Linear Accelerator Project Update

For the design and construction details of ion guns, necessary for more advanced designs and lower vacuums.
Post Reply
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Linear Accelerator Project Update

Post by Dennis P Brown »

This post concerns (mainly) the deuteron gun for my accelerator project. The project is really just a glorified ion gun so posting here seems appropriate.
The system uses a Van de Graaff to create the electric field that will accelerate the deuterium ions. The main part of the post concerns this gun.

While the Van de Graaff I have modified isn't exactly as reliable as I expected, I have got it to run as high as 250 kV (rough guess); such being the case, I decided to finish and test the main deuteron gun for the accelerator. I am enclosing a picture of the power supply that is contained in a Steel Dome that will be charged by the VdG to accelerate the deuterons (Image 3).

The power supply system for the gun is straightforward (again, see image 3): a 12 volt battery (the large black object in the upper part of the dome) is mounted in a case that supported by the accelerator tube. A potentiometer (that can be adjusted using an external plastic rod) is used to control the volt/current that is fed to the first step up voltage/converter system. This power source converts the 12 volt DC into 3,000 volt AC. In the picture, that is the small silver box below the battery and in front of the white bellows object. Also, this converter has a "High Voltage" sticker on it (the case can "float" at the 3 KV voltage.) This power supply then feeds my home made voltage multiplier that raises this AC 3 kV output up to at least negative 25 kV. This voltage multiplier is the larger "white" bellows system just under the battery/case (the voltage multiplier is filled with oil so I mounted it within a plastic bellows case. This makes filling easy and provides for expansion of the oil with temp. The oil filling was done due to the very tight constraints on spacing within the dome. Also, the dome forms the "common ground" for the electrical supply and as can be seen, the battery takes up a lot of room.)

This high voltage ionizes the deuterium gas within the gun. I've posted that gun design before but it is a modified "Penning" system (Image 1 for the gun by-itself on the end of the glass section of the accelerator tube. No magnet is in this image either). A large permanent magnet is used to enhance the ionization of the gas between the input needle capillary tube (ground and source of the deuterium gas) and the dome exit tube (cathode) that is charged to negative 25 kV.

The gun appears to work well - at 75 microns I get a nice plasma which changes color when I add deuterium gas (the accelerator is currently being tested after some work done on the system and is already at 5 x 10^ -6 torr. This just required one hour of pumping even though the tube had been vented to atmosphere just before pump down. So, leakage isn't, so far, an issue.)

Once the VdG operates in a manner that is more reliable, I will attempt to create a deuteron beam and accelerate it to a glass target (I have a system that allows a glass cover slide to be inserted into the beam path without affecting the vacuum. The glass slide will fluoresce if struck by the deuterons allowing me to both see the beam and check focusing ability of the lens). The accelerator tube/Einzel lens system also uses my new "air" resistor system but I will provide another post (and pictures in the appropriate forum section) on that subject separately. The accelerator resistor system and Einzel lens collector system were both problems considering the requirements - both accelerate the beam and focus it. Still, while I had to try a number of designs, this current one appears to create a far more more uniform electric field along the accelerator tube and also provides good charging to the Einzel lens. At least in tests with the VdG and a mock-up accelerator tube. The measured field was very uniform over about a 1/3 of the tube length (about 25 cm) and I was still getting good charging of an Einzel lens 50 cm away.

Also, I am including a picture of the complete accelerator system (Image 2) to give an idea of the location of the accelerator tube dome relative to the rest of the device - hope this doesn't make the posting too large.
Attachments
Image 1: Penning Gun without magnet
Image 1: Penning Gun without magnet
Image 2: Accelerator with part of the shielding system blocking most of the accelerator tube) and support vacuum system (all bench mounted)
Image 2: Accelerator with part of the shielding system blocking most of the accelerator tube) and support vacuum system (all bench mounted)
Image 3: Power system for deuteron gun system within half the steel dome assembly
Image 3: Power system for deuteron gun system within half the steel dome assembly
Last edited by Dennis P Brown on Tue Jul 22, 2014 9:44 am, edited 2 times in total.
User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 7:05 am
Real name:

Re: Linear Accelerator Project Update

Post by Chris Bradley »

What ratios of deuteron species are you thinking you'll get? Looks like it could be a recipe for 'deuteron soup'?
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Linear Accelerator Project Update

Post by George Schmermund »

Dennis - It looks as though your making some headway! I have a few question about your rig.

How are you arriving at the approximate HV numbers for your VdG. I realize that you suggested the numbers are guesses, but they must be based on something.

Is the ion source demountable? Your images don't indicate how it connects to the accelerator tube.

If the ion source is demountable, have you substituted it with an ion gauge for the purpose of finding out what the real pressure is at the source end of the tube?

What level of beam currents are you aiming for?

Do you know what the total charging current is when the VdG is working properly?

You mention that you're measuring the field magnitude and its uniformity along the tube. How are you making these measurements?

Sorry for the questions if you've already answered them elsewhere. Sometimes I fall behind the curve......
Anything obvious in high vacuum is probably wrong.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Linear Accelerator Project Update

Post by Dennis P Brown »

Mr. Bradley - if I get what you mean by deuteron ratio, then until I measure the current on target, the total number of deuterons accelerated will not be known; of course, only the ionized deuterium will be accelerated so that can be the only species impacting the target that will produce a measurable current for the deuterium gas - non-ionized gas of any species will not create a current. My best guess will be to get 10^5 to 10^7 flux as a probable range but that depend on the success of the Einzel lens and electric field uni formality - neither can be predicted. As for the extremely small trace gases of nitrogen and oxygen at 10^-6 torr, I'd think it would be safe to ignore those tiny fractions since the feed pressure of the deuterium gas will be greater and these far more massive ions will have trouble remaining within the beam line due to both the weak focusing of the lens and I'd think space charge effects by these massive ions (note I qualified this statement so please try and keep it civil.)

You generally have very good input/knowledge so if you have some ideas on this subject I'd very much appreciate them posted here for both mine and others benefit.

As for loses of the deuterium ions due to collisions, absorption, recombination, etc that will be unknown since I will not measure the true input deuterium flow volume so an output current will be the only measurement of the deuterons on target. I'd answer the second question but I am unsure of what the reference to deuteron soup means? Again, maybe the issue of the N2/O2 ions?
Last edited by Dennis P Brown on Tue Jul 22, 2014 9:23 am, edited 6 times in total.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Linear Accelerator Project Update

Post by Dennis P Brown »

George Schmermund wrote: How are you arriving at the approximate HV numbers for your VdG. I realize that you suggested the numbers are guesses, but they must be based on something..
I measured the arc length of the normal spark between two polished globes (one is my Van de Graaff globe fully charged) - I assume about 25 kV/cm; this isn't a reliable number but gives the best guess.
George Schmermund wrote:Is the ion source demountable? Your images don't indicate how it connects to the accelerator tube.
Aside: I included a MS Power Point slide (for those with power point who can view these types of files) that identifies the various components.

My accelerator system uses two globes very much like that found in the 1959 Scientific American article - see image 2 of the pictures provided in the original post above. Of these two globes, one globe contains my gun/supply and is directly attached to my accelerator tube and the other is my Van de Graaff (VdG) unit - again, see image 2. The globe containing the Penning gun and HV supplies I call the "Charging" Globe simply since that is its only function - to be charged by the VdG.

So this "charging globe" is directly attached to the end of my accelerator tube. In all pictures only "half" the globe is shown due to the need to display the gun system contained within. In the pictures shown in the original post (Image 3 above) the accelerator tube can just to the left of the "half" globe holding the gun/power supplies; a small section of this accelerator tube can be seen and it has has copper rings along its length. Note: the actual penning gun isn't visible in any of these pictures except for image 1 but that is only of the gun by itself - that gun is contained in the "charging" globe behind all the electrical/battery components in image 3 above. The battery case sits on a machined teflon support block that is attached to the end of the accelerator tube. The gun's battery/high voltage system components within the globe hide the gun and accelerator tube end.)

I download two older pictures of the accelerator tube for this post - the first picture shows the inside of the charging globe and a small piece of the accelerator tube to the upper left. The second picture shows mostly the accelerator tube (no resistor rings) and a small piece of the globe that contains the gun system is just to the right side of the image.

My source is fully contained within the charging "globe" except for the gas fed line which runs from the globe to my needle valve located some distance away.

The deuterium gas is fed into the globe by a teflon tube somewhat visible in the center of the globe in the picture - the gas is ionized by a simple Penning gun design within the accelerator tube mostly located within the globe (the end of the cathode tube does reach out of the charging globe;) the various high voltage sources/battery are also within the globe and displayed in the picture. I have tested issues of shorting from the teflon tube but below 10^-4 torr, I see no effect.
George Schmermund wrote: If the ion source is demountable, have you substituted it with an ion gauge for the purpose of finding out what the real pressure is at the source end of the tube?
My pressure gauge sensor is located at the opposite end of the accelerator directly connected to the tube by a very short length adapter. I have only measured the pressure with and without admitted deuterium gas (base pressure 5x10^-6 torr and after deuterium flow, 7-9 x10^-6 torr.) Never with ions accelerated, however. I realize that once the ions are being accelerated, there may or may not be an issue of gradient effect but that isn't possible to address using a Van de Graaf system.
George Schmermund wrote:What level of beam currents are you aiming for?


Similar devices have routinely achieved 10^5 - 10^7 deuterons/sec; whether I get those levels will be problematic - especially relative to the VdG performance which I cannot yet control consistently.
George Schmermund wrote: Do you know what the total charging current is when the VdG is working properly?
I have indeed tried to measure the charge by using a micro-ampere meter to give me a current to calculate that value but all attempts, even when the VdG is working (tested when the VdG produces massive sparks) have been nil. Either I am doing it wrong or the system cannot charge when the meter is connected (analog meters are all that I am willing to risk.) I always connect the meter before charging but that may be the issue - I realize that to use an analog gauge connected to a spark output wouldn't measurement anything. As for using a digital meter at all I will not risk that meter with those voltages - maybe a scope and some giga-ohm resistors on a shielded rod would capture the current/time image but I gave away my oscilloscope a few months ago.
George Schmermund wrote: You mention that you're measuring the field magnitude and its uniformity along the tube. How are you making these measurements?
I measure the voltage field with a 1Giga ohm resistor/HV probe - if the voltage stays constant over the measured range, I assume the field is constant. I connected a plastic tube to my VdG. I charged the VdG and I did a "null" test (no resistor ring. s) Using the probe I measured the "potential" along the tube and the values dropped off extremely rapidly - the meter was reading zero withing two or three cm from the edge of the VdG along the test tube. With my "air" resistors ring system installed along the test tube I was getting a steady reading (10-15 kV) along a 10 cm range down the tube. I attributed this result to the electric field being more uniform to the resistor rings installed along the tube.

As for checking the charging value of the Einzel lens that was primitive - I used a grounded wire to check for a spark from the lens; if one occurred, the lens was charged - these varied from massive (rings nearest the VdG) to just detectable along the test tube's length.
George Schmermund wrote: Sorry for the questions if you've already answered them elsewhere. Sometimes I fall behind the curve......
No problem for me and I hope this partly addresses your questions. I would ask that you please do not use any sarcasm in addressing my answers (your previous usage was not appreciated. However, I very much want your project to succeed and will attempt to answer your questions as best I can.)

In that regard, I am doing the best I can with the very limited budget & primitive equipment I have available. As such, rough measurements using any means available is the best I can currently provide (at least until I get a measurable beam current) - it is a given that I cannot provide scientifically acceptable data that could be peer reviewed.

A few words about possible first light - that solely depends on the VdG performance (all other systems are working well - vacuum (10^-6 torr), gun & its HV supplies, gas flow/feed system, support electronics and critical safety systems (shielding, x-ray detectors, anti-electron back flow suppression system, and my two diagnostic devices: one to measure beam current and the other to image the beam on a glass cover slide) - when the VdG does operate consistently (I assume humidity is killing it most the time) I intend to run the device at first opportunity.
Attachments
Accelerator.pptx
Image with idenifications
(128.76 KiB) Downloaded 434 times
Deuteron Gun system within the accelerator's "charging" globe/sphere (center of image.) Just up and to the left of center is a portion of the accelerator tube
Deuteron Gun system within the accelerator's "charging" globe/sphere (center of image.) Just up and to the left of center is a portion of the accelerator tube
Accelerator Tube (center with aluminum Einzel lens) with a small section of the globe/gun system just within the picture (to the extreme right)
Accelerator Tube (center with aluminum Einzel lens) with a small section of the globe/gun system just within the picture (to the extreme right)
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Linear Accelerator Project Update

Post by Richard Hull »

Dennis you certainly have poured your heart into this effort and that is what it takes to be successful. I have not commented as this is an area for me where it is better to look, listen and learn. All the best on getting your system up and running.

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
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 7:05 am
Real name:

Re: Linear Accelerator Project Update

Post by Chris Bradley »

Dennis P Brown wrote:Mr. Bradley - if I get what you mean by deuteron ratio, then until I measure the current on target, the total number of deuterons accelerated will not be known; of course, only the ionized deuterium will be accelerated
D+, D2+, D3+? If you find you have made a D3+ rich ion source, then you may find the bulk of your ion have a lower velocity than the acceleration potential might suggest, and the collision energies will be different.

It's a black-art to me which ion species are dominant in which ion source designs. I think Doug C. went into this in a big way, and ended up with his magnetron source to try to avoid D2+/D3+.

I guess it'll be as easy just to try it and see what happens. Good luck!
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Linear Accelerator Project Update

Post by George Schmermund »

Dennis - For this project to work properly (if at all) you're going to have to establish a 'current budget'. The overall primary thing is to know what the total charging current of the VdG is. The easiest way to do this is to use a micro-ammeter between ground and the top comb. there is no need to have the HV terminal in place as this test is a simple short to ground. This current will be your 'budget'. The test will also allow you to tweak the placement of the combs for highest charge transfer to and from the belt without building up HV. Since this test is done without the HV terminal in place you can safely view the action up close in a darkened room. You'll probably be surprised at the light show!

When you've established the maximum current that can be reliably generated, you can then calculate if the charging rate will handle the burden of the tube's equipotential voltage divider string, corona voltage stabilizer, omnipresent extraneous corona leakage, etc.. Anything left over is all that you'll have to make a beam with. In the end you need to know that the charging current is really available at the voltage you plan to accelerate the beam with. The higher the voltage gets, the higher the burden gets.

Without knowing what these numbers are, to even a first approximation, I don't think you'll have much success moving forward.
Anything obvious in high vacuum is probably wrong.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Linear Accelerator Project Update

Post by Dennis P Brown »

Mr. Bradley - your educated guess's are far better than mine - but there is little chance my system can produce an ion stream of usable deuterons that is either mostly deuterons nor remotely uniform - I'll be lucky to get a detectable beam at all without all types of adjustments, repairs, and other wastes of time/money/efforts before (if at all) a deuteron beam of sufficient intensity is achieved that my detection systems will detect/display a real current. Until then, my ability to measure these parameters and properly tune the device will be difficult to near unworkable.

Until this post of yours, I did not consider now many more ion species would most certainly be produced - and now the issue of the Oxygen & nitrogen ions concerns me. Unlike deuterium, these will have more electrons to lose (creating very negative ions) and the space charge will be an issue for my Lens (Collection of charges on the Einzel lens is more an issue than I considered since these ions will tend to collect at the lens and remove electrons creating net positive charging effects maybe disrupting the lens performance) ; until your question, I did not consider or realize the magnitude of these ions relative to the deuterium.

This really gets to the need for flushing the system well with deuterium gas (under high vac, of course) to get as much O2/N2 out as possible.

Thanks for both the question/comments (again, very useful input) and the wish for luck! I'll need it along with inputs from the people here.
Last edited by Dennis P Brown on Wed Jul 23, 2014 6:21 am, 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: Linear Accelerator Project Update

Post by Dennis P Brown »

Mr. Schmermund - some of the issues you raise have been a concern for me and some I have not considered (and would have bitten me.)

I will yet again try and measure the Van de Graaff's current; I will attempt again to connect a micro-ampere meter but this time I will use a 100 meg or higher resistor connected to a polished metal sphere (placed in contact with the VdG's sphere) with the meter safely near ground potential to get a proper current measurement - in this manner I should be able to use my digital meter safely. Really wish I still had my oscilloscope.

I cannot tell you how difficult dealing with the comb design and placement of these critical components has been while working nearly blind - proper feedback is critical. Simply using arc length voltage has been almost useless since this crude measure has been all over the possible ranges - current (as you correctly point out is critical for any chance of Linear Accelerator operation) has been a major issue that I have (mostly) ignored due to my failure to measure this parameter. Now that all other accelerator systems are working extremely well I will place all my efforts into optimizing the VdG - ridiculous that such a seemingly simple off-the-self device becomes the defining issue for this project. As you correctly point out - current (total charge capability of the VdG) is everything for any LA operation (hence the critical concern of losses.) I have been focusing (incorrectly) on the voltage performance alone with only secondary concern for the charge capacity of the VdG and have previously ignored corona issues relative to the accelerator tube (of course, I have just started to deal with that problem in a serious manner and have made some progress.)

With this point in mind, I wonder if I should force the VdG to remain at the lowest possible voltage consistent with useful acceleration potential? It would appear that higher voltages would logically led to higher corona losses by the accelerator tube and VdG globe (all other issues remaining constant.) In that regard, I would think my only really "free" parameter to adjust in this regard is either the addition/removal of corona rings on either/both the VdG/accelerator and the design of said rings (not just ring size but use of corona needles to equalize the voltage.)

Of course, the power supply that provides both the VdG charging and polarity control will impact the VdG overall performance. I use a 60 ma 15 kV NST with a diode to provide polarity control - be lucky if it produces just a few milli-ampere at its real operational voltage of 7- 9 kV. It is rather late to consider alternatives (my 40 kV Glassman supply is, unfortunately, a negative system - great for a fusor but useless for this application!)

Any thoughts/ideas would be appreciated and thanks for the questions and issues you raised! That helps me to both clarify and direct my efforts to get this monster to first light ... . You may very well succeed in getting first light before I get this beast working.
Last edited by Dennis P Brown on Wed Jul 23, 2014 6:54 am, edited 4 times in total.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Linear Accelerator Project Update

Post by Dennis P Brown »

Richard, thank you for both your encouragement and comments! This project has been a constant battle but really a learning experience in more area's of applied technology (and buy cheap on ebay!) than I was ever prepared to address. Really would have been hopeless without the comments, suggestions and ideas from the many extremely talented and smart members of this forum. Really goes to show the valve of the Fusor Forum!
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Linear Accelerator Project Update

Post by George Schmermund »

Dennis - I think that you're missing the point about measuring the charging current on your VdG. Voltage has nothing to do with the measurement. You need to remove the HV terminal so that you can get to the top comb. Leave the terminal off. The object is to short the comb to ground through a microammeter. This can (and probably should) be an analog panel meter. They are relativity expendable. Don't use any resistors. Be cautious with the meters that have a plastic case as they tend to charge up on dry days and make the needle stick. You can check this by just running your finger across the front of the meter. If the needle moves and then sticks, breath on the plastic and the needle should return to zero when not connected up.

I'll assume that your 'off the shelf' VdG was designed to charge itself triboelectrically. If it still has the same belt and roller system you can start your current measurements with the original setup. Don't use any external charging. The polarity is inconsequential at this point. You're just trying to get an indication in order to understand the measurement.

The important thing is to remember that you only need to measure the charging current at the top comb. There is no need to try and produce any voltage at this stage. The only resistance between the comb and ground should be the meter movement. Again, making this measurement in a darkened room, without the HV terminal in place, will be very instructive.
Anything obvious in high vacuum is probably wrong.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Linear Accelerator Project Update

Post by Dennis P Brown »

First off, again, thanks for your posts and especially this one! You are absolutely correct and I have approached this aspect of my project totally wrong (I foolishly considered this aspect of the project as just too simple to worry about ...not exactly the best idea no matter the excuses and the best way to fail in any case...hopefully, a lesson I will not repeat any time too soon.)

Hate to say it but I never thought of directly measuring the current from the top comb directly with the hemisphere off! And I might add, this time I do deserve some sarcasm! LOL ;)

As you also correctly point out, the polarity issue is totally irrelevant at this stage. My efforts in that quest has created more issues/problems than I need right now when what is paramount, as you have posted, is an operating VdG. So I will return the device back to its correct/original configuration (something I should have done a while back ... ) and attempt to get things working properly. For something so simple I've really managed to screw it up - might make space cadet after all ...
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Linear Accelerator Project Update

Post by Dennis P Brown »

Converted the previous PowerPoint image to a PDF file. Identified some other components.
Also, I should note that the wiring system allows rapid dis-connect so all these components can be quickly removed for service. The "rat's nest" of large diameter white wires bundled together near the bottom and center of the image are common ground wires for the various HV power supplies. These in turn connect to the Rheostat's common ground point. The lone HV feed-line for the gun is not visible in this image.

A minor point: The high voltage for the gun enters the accelerator tube via a very tiny hole drilled through the side of the glass vacuum tube; this hole was then completely sealed with epoxy using minor external pressure to force the epoxy to spread away from the hole creating a wider barrier around that hole. Then while the epoxy was drying, a ultra thin needled was pressed into the epoxy to make direct connect with the metal tube that forms the end of the gun. This created a clear hole to access the gun's HV tube inside the accelerator's tube - even when it is under high vacuum! This "open" channel is not exposed to the vacuum and the metal needle can be removed/replaced as needed.

The ground for the HV deuteron gun is just the gas inlet metal capillary tube - the teflon feed line connects to this gun via a compression fitting.
Attachments
Accelerator.pdf
Accelerator Penning Gun and High Voltage System for the Electro-Static Accelerator Project
(202.57 KiB) Downloaded 464 times
Post Reply

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