Micro-cyclotron progress

[George Schmermund]

The micro-cyclotron (uC) has had slow progress up until now do to the steep uphill climb necessary for learning something about RF circuits. I've spent most of my life mainly with mechanical/acoustical/chemical/nuclear devices and related systems that function in the DC to 1 MHz realm. Some of the NIM and detector stuff is faster, but relatively predictable because it mostly uses hardware that has already been designed and tested for 'plug and play'. Learning a new discipline at my age truly taxes the few sober hours I can set aside each day for the purpose of burning in new neuronal pathways.

Anyhow, I'm now developing an intimate relationship with my HP Q-meter and RF vector impedance meter. Over the years I've become a devote believer in the philosophy of "never calculate what you can measure". This imprint has never served me as well it does now. RF stuff is full of non-intuitive pitfalls. I also believe in keeping a good assortment of calibration standards within arms reach as a means of maintaining sanity.

The images show the progress to this point in the uC project. They represent the second generation of design attempts. The ''circuit" was drawn by hand onto the double-sided PCB. The board was then covered with a layer of mylar packaging tape. An Exacto knife was used to cut the pattern. The tape mask was removed where the copper wasn't wanted and then the board was etched with mix of HCL/H2O2. The etching mix works very well and has the added benefit of being clear. It's interesting to watch the progress.

The fabrication part of the uC to this point has been so easy that I almost feel like I'm cheating. (Magnet assembly nightmare excluded!) I guess I'll think otherwise if I can't get it to work. So far this is still a back patio construction project needing no special tools.

[Todd Massure]

Looking good George. I'd like to see the whole set up in person some time.

[John Futter]

George
You oviously know approximately what your magnets strength is and what you have for HT.
What is your expected RF drive frequency??
If its in the 1-30MHz range then a ham rig will do well as the RF exciter if not then there are other options that I might be able to point you to, I haven't spent 30 years in the RF game for nothing.

All looks like eye candy @ the moment
keep up the good work

[Starfire]

Eye-candy indeed - I congratulate you on fine work - RF work is almost a blackart and UHF layout almost mechanical rather than electrical - good luck George and thanks for sharing.

[George Schmermund]

Todd - When this project is complete it should fit into a lunch box. I just hope you don't want to get together for a show and tell at the airport.

John F - The magnets, when in the yoke, measure to be .94T at the center and tapers off slightly at the edge. I was concerned about the field gradient at first, but it now seems that it's a plus because the beam spiral spacing will widen as the radius extends. This effect should make it easier to intercept just a single beam of highest energy at the collector.

I used this calculator for a ballpark estimate of aim point frequencies. http://www.vk2zay.net/calculators/cyclotron.php . I'm using signal generators for the basic testing. When it's show time I'll use my Kenwood TS 930S transceiver and AT 230 antenna tuner for the real test drive. My power requirements are minimal because all I want at first is a measurable beam. I've got some Keithley stuff for that test.

John H - You're right about the witchcraft part!

[Todd Massure]

Handy frequency calculator, thanks.

[Carl Willis]

Nice progress George.

I'll be interested especially in seeing how you seal this puppy up and what you install for an ion source.

Any idea what the low-power Q of your dee circuit is right now? I think the PCB material near the dee gap is going to be significantly lossy and is going to be a challenging outgassing, arcing, and heating issue. You're certainly pushing the envelope (which will make it all that much more impressive if you can get the thing to work)!

-Carl

[John Futter]

George

fine I didn't realise you were a ham and I'm sure the TS940 can be made to transmit anywhere between 1.6 and 30MHz by cutting a diode inside
thats what I have done to my TS440S
73's ZL2TUD

Edit
G10/ FR4 properties are fine up to about 200MHz I have used printed caps in output filters at the kilowatt level for FM broadcast xmitters that run 24/7 with no appreciable heating of the printed caps. Printed caps at 400-800MHz television service are limited to about ten watts so are OK for driver stages not output stages. the printed output filter caps for TV use I change to Polyflon pcb material which is copper clad virgin Teflon.

[George Schmermund]

John F - I'm not a ham (well, not of the amateur radio sort). I just wanted to keep a nice rig around the lab just in case; for after "the big one!" (bomb, earthquake, frankenstorm, take your pick). Looks like it will be handy before then. Don't worry, the experiment will be well shielded for RF leakage. Besides, it may end up in an industrial band anyway.

[tligon]

So many projects, so little time.

I think I NEED one of those!

I wonder how small you can go. Buddy of mine used to think you could make cyclotrons out of molecules such as ATP.

[George Schmermund]

Carl - My plan for sealing the top cover on is the same as the bottom: solder. I'll close the dee gaps with small pieces of glass and epoxy them in. With a small mirror I should be able to see any lightshows that may occur.

The ion source was originally going to be a modified nuvistor triode. Now I realize that it was not as good a choice as a simple biased hot filament. I'll start with some light bulbs of the grain of wheat variety. A separate small test system will allow me to measure emission currents. Cutting the glass bulb away shouldn't be a big problem. This same test system should also allow the design of the world's smallest hot cathode ion gauge. This would be just to keep everything to scale.

The Q of the dee sections and coil ranges from 66 at 12 MHz to 59 at 14 MHz with the top clamped on. I'm aiming for successful operation in that range. The homemade cyclotron at Houghton claims a Q of 22. The one at Rutgers claims 150. (Rutgers calculated a Q of something like 1600. Ha-ha). I saw another reference (that I can't find now) which stated that most cyclotrons operate with Qs of between about 10 and 25. Seems the values are all over the place. No measurements that I've seen so far have claimed to use a Q meter. Poor guys. Well, I'm happy with the numbers I've got for the time being.

I've tested the dees with an insulation breakdown meter and they passed with flying colors - 5 KVDC and 3.5 KVAC at 25 KHz. I don't think I'll be seeing those voltages on the dees because I'm only after a measurable beam of nano or pico amps. A clean and well defined beam plot is all l need to move me on to the new and improved machine that is now festering in my head.

As far as outgassing goes, we'll see. I don't plan on the device overheating. If outgassing really interferes (which I doubt), I can easily evaporate a layer of silicon monoxide over the exposed material in the gap.