about 50-200 volts at up to 1 amp and you will sometimes have to have an extractor voltage with is very high voltage by comparison to the other supplies (~1-20kv)

5. Detector?? not needed........??

6. Traditionally, there are a lot of very special machined do-dads in these puppies. Adroit amateurs can keep this work down with careful design and use of available materials.

Richard Hull

[Brian McDermott]

There is another post to this ion gun forum regarding an ion source that requires only one power supply, and can generate adequate ion currents.

[MARK-HARRISS]

G'day Richard, I believe dental supply shops have really tiny TIG electrodes (at dentist's prices) for use by orthodontists when they fabricate titanium and stainless braces from wire stock. Also I read that thoria is a natural contaminant of tungsten so just by buying the wire you may be able to get it
with thorium anyway depending on the source.

I used to have reels of it for photocopier corona wire repair about 7 years ago but it's of the order of a few thou and to small for this sort of work.

Mark H

[MARK-HARRISS]

Thanks Dave, I'll do just that: I mainly hope to find cheap commonly available emitter materials as all the real research on filaments has been done in the 1920's and 30's and there's no point reinventing the wheel. I have been doing some interesting reading about filaments at: http://adswww.harvard.edu/

My main understanding of thorium tungsten filaments is that you can operate them hot enough to have a surface coating of thorium to provide the emission or at a significantly reduced power you can operate them if you have carbon in the filament to reduce the thoria in the filament. I have read that the ceramic transmitter valves like the 4CX's etc all do this and they die when the carbon runs out but can opened and recarburised.

This article is a good read: https://dspace.mit.edu/bitstream/1721.1 ... 718339.pdf

If you could get straight thoria ThO?? you could mix in some small amount of molybdenum powder to make it conductive and get an amp per cm2 of electron emission.............hmmmm, burn some gas mantles + molybdenum powder and press it into a mould......... hmmmmmm