Software for drift tube accellerator design

For PC type software that runs under some PC opsys.

Software for drift tube accellerator design

Postby Doug Coulter » Tue Oct 11, 2011 3:40 pm

While watching markets (ugh) today, I wrote up a little perl script to calculate the lengths of the tubes for a drift tube accelerator. This is a little rough, and makes some assumptions, but gets to the right (hand checked, ugh) numbers. It is assumed that you are putting in ions at some DC injection voltage, which you can specify; the program has defaults I've set for about what I think I want for the first go. It's assumed that the first tube, the output of the ion source, is at ground, followed by some even number of tubes driven out of phase with RF, followed by a final grounded tube. The voltage after the first gap is thus ion source volts + RF peak volts, the next gap gains 2 times RF peak volts (one tube at + peak while the next is at - peak volts), and so forth till the last gap only gets one more RF peak volts as it's ground, not the opposite phase. The length of the first tube is really the length from it's front face to the middle of the gap to the next tube, and after that the lengths are middle-gap to middle gap (the actual metal tubes will therefore be the number output minus one gap length, which will be short for all but the first, focus, gap. So, it's a little rough, what I like to call research grade code -- because the point of writing it is to get the answer, once, not for it to be pretty and long maintained "production" code which hews to a higher standard (unless you work for Adobe or Microsoft).

Here's what a run looks like for the defaults, and assuming I'm accelerating D.
drift.png
Screen shot of run


And here's the code itself.
drift.zip
perl script for drift tube design
(2.43 KiB) Downloaded 340 times


So, I have to munge the gap corrections to the lengths by hand (and maybe I'll be virutously lazy and do the conversion to good English (oops, now only American) inches throughout to make that a little easier, but this is the basics. I tried to use the latest numbers from Terman for the factor, and from Wikipedia for the masses, and I subtracted one electron mass from the atomic weight of the ion involved.

Drift tube accelerators aren't very flexible. They act like a mass spectrometer, in that only ions of the right charge/mass ratio for the other settings will be accelerated, and they get somewhat pickier about that the more tubes there are, though there is some "bunching" effect that allows them to work a couple percent "off" at reduced performance. You could "select" for a different ion mass (actually charge mass ratio) in one by changing the frequencies and voltages up to a point, just like you can accelerate more than one type ion in a cyclotron if the ratios can be made to work out right, but that's not going to work well over any large range. And that will be another program -- given a particular accelerator mechanical design, show the DC and RF volts and frequency for some other ion type...because in this one I want to be able to do all of H, D, T, and 3He, since I have all of those, and if I can, I'd like to be able to use the same RF frequency for all - which means fairly major changes to the other numbers will be required. But you have to have this first, so here it is.
Since most of the focusing action will happen at the first gap (and before in the ion source) the inter-tube gaps from then on might as well be pretty small - just below arc-over spacings, since so little percentage of energy is added at each following gap as a function of the energy entering each later gap. Like a linac, most of the focus is right at the beginning, and the focus effects of following lenses can be pretty much ignored -- barely keep up with coulomb beam expansion from self-repulsion at any decent beam current.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Return to PC

Who is online

Users browsing this forum: No registered users and 1 guest