Accelerator Tube Revisited

For the design and construction details of ion guns, necessary for more advanced designs and lower vacuums.
Post Reply
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Accelerator Tube Revisited

Post by George Schmermund »

I've recently been sorting through some old boxes full of started, but never finished projects. One of my past favorites is the accelerator tube from the "Back Porch" series of a couple of years ago: viewtopic.php?f=12&t=5023. Having found the parts and a partially completed tube, I was swooped up once again into the vortex of past fantasies. There never was much interest in the construction method, at least not on the level where anyone tried to utilize the information, so it lost traction and went dormant along with many other madcap projects I've posted here.

Now that I've abandoned projects that exclusively uses hand tools, I'm free to get back into the machine shop and make chips! In an effort to resurrect the accelerator tube idea and proceed on to the making of a small beam on target instrument, I've chosen to replace the polyethylene insulator with Delrin. There may be better choices, but I have a good supply of it and Delrin is a pleasure to machine. The overall scheme is still the same as far as pressing the parts together into as many stages as needed/desired. I'll post some photos as things progress.

Naysayers: Call your Aunt Sally and begin stuffing those straw men!
Attachments
1-P1030826.JPG
Anything obvious in high vacuum is probably wrong.
JakeJHecla
Posts: 91
Joined: Fri Aug 26, 2011 6:19 pm
Real name:

Re: Accelerator Tube Revisited

Post by JakeJHecla »

I'm greatly looking forward to the outcome of this! What's the intended voltage source?
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Accelerator Tube Revisited

Post by Dennis P Brown »

I am very pleased someone else is trying an accelerator project!

So, hello and I really hope your project starts picking up speed.

I get the basic idea but could you maybe give a drawing of the complete accelerator tube? I'd like to see what you exactly have in mind - good ideas are always worth stealing ... I mean copying (lol.) Vacuum sealing is an issue that I'd like to see how you are addressing.

Will this be used as a deuteron injector for a fusor or are you trying for something else? I need a collimated neutron source so I am putting a lot of effort into this project. Curious to see what someone else has in mind ... .

Also, my Van da Graff has crashed due to humidity and an old belt (can get it working well but only at times.) I am considering a high voltage multiplier - 300+ kV range. Are you going to continue to use a VdG? If so, any special mods for the device? For instance, polarity is critical but I am not impressed with current examples used to spray the belt to control the VdG charge - any ideas on this issue?
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Accelerator Tube Revisited

Post by George Schmermund »

This is a concept experiment. There are no drawings. It will work with any constant potential DC supply.

An interference fit should be enough to seal each section. A helium leak test will show if that assumption is true.
Anything obvious in high vacuum is probably wrong.
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Accelerator Tube Revisited

Post by George Schmermund »

I've decided to start the tube project with a few simple tests. A Delrin collar was machined to allow a .003" interference fit with the Cu fitting. Using a compression gauge in the mill's vise, I can see what pressure needs to be applied to get the fitting into the collar. This also allows the measurement of creep over time.

Without the second fitting compressed over the outside of the Delrin collar the creep factor is high and won't maintain the pressure needed to assure a good vacuum seal. All this is quite obvious, of course, but the numbers have to start being collected somewhere if progress is to be measured. I'm confident that the Delrin can be cold extruded into any surface irregularities in the surface of the Cu fittings and will produce a helium tight seal. Time will tell.
Attachments
1-P1030831.JPG
2-P1030832.JPG
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Accelerator Tube Revisited

Post by Dennis P Brown »

Interesting - such details are informative. I, too, am interested in the He leak check results. That will be some tube if it holds vacuum - the design is both extremely cleaver and very clean. The design looks to have a lot of potential. I look forward to both the vacuum and accelerator application testing.

Any ideas on the ion gun design?
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Accelerator Tube Revisited

Post by George Schmermund »

I spent some time on the lathe this afternoon and turned down the Delrin collar to produce the same .003" interference fit for the larger 3/4" end of the fitting. The smaller 1/2" diameter test yesterday allowed the fitting/collar pressure to build up to a constant 100 pounds during the insertion. It quickly decayed to about 40 pounds when the insertion movement stopped. Pressing the two fittings together today required a constant 400 pounds during insertion. when the motion stopped the pressure dropped to about 340 pounds. we'll see where it settles to over night. If it will maintain at least 100 pounds, I'll do a leak test and proceed from there.

Unlike the agony expressed by others building similar projects I must say that I find this kind of work to be the highest form of entertainment. Well... in a machine shop, that is.
Attachments
2-P1030835.JPG
1-P1030838.JPG
Anything obvious in high vacuum is probably wrong.
Peter Schmelcher
Posts: 228
Joined: Sat Nov 13, 2010 1:56 am
Real name: Peter Schmelcher

Re: Accelerator Tube Revisited

Post by Peter Schmelcher »

George I’m with you. It should be about the journey and not only the destination, however, at times I wish the journey was flatter.

Looks like you have a nice Bridgeport in the shop.

When I started surface grinding on my mill, I cobbled together some way covers. I imagine you will get a smile from the materials, an Ikea roll up blind and 0.002” SS tool wrap. The ways stay oily and clean up is easy.

-Peter
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Accelerator Tube Revisited

Post by George Schmermund »

Peter - Yes, the ways are always covered when I'm machining. They're uncovered now because I'm installing a DRO and need access to the back of the table.

As an aside, I think it's interesting that the utility of having ready access to a lathe and mill is generally overlooked on this forum. When I inquired about this some time back I was informed that 'there are plenty of other forums where the focus is on home machining. They should go there.' This is true, but how are those folks who are new to construction projects even to know that they can buy useable machine tools at places like Harbor Freight for dirt cheap prices? How are they to know to even look? Oh, well, go figure.
Anything obvious in high vacuum is probably wrong.
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Accelerator Tube Revisited

Post by Dennis P Brown »

Machine work can be a more relaxing experience compared to working with the high vacuum or high voltage systems - that is for sure. I just finished some lathe work on parts for my deuteron gun assembly including reworking some KF-fittings used for that application. I have to admit, it was the easiest work I've done to date. I have some minor finish work on some plastic pieces to mount the battery in with the gun and if my high vac flange arrives tomorrow (appears it will) - I'll assemble the completed gun and install it on the accelerator tube along with its high voltage supply (damn thing has bitten me enough - be glad to get it away from me and sealed up in the accelerator's steel globe).

From then on, no more fun machine work ...rather vacuum testing/leak detection, and as needed, sealing. These operations are no fun. Still, getting close to first light.

Can't wait to see your entire tube and the vacuum test results (I have little doubt it will be successful. The work you are going looks very exact.)
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Accelerator Tube Revisited

Post by George Schmermund »

Dennis - Good vacuum joints, whether demountable or permanent, never need 'sealants'. Attention to detail before assembly is all that's necessary. As far as working with high voltage goes, I never approach it without at least a few beers in me. High voltage is too scary to work with when sober!
Anything obvious in high vacuum is probably wrong.
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Accelerator Tube Revisited

Post by Dennis P Brown »

There are real machinist like you - those who can make precision parts that can be pressed for high vacuum fit: that is really impressive (no pun intended.) I really enjoy the photo's and work - that accelerator tube is not just a scientific part but art work, as well. Your's is a skill that I really appreciate since I can't even use a micrometer properly while machining ... .
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Accelerator Tube Revisited

Post by George Schmermund »

Dennis - For the record I want to say that I'm not a machinist by any stretch of the title. In fact, to assume so would be an insult to the trade. I do have some nice machine tools, though. Being a hard core science hacker I've been driven to make my own parts since high school. Having the ability to work with tools is the only way I have for realizing the ideas and projects that are constantly fermenting in the darker corners of my imagination. Also, I think that making things is fun!

For anyone interested in learning how to use basic machine tools I can recommend spending some time on YouTube channels like http://www.youtube.com/user/mrpete222/videos. There are many other channels that offer basic information about using manual machines and hand tools, but Tubalcain/mrpete222 is one of the best. Practice makes perfect!
Anything obvious in high vacuum is probably wrong.
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Accelerator Tube Revisited

Post by George Schmermund »

I finally got around to playing with the tube again. After a few days in the vise setup the pressure stayed at ~ 250 pounds. That was a good sign in the overall scheme of things. I setup a quick leak test on one of my trusty Veeco MS-20 instruments and... well, the tube leaks! Not a lot, but more than I thought it would. There are several simple remedies that can be used to correct the leaking (short of sealants, of course), but I was surprised to find that the Delrin was not as conformal as I thought it would be.

The first remedial step will be to use some heat while the parts are under pressure. This should improve the flow characteristics. The same parts can still be used for the new tests. If that works I'll be able to finally justify making a small injection molder! I always learn more from the experiments that don't work than the ones that do. Many times success keeps you sound asleep.

Well, it's Saturday evening and my gin bottle is beckoning me. I'll report progress when it happens.
Attachments
1-P1030847.JPG
Anything obvious in high vacuum is probably wrong.
Peter Schmelcher
Posts: 228
Joined: Sat Nov 13, 2010 1:56 am
Real name: Peter Schmelcher

Re: Accelerator Tube Revisited

Post by Peter Schmelcher »

I wonder about the rms finish on the copper fitting, I would expect it is not that great. You might dress the surface with a diamond.
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Accelerator Tube Revisited

Post by Dennis P Brown »

Sorry to hear that - I too am chasing a leak on my new gun; luckily it isn't the brass KF flange that is epoxied onto the end of the glass accelerator tube!

A word of warning with press fitting copper and ceramic at elevated temperatures - the coefficient of thermal expansion (CTE) of copper is huge compared to the ceramic. As long as the copper temperature is well below that of the ceramic part, this method will work. Otherwise, the copper may crack the ceramic but in any case, upon cooling, will pull away from the ceramic.

Post edit note: first off, sorry if this is an obvious issue to you but many here without experience may not be aware of this problem when they use ceramics and copper in making parts and have to do a heat treatment. CTE mismatch can cause failure in a composite part if it is heat to sufficient degree.

Mitigating this problem can be done by using lower CTE materials like steel. Aluminium, of course, is even worse than copper.
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Accelerator Tube Revisited

Post by George Schmermund »

After the disappointment of the tube compression experiment I got to thinking about alternative methods of assembly. The amount of machining that would be needed to fix the sealing problem got the lazy side of my mind into gear and I decided to fall back to conventional methods of putting these type of parts together. I haven't abandoned the parts choice that I'm now using because I still think that the project will be successful once the joining method is improved. I spent some time in the shop today making another set of electrodes and an insulator.

The new insulator has about the size and shape of what the final pieces will be like and I'm continuing with the Cu fittings for the time being. Eventually I'll probably machine similar, though custom, electrodes out of aluminum. The Cu fittings are cheap, readily available, and will allow easy attachment of the bleeder resistors by soldering them on.

I've machined a groove into the Delrin at the base of where each Cu fitting seats when pressed into the plastic and will use this groove to hold a small amount of JB Weld. When the parts are compressed together the epoxy will act as a permanent o-ring on each part. I've also added some grooves on the OD of the Delrin to act as corona barriers. Delrin is such a pleasure to machine that I consider it to be the 12L14 of plastics!

The photos show what the new parts look like. As can be seen, the electrode extends 0.100" past the Delrin insulator. This should reduce the tendency of scattered beam charging.

After another leak test I'll be ready to start testing the high voltage breakdown limit.
Attachments
3-P1030856.JPG
2-P1030855.JPG
1-P1030850.JPG
Anything obvious in high vacuum is probably wrong.
User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 7:05 am
Real name:

Re: Accelerator Tube Revisited

Post by Chris Bradley »

Have you considered using a rubber joint? For my 'stub' connections I have cut pieces of pure nitrile sheet into short strips that can just reach all the way around the piece to be fitted when held in the hand and that is slightly thicker than the clearance between the two parts. As you work the parts together, the rubber is squeezed and compressed, and the ends of the rubber will be pressed together. It self-seals a treat. In your case, you could even apply a little surfactant to the rubber to allow ease of assembly, then wash all the residues off.

Example: viewtopic.php?f=10&t=3995&p=25632#p25632
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: Accelerator Tube Revisited

Post by John Futter »

George
I have used Sikaflex MS sealant to hold my accelerator sections together instead of the High Voltage Engineering Ltd use of PVA glue.
I needed more strength than the PVA afforded as I have quite a few pounds of stainless cantelevered off the end of the accel section.
I saw somewhere that NASA uses the sikaflex MS as it has extremely low outgassing once cured
pics at the end of this thread on another site of what I have done to join the glass discs to the Al plates http://www.coultersmithing.com/forums/v ... f=45&t=757
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Accelerator Tube Revisited

Post by George Schmermund »

Thanks for all of the suggestions concerning progress on the tube. I was going to JB Weld the parts together today and let them spend a day in the mill vise. The epoxy is my first choice because it's what I have on hand. Any port in a storm! I decided to rethink the testing schedule and chose to do the HV breakdown test first. There's no need to proceed forward with this prototype if it can't hold off a useful voltage in air. The gap between the two electrodes is only .072" and the radius on the inside electrode is .050". Reflecting on these dimension made me a little nervous.

As can be seen in the photo, the tube passes with ease at the 5KV mark. The test was run for 5 minutes using DC. The temperature was 70 deg. F and the humidity was 48%. The barometric pressure was 1011.2 mb. Having previously spent some time polishing the electrodes where breakdown was most likely, It seems that the time was well invested. I had looked up several sources regarding HV breakdown tests and found most of the information quit contradictory and basically useless. I fell back to my old mantra: "One test is worth 100 expert opinions".

I can now finish joining the parts and then move on to leak testing and the real HV test under vacuum.
Attachments
1-P1030859.JPG
Anything obvious in high vacuum is probably wrong.
Peter Schmelcher
Posts: 228
Joined: Sat Nov 13, 2010 1:56 am
Real name: Peter Schmelcher

Re: Accelerator Tube Revisited

Post by Peter Schmelcher »

George these attachments are on HV insulation design and there worth looking at. I think the work was done for the EU space program.
Attachments
HV vacuum Metis123201.pdf
(909.28 KiB) Downloaded 620 times
HV vacuum Metis123209.pdf
(740.29 KiB) Downloaded 503 times
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Accelerator Tube Revisited

Post by George Schmermund »

Peter - Thanks for the links. Their experiments are quite detailed and very interesting to read. I'm sure I'll be able to use the info in upcoming projects.
Anything obvious in high vacuum is probably wrong.
George Schmermund
Posts: 262
Joined: Tue Aug 15, 2006 8:51 pm
Real name: George Schmermund
Location: Carlsbad, CA

Re: Accelerator Tube Revisited

Post by George Schmermund »

I've finally gotten around to doing another test for breakdown voltage on the accelerator tube electrodes. I was somewhat skeptical about the hold-off ability of the Delrin insulator, but my intended needs aimed only as high as 10 KV/electrode under vacuum. The gap between electrode surfaces is only .072". It was a pleasant surprise to find that the section was good (so far) to 15KV! The aim is for a 10 section tube, so this arrangement should be good for 150 KV with a length of less than 1 foot. The maximum hold-off voltage will be determined in the next round of tests. These tests are being done using a 40KV Spellman HVPS and it has the floating ground option. This option allows the measurement of current far below the metered current on the front panel. Using a Keithley picoammeter, leakage from corona and field emission can be easily evaluated.

Tests are under way to determine the best way to make an equipotential resistor string for the tube. The original plan was going to use 200 Meg resistors when this project first got started, but it will require something closer to 1G ohm/electrode to keep the current drain reasonable for the VdG voltage source.
Last edited by George Schmermund on Fri Jul 04, 2014 2:13 pm, edited 1 time in total.
Anything obvious in high vacuum is probably wrong.
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Accelerator Tube Revisited

Post by Dennis P Brown »

Since you are aiming for 150 kV for your tube, how will you achieve that voltage?
Post Reply

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