DIY Cyclotron

Post links to other interesting fusion or alternate energy sites here.
Post Reply
MarkS
Posts: 250
Joined: Sat Jun 30, 2007 8:07 am
Real name:

DIY Cyclotron

Post by MarkS » Mon Dec 22, 2008 8:50 pm

Interesting article from back in 1947 about four high school students who built a Cyclotron.

http://blog.modernmechanix.com/2006/06/ ... is=XL#qdig

Starfire
Posts: 1482
Joined: Wed Oct 24, 2001 6:14 pm
Real name:

Re: DIY Cyclotron

Post by Starfire » Mon Dec 22, 2008 10:15 pm

Wow - they used a glass 'Mercury' diffusion pump - reminisce of the Amateur Scientist

winterhaven
Posts: 437
Joined: Thu Mar 11, 2004 5:38 am
Real name: Todd Massure

Re: DIY Cyclotron

Post by winterhaven » Tue Dec 23, 2008 2:50 am

Thanks for that.

I was also recently looking at the one in "The Amateur Scientist" that was referred to by one of the people commenting in the article and found it very interesting too. Both seemed to use the "pancake" design for cooling which seems like a good way to go.

Todd Massure

User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 7:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: DIY Cyclotron

Post by Doug Coulter » Wed Dec 24, 2008 3:34 pm

I am also working on a DIY cyclotron. You can see a picture of the magnetic structure at
http://www.coultersmithing.com/fusion/fusion1.html
or in this post. The coil shown in the gap is for NMR field measurement. The coil form contains a well I can fill with water for that purpose.

I find it amazing that few inventors seem to realize just how far permanent magnets have come since the days of "get hundreds of tons of iron and water cooled copper windings" to make a serious field.

Older designs had to have the magnet structure outside the vacuum tank. This meant more cubic area of field had to be made, which in turn made the magnet structure have to be even larger. Kind of a divergent situation.

A new design could have the magnet structure inside the vacuum. Permanent magnets don't create heat either (but may need cooling due to particle losses). If you can put the magnets in the tank, you get a benefit from the above divergent situation instead -- you can make the gap short since you don't have to make room for tank walls inside it. This means you don't need as much coercive force to get the same field in the gap. The one pictured fits through a 6" door on my large tank.

The cyclotron (not yet running) pictured gets over 15k gauss in the 1/4" gap, as measured by NMR.
This is comparable to many of the older designs. I have made the Dees and am getting ready for testing. The Dees are made of titanium, which will hold significant deuterium in solid solution at reasonably low temperatures. This means I don't have to extract the beam at all, I can just let it hit the walls at the outer circumference. I calculate I should be able to get on the order of 250 keV energy with this design. I used custom made NeFeB magnets of the highest grade (eg not real cheap, but cheaper than the other alternatives) to get this far. I will of course have to "shim" the pole pieces by shaping them further on the lathe to get reasonably low beam losses at the outer range of the radius -- but in this case, the beam striking the walls is what I want anyway(!). This design also means that I don't have to have high DC voltages going into the tank via feedthroughs, just some RF at low impedance (ham transmitter output) that will be stepped up by a Pi-type network inside the tank.

Isn't it interesting that with a new approach, the same thing that made the older designs big and unwieldy can instead allow them to be made small and cheap? Divergence in design parameters can go either way!

In short, this should exceed the performance of a borehole neutron activator tube by a wide margin, which itself exceeds the performance in neutrons/watt input of anything so far shown here by quite a bit, if I can believe the numbers posted here and in the Philips book that includes borehole tube design. I will probably have to add liquid cooling to this before it can be run for any duration.

You can get the Phillips data at:
http://www.coultersmithing.com/pdfs/dox.html
where chapter 8 contains the information you may want to look over. Download anything there quick before the copyright cops get me, the site has plenty of bandwidth.

I will of course proudly post when this thing sees first light. Right now I am working on diagnostics so I can tune it in the various ways it will need, and to keep myself relatively safe if and when it starts to work. It has been a job of work to obtain all the bits and pieces needed for this, and thanks to BillF who has helped a lot with that.

I am also working on Farnsworth type designs (cylindrical rather than spherical, and using the math to actually create a focus in the middle, rather than a random grid design) and Dense Plasma Focus, and making small progress on something just about every day. Lots of baby steps yet before the grand announcements!
Attachments
cyclotron.jpg
cyclotron.jpg (90.87 KiB) Viewed 4542 times
Why guess when you can know? Measure!

User avatar
Chris Bradley
Posts: 2931
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: DIY Cyclotron

Post by Chris Bradley » Wed Dec 24, 2008 7:18 pm

You mention N42 on the web link. I am surprised you can get 15,000Gauss with these, I understood their Br max is 13,200. The calculators on http://www.magnetsales.com/Design/Calc_ ... scYoke.asp would suggest about 9,000Gauss in a 0.25" gap (I'm assuming they are ~2"x0.5"?).

One thing I would certainly like to know - how did you mount these on this yoke? I bought a couple of 65mm x 30mm x N45H (120degC rating) 'discs', but thought better of yoke-mounting them partly because of the skull-and-crossbones on the carton! I wrapped them back up and thought I'd take my time over working out the logistics of mounting these magnets, with a rated lifting capacity of 400kg combined. I figured I'd think long and hard over getting my hands anywhere near something loose with that much force - I'll be doing some experiments with ferrites first (including how to mount them safely).

Have you fixed these onto the yoke, or are you relying on them to hold themselves onto the yoke?

You mention p+7Li at 250keV. Not sure if that is going to fly at that 1millibarn fusion cross-section for that energy level (will you get to 250keV with this size of device?) but I did notice some interesting arithmetic related solely to the p+7Li reaction which I'll post separately in due course, thought it is related to 3MeV energy levels where this reaction peaks.

best regards,

Chris MB.

User avatar
Carl Willis
Posts: 2841
Joined: Thu Jul 26, 2001 11:33 pm
Real name: Carl Willis
Location: Albuquerque, New Mexico, USA
Contact:

Re: DIY Cyclotron

Post by Carl Willis » Wed Dec 24, 2008 9:40 pm

Hi Doug,

That's a nice little core. Am I correct in understanding that you will use permanent magnets with it? Is that them visible near the poles in the photo?

Tim Koeth and some friends at Rutgers built a ~1 MeV proton cyclotron a few years ago. His website is here: http://www.physics.rutgers.edu/cyclotron/ This is probably the premier example of a functional "amateur" cyclotron that I know of. I'll be visiting Tim and his toy in a few days for New Year's and it should be fun. I'm hoping he might allow the placement of a bit of Teflon in there to check for F-19(p,ag) reactions.

Good luck with your cyclotron! Should be interesting to see how it works.

-Carl
Carl Willis
http://carlwillis.wordpress.com/
TEL: +1-505-412-3277

User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 7:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: DIY Cyclotron

Post by Doug Coulter » Thu Dec 25, 2008 12:42 am

Thanks Chris and Carl. Those are some very nice links!

Yes, the magnets are at the top and bottom of the yoke, and the preliminary pole pieces extend from them to the middle. This was designed to fit through the 6" hole in the doors of my tanks. It looks to me like I may have to make a slightly bigger version to get the field strength/energy I want, but this one should work, just not at as high an energy as I'd like. Should be enough to learn on, though.

As we used to say in engineering "version 2 is the first one that works right". So the intent with version 1 is simply to learn enough to make version 2 work well.

What I intend to do is test with D and allow (as if I could prevent) the high energy ions to hit the walls, which will also contain some D, being made of Ti, which holds D well below a couple hundred degrees C. This simplifies the design quite a lot -- no beam extraction...At first I plan to do very short (~ 1 second) runs to avoid the need for cooling and plumbing. I am hoping I will be able to discover the ion resonance frequency at low power by the fact that it will change the RF loading as my attempts to measure the field via NMR weren't very accurate.

Here is a picture (lousy focus) of the Dees I made here in the shop, all from Ti. I bored a hole in a Ti plate, then spot welded thin Ti sheet to both sides, then sawed it in half. I plan to support and insulate them with some thin alumina plate stock I have.

At worst this should work as well as a neutron borehole tube, but without the need to have real serious HV in the room, and going through feedthroughs, just RF. I plan to feed in the RF at about 50 ohms, then use a balanced Pi type network to get the high Dee voltages inside the tank. I've managed to acquire some large linear amplifiers that should handle the frequencies of interest, and have built a programmable VCO to run them.

Anyone interested in early designs for neutron borehole activators can check
http://www.coultersmithing.com/pdfs/dox.html
and use the link for the Phillips book, chapter 8, which is the best info I have. From that it looks like I might have made the Dees out of something else and evap'd some Ti on them, but Ti alone should work if there's plenty of deuterium around to compensate the fact that the D will kind of soak in deeply. Or so I hope. If not, well that's one more thing I'll know instead of having to guess from incomplete or poorly documented scientific papers.
Attachments
dees.jpg
dees.jpg (26.43 KiB) Viewed 4542 times
Why guess when you can know? Measure!

User avatar
Chris Bradley
Posts: 2931
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: DIY Cyclotron

Post by Chris Bradley » Mon Dec 29, 2008 2:55 pm

I've just come across, and am reading through (will finish reading it later), a nice BSc thesis submission on small cyclotrons, which includes some salient details of the El Ceritto cyclotron mentioned here, along with several others including much detail of the author's own at Houghton (which is aiming at DD fusion, beam-target[embedded] with it):

http://www.houghton.edu/academics/progr ... sFinal.pdf

and also a comment that these smaller DIY cyclotrons just use one dee, which I wasn't aware of (the chamber wall/'rest of universe' ground being the opposite end of the RF drive - not sure about the RFI emissions from such an arrangement, though, if the grounding isn't near perfect?!!).

best regards,

Chris MB.

User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 7:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: DIY Cyclotron

Post by Doug Coulter » Tue Dec 30, 2008 4:03 am

That's a nice link, Chris. It explains more clearly than most of the papers I have, in plain units too.
The Rutgers site Carl gave us a link to is also very good.

I myself do not plan to do the single ended D with a dummy, but we'll see what works. From the Rugters data, it seems not to have helped them since is wasn't as symmetric as it could have been.
Shouldn't be a real killer to make equal amounts of both RF phases. I just ordered a magnetometer so I can pre-tune the RF stuff to the right frequencies for my magnets. Easier than making it all broadband.

The best ref I have on this is "Introductory Nuclear Physics" by David Halliday, that I picked up in a used bookstore. Hard to find, but way worth it. It's from the late 50's. What's fun about it is that of course things like quarks and color weren't known, and this book is very honest about what was known and what was merely speculation. Makes one trust the rest a lot more. In any case, they gave real working parameters for a few cyclotrons, so I could check my design and make sure I was using the correct units throughout (no factor of Pi, or C or meters vs cm or something missed).
Why guess when you can know? Measure!

Post Reply