Deuterium Accelerator and Electro-Static Power Supply

For the design and construction details of ion guns, necessary for more advanced designs and lower vacuums.
Post Reply
User avatar
Finn Hammer
Posts: 298
Joined: Sat Mar 05, 2016 7:21 am
Real name: Finn Hammer
Contact:

Re: Deuterium Accelerator and Electro-Static Power Supply

Post by Finn Hammer »

Dennis,

Sorry to hear you missed the corona at the tip of the needles. I was wondering about the possible difference between the 60Hz you are using, and the 24kHz that I am using, but thought better about mentioning it beforehand. It shouldn't really matter, since the supply is rectified, but who knows?
The bottom pulley is made of metal, and grounded, right?
Mine is, and I have covered it in electrical tape, to avoid the alluminium smearing on the back side of the belt, which you also have experienced.
Cheers, Finn Hammer
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Deuterium Accelerator and Electro-Static Power Supply

Post by Dennis P Brown »

Hello Finn, and I will soon see if the applied frequency matters using another supply I happen to have on hand (again, being a bit of a "pack-rat", pays off.) Using this I can perform the test you outlined (my preliminary day-light test tends to indicate it does): this "supply" consists of a ZVS (Zero Voltage Switch; a very nice Chinese built unit) which is a high frequency driver (supplied by a simple 12 - 30 VDC high current supply; luckily, I have such a DC supply.) I use this ZVS to drive a flyback - also, fortunately, this outputs a positive 35-50 kV (my charging unit must be positive.) While it tends for higher voltage (even when driven in the ZVS's lower operating voltage range), I will put it though my 150 M-ohm resistor series and see what this can do via a corona discharge tonight. My lower VdG's corona needle assembly is a few inches above my lower bearing but has a "grounding" plate across from that - the belt is between these two "electrodes".

A quick bench test using my HV probe as a grounding point on the flyback's output lead yields a strong corona field of 12 kV about 5 mm from the HV probe's pickup point; one can really hear the high frequency corona "buzz" as the probe gets close. A good indicator of what you are describing - not dark currently, so that test will be later.

I really need to replace my belt but frankly, sites are all over the map with what works and what's available (yes, the company you mention does appears to have been sold to a Chinese's company and their web sight does not appear to support English.) Guess I may end up ordering various materials and trying them.

Aside: I should not have gotten back to this old project because now, frankly, I am a bit angry at it for not working. So, I guess I will keep at it. That attitude worked well for my fusor and neutron generation/detection efforts, and even the very recent HV multiplier rebuild (my goal was 150 kV and it appears that I did achieve that (to no avail - the unit didn't have enough power)) but I did achieve that voltage, so I can cross that goal off my left over project list - lol.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Deuterium Accelerator and Electro-Static Power Supply

Post by Dennis P Brown »

Well, the good news, when I used a super sharp needle and this ZVS driven (high frequency) flyback system, a corona discharge was easily visible even in just low light. That settles the frequency question very clearly; 60 Hz is not going to work very well at all compared to high frequency supplies.

Strange that two or three people on the web have written extensive articles on sources and did not examine this issue and discus it. I know these people suggested 60 Hz NST systems but then again, I've only seen them talked about in theory, not necessarily used by them. But then, they might have decent belts and then, even a system that yields no visible corona might still charge a good belt material enough to give satisfactory results. Still, I have no doubt that the brighter the corona from the needles/pins, the more charge is being deposited upon the belt.

The bad news for my real device test, is that on the Van de Graaff, all I was getting was about one micro-amp current transported from the dome.

Again, most likely the belt material is the primary issue here. To a lesser extent, I noticed the ground plate for the corona is not in the correct position at all (an oversight when building) and that is not easy to address (but will be. Also, I will install the ZVS/Flyback in the unit as well) Still, some tenable but none-the-less, real progress.

So, getting and making a new belt is priority #1; it is just finding a material that will really work. Anyone with a source, please weigh in. Thanks!
User avatar
Finn Hammer
Posts: 298
Joined: Sat Mar 05, 2016 7:21 am
Real name: Finn Hammer
Contact:

Re: Deuterium Accelerator and Electro-Static Power Supply

Post by Finn Hammer »

Dennis,

Great news on the corona issue. The Habasit site looks like a dead end to me, it looks like the company has folded, and is just holding on to the URL for future reasons. Just plain impossible to get to a place where you can place an order, or even get material descriptions.
Come to think of it, the last batch of belts I ordered might be from another company, but from the same distributor. It may take some time to find the name, I will have a look for it tomorrow, have to dig the old invoice out of files.
The belt is important. Some people use rubber, some rubber bands used by physio therapists, and it works of course. But if you want to run fast and long, then nothing beats there conveyor belts. I had mine running at 6000rpm on a 4" pulley, that is 36 Km/hour and the only limiting factor here was the lack of ability to increase the revolutions of the motor. The top roller running op to 12000rpm might become a problem in the long run, though.
I hope I can get some info on the belts.

Cheers, Finn Hammer
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Deuterium Accelerator and Electro-Static Power Supply

Post by Dennis P Brown »

Thank you again Finn. I ordered some white transport belt grade material from a source recommended by someone else that builds VdG's.

For entertainment, I repaired my high voltage multiplier (of course I cracked the seam and oil was leaking) and connected it to my "accelerator" sphere (10 ich diameter.) I manage to get a stable charge up to around 100/120 kV by indirect measurement via the electric field on the HV probe; that is, when the probe read/achieved 25 kV when it was set four centimeters away (this was based/loosely calibrated on a previous run without the sphere and I had the HVM at near zero load and then I read the variac scale to determine the real voltage. Aside: 20% full scale is directly measured as 50 kV on my HV probe.)

In any case, I measured the sphere's voltage more directly via air insulation break down - an arc 5 cm long occurred (so this indicates about 100 - 150 kV depending on humidity.) This arc went to one of the accelerator's focus lens (a bit confused why that conducts since it is way out on a ceramic tube so not exactly grounded to anything directly.)

Finally, I did do a direct voltage measurement upon the sphere with a voltage divider (5 G-ohm to 1 G-ohm and got a 25 kV reading on the HV probe before break down resulting in voltage loss. Again, supports this high voltage of 125 kV. (scale of variac was about 35%.)

All in all, far better performance than I would have suspected considering the cap limitations - looks I might be close to 140/150 kV upper range I hoped for and worse case, at least over a 100 kV.

In any case, the variac and HVM saturated at 35%-40% of full scale: after that, the voltage would not increase due to severe arcing.
I was surprised to see internal shorting within the metal sphere!? I didn't think that could occur ...well, learn a bit more.

My next step (weekend?) is to get the accelerator tube under moderate vacuum (10^-4 torr or better) and seen what happens when I charge the sphere again. See what breaks down and how and why there is internal arcing in a metal sphere!

In any case, these results indicate I am in the voltage range I feel that would allow me to get the accelerator to work - then again, been fooled before but these measurements, unless one see's issues I'm missing, indicate my measured values are likely valid.

So, anyone want to weigh in and provide better insight/measurement methodology to refute or support my efforts at some really high voltage measurements?
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Deuterium Accelerator and Electro-Static Power Supply

Post by Richard Hull »

One must remember and realize that a 100kv reading off the spark length is worthless under any true load condition, no matter how minor the load happens to be. The entire charge and, thus, the voltage on the sphere, at break down, instantly disappears from that sphere and it must be charged again. The sphere is the source of the ultimate voltage as an isotropic capacity. How fast the charge and, again, the voltage, can recover is controlled solely by the rate of charge delivery by the belt and its efficient transfer to the sphere.

This is readily seen by moving the discharge, (grounded), ball closer and observing that the spark is more frequent. Amazing isn't it! A real loaded voltage might best be obtained by measuring the spark length that the VDG can deliver at a virtual continuous rate. Can it do this at 1CM?

Even the least interested by-stander can tell that a 10cm spark off a VDG delivered every 10 seconds is white hot and noisey..POW! A s---pot load of current was in that lightning bolt! You bet it was! As the discharger is moved in closer, the spark becomes more frequent and a bit weaker,.... less current. As it comes to a point of close proximity to the big sphere, the spark is almost continuous and the noise from the spark is grossly reduced, as is the current. Under any load actually connected via a wire lead to the sphere the voltage and current plunge even more. Simple physics

Unfortunately, a continuous load, as in an accelerator, might reduce the continuously available voltage, and or current, far below that needed to achieve a specific goal with that accelerator.

Trying to get a VDG to do real scientific work demands far more than getting a hot, snappy, 9cm spark every 8 seconds. Half of all useful VDG work will involve insulation, the other half, a study in rapid electrostatic charge transfer to a large enough and well insulated, isotropic capacity.

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
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Deuterium Accelerator and Electro-Static Power Supply

Post by Dennis P Brown »

Thank you, again Richard. All good points that offer better understanding of the real situation for such systems; not obvious for people like myself with near zero knowledge of high voltage systems. Learning about these very high voltage systems is tricky; issues such as the length of time between arcing is a point I hadn't considered and is rather critical in order to properly determine a valid sustained voltage/charge. Just as I forgot that since the accelerator is under vacuum (10 - 20 torr) it is an excellent conductor (duh; forgetting that is not something I should have done); hence, the internal arcing is rather obvious (accelerator dome/sphere to the internal accelerator tube) and also to the metal electro-static lens 5 cm away from the accelerator's sphere. Also,this arcing, as you point out, eliminates all significant stored charge on the accelerator's sphere and the arcs are causing severe loss of the accelerator''s voltage and charge on the sphere. Your discussion of distance and arc strength/time between arcs was very clear and encompasses the key points I hadn't properly considered.

My method of using a voltage divider was a more realistic measure and I still think the accelerator's sphere is close to the 100 kV value; that is, until the arc over/short does occur. I do get a strong electric field to form (between arcs, that is.) I do believe that the arcing issue can be fixed by reducing the system's vacuum to under 10^-4 torr. Then I will see what sustained voltage/charge can be achieved on the sphere (then again, maybe not.)

Just as in mastering fusor operation and the issue of creating a stable plasma, a learning curve is involved. This necessitates getting feedback from someone familiar with high voltage systems in order to learn the techniques and correct mistakes (both in understanding and execution) to determine a valid charge & voltage that is sustained rather than what I want or just assume.

While my primary focus will, for now, remain repairing and upgrading the Van de Graaf so that it at least functions and provides some realistic charge/voltage levels while isolated - and maybe, with some better design features - in the future can operate the accelerator, I will continue to explore the issues of measuring the high voltage created by my voltage multiplier. More to the point and relative to your post my determining if a significant charge/voltage can be maintained on the accelerator's sphere using that device will be more involved then I suspected.

Again, thank you for both the insights and clarifications - they save me a great deal of trouble and prevent me from wasting a lot of time.
Bruce Meagher
Posts: 148
Joined: Thu May 12, 2011 11:25 pm
Real name: Bruce Meagher
Location: San Diego

Re: Deuterium Accelerator and Electro-Static Power Supply

Post by Bruce Meagher »

Dennis,

Nice to see that you’re continuing your electrostatic accelerator efforts. As Finn mentioned you want to get as much charge on the belt as possible. Belt material (with high surface resistivity), width, speed and comb design will determine how much current your VdG will get to the terminal. Unlike a fusor you’re not going to draw a ton of current. You have your ion source, equipotential resistor divider network, and corona losses. I think of it like a link budget, and as long as your VdG can supply more current than your budget (for a particular voltage) you’ll be in good shape. One key element to continuous stable operation is pulling off extra charge from the terminal. On the pelletron there are adjustable corona needles used to set the voltage and keep the terminal's voltage from breaking down (see attached pics).

You might consider grounding the terminal through a current meter and focusing on improving your charging rate.

Also attached is a paper from the web on a particular Van da Graaff accelerator design from the '70s. It has some useful info if you haven’t seen it already.

Finally, there's a sweet Van de Graaff accelerator for sale right now on eBay! At $10k it might be out of some people’s budget, but it’s a 400KV unit from High Voltage Engineering (the company founded by Robert Van Da Graaff in the late ‘40s). It’d be an excellent setup for high energy physics experimentation. https://www.ebay.com/itm/Small-Van-de-G ... SwjqVZL24Q

Bruce
Attachments
Corona probe on the pelletron
Corona probe on the pelletron
Inside an empty pelletron chamber with corona probe extended
Inside an empty pelletron chamber with corona probe extended
05_chapter 2.pdf
(1.11 MiB) Downloaded 1164 times
Last edited by Bruce Meagher on Wed May 20, 2020 8:39 pm, edited 2 times in total.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Deuterium Accelerator and Electro-Static Power Supply

Post by Dennis P Brown »

First off, I appreciate any comments/help because this project has been waiting for a few years now (obviously from my complaining), only because the Van de Graaff is the issue with my accelerator project (but really, all such projects based on either a VdG or voltage multiplier - people have misrepresented their results in past articles on-line (not a reference for any here.)) Back to the matter at hand, as you mention, current generated vs. the losses in a VdG are the driving concern. That will occupy me and determine success or failure.

Also, by the way, thanks for your post relative to this topic!

The VdG unit you posted is a bit outside my budget constraints (lol.)

I am determined to try and get this accelerator working (if possible.) Yes, to all your points - I have a 0 - 2 and 0-100 micro-amp meters (both analog so ok with the voltage) that I am using to determine output in a manner you suggest. While I solved one serious problem that I had overlooked in the past (and this issue has never really been addressed in the literature, for that matter) and that was the spray supply frequency - that matters a good bit as I discovered in real world testing.

By the way, any source for belt materials that you have experience with or knowledge of would be useful.

I am currently rebuilding my monster Van de Graaf to accommodate the new belt (the new material arrived) - also, I increased the system's drive speed via a larger drive sprocket but now I am having issues with stalling the motor; to address that issue, I need a belt with a good bit less drag (now have) but my motor mounts do not adjust enough to accommodate the slight size change so, back to redesign/rebuild in order to deal with that new issue.

I also have a small VdG and that now is out-of-mothballs, and on my 'to-do' assembly line (currently, that list has my neutron scintillation detector which is currently priority #1 - that scintillation system's wiring and machine work have been a significantly easier but also bigger pain than I thought but is now done. On the plus side, I have discovered a good bit about problems/repairs relative to my entire HV supply/counter/pre-amp systems used for all my neutron detectors (the pain part). Never gets easier ... .)

Having the small VdG should enable me to do some experiments on using this device due to these will be a lot cheaper, faster and easier builds, while I continue (as time windows open) the mod's on the giant VdG.

Mean-while, I have run my fusor and conditioned it - it, at least (unlike my He-3 detector project), is working very well ...but wondering what shoe will fall next, equipment wise.
Post Reply

Return to “Ion Gun Design and Construction (& FAQs)”