Linac Design - Electron Beam Acceleration and Collimation / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Linac Design - Electron Beam Acceleration and Collimation

previous 1 2 3 next

Move Thread

LAN_403

colin heath

Tue Dec 27 2011, 12:37PM

Registered Member #123 Joined: Fri Feb 10 2006, 12:58PM
Location:
Posts: 162

Sounds a goods project.

Basically build a cockcroft walton multiplier then tap each stage of cockcroft to each stage of accelerator electrodes. This gives good accelration at each stage and should if I remember keep the beam fairly well guided to target. Just be careful with something more powerful than van de graff with Xrays and other nasties.

Lead and distance are your friends for Xrays and Boron with water for neutrons depending what you plan to smash

MArked One

Wed Dec 28 2011, 10:08AM

Registered Member #3688 Joined: Mon Feb 14 2011, 07:39PM
Location: Europe
Posts: 38

Greetings.

Thank for the input. I'll make the tube supports, electrodes, etc from aluminum to reduce the x-ray hazard. Perhaps some lead shielding will be in order, too.

What about powering it off of a marx? The peak current should be substantial, but what should be done to the cathode feed to enable pulsed duty?

again, thanks for the input.

colin heath

Thu Dec 29 2011, 09:32PM

Registered Member #123 Joined: Fri Feb 10 2006, 12:58PM
Location:
Posts: 162

hi,

I think the pulsed approach will lead to higher X-rays etc but will get more power. stainless steel is another great material for X-ray reduction.

have you been on fusor.net as they have many experienced members there? i just tend to build things then learn later which not a good idea with high energy stuff lol!

I was proposing a linac with Marx with defib caps but would have to bury it to make output radiation safe so never pursued!

jpsmith123

Fri Dec 30 2011, 06:31AM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

There are quite a few examples of pulsed accelerators around; usually using either a Marx generator or a "Tesla transformer" to directly or indirectly drive an "electron beam diode" (i.e., a single gap accelerator). Here's one where the Marx is driving the cathode of a "vircator".

Since the Marx generator would usually have a fairly low impedance, the cathode will generally need to emit several kiloamps of current...so you'd be using a cathode in an "explosive emission" mode.

Steve Conner

Fri Dec 30 2011, 07:44AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

I always thought a Tesla coil driven accelerator would make an interesting project. You could use a multiple gap tube by connecting the grading electrodes to taps on the secondary coil.

You'd maybe need to modulate your electron or ion source so that it only emitted at the right times.

jpsmith123

Fri Dec 30 2011, 04:42PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

Below are a few examples. If you have access to the journal "The Review of Scientific Instruments", there are lots of good papers there regarding tesla coil driven (and other type) ebeam accelerators, and some of them are relatively simple in design and construction.

Here's a link to someone's thesis involving a fairly spiral generator driven ebeam generator:

MArked One

Mon Jan 02 2012, 10:09AM

Registered Member #3688 Joined: Mon Feb 14 2011, 07:39PM
Location: Europe
Posts: 38

Greetings.

Very interesting, especially the VIG-driven one. I'll go and talk to someone with RSI access and look at the other papers as well.

Thanks for all the help, everyone.

I have started to build the case, and shall (with luck) provide some pics shortly.

Will an acrylic tube serve as a vacuum vessel in terms of vapor pressure (assuming chemical treatment)?

jpsmith123

Tue Jan 03 2012, 05:01AM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

A pulsed electron accelerator would need to be designed as a system.

For example, whatever you use to generate the high voltage pulse, say, a Marx generator, will have a certain output impedance (in this case a relatively low impedance Z = (L/C)^0.5 ohms) that will need to be approximately matched to the "diode" load.

From the papers I've seen, usually the load is adjusted approximately for "critical damping".

Anyway, since this implies high current pulses from hundreds to thousands of amps, this usually means an explosive emission cathode. And the diode impedance is a function of the diameter of the cathode and the spacing between the cathode and anode foil or whatever. (I've seen pulsed accelerators operating at the megavolt level with acceleration gaps that are only a cm long or so, and some even less).

As I see it, with these kinds of short diode gaps, the vacuum in the diode doesn't have to be very high. I could be wrong, but considering that the mean free path of an electron in a gas at 1u pressure is about 29 cm, and if your gap is only 1 cm or so (or even 2 or 3 or 4 cm for that matter), at 1u pressure, your gap is still many times smaller than the mfp. I think you could easily get that kind of medium vacuum with a good two stage pump and a molecular sieve trap).

Also, since a plasma forms on the surface of the cathode, and it slowly expands across the gap and eventually shorts out the diode, the pulse width and gap distance have to be chosen so that the beam pulse has already ended before the so-called "plasma closure" occurs.

(I have a paper somewhere that goes into some detail on these considerations. I'll have to look for it and post it).

Lastly, I think you could use acrylic for the diode insulator, i.e., feed-through and wall. (I know I've seen acrylic belljars and base plates designed to be used for moderate vacuum work). IIRC the vapor pressure of acrylic at room temperature is somewhere between 10^-6 and 10^-7 torr, so that should be ok for your application I would think.

MArked One

Thu Jan 05 2012, 10:13AM

Registered Member #3688 Joined: Mon Feb 14 2011, 07:39PM
Location: Europe
Posts: 38

Greetings.

Oh wow, the requirements for a pulsed accelerator are quite interesting.

Do you know what limits the size of the acceleration gap? Just the impedance?

I have looked at explosive emission cathodes, but have found no designs that I could build. Do you have any suggestions? Perhaps a graphite rod that's heated with a capacitative discharge? Or perhaps I could use a tungsten thermionic cathode in pulsed mode? (And/Or try to contain the resulting electron cloud with an electric field?)

I will try to secure a diffusion pump, so attainable vacuum might higher.

EDIT: I'll try to dig up a paper I read on a proton linac that used an acrylic tube and a VDG. They used a diff. pump, so acrylic should be good for high vacuum if cleaned. (they did use wax as sealant, and I'm using epoxy)

I will try to post some pics if anyone's interested in a few days.

Proud Mary

Thu Jan 05 2012, 11:11AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

MArked One wrote ...

I have looked at explosive emission cathodes, but have found no designs that I could build. Do you have any suggestions? Perhaps a graphite rod that's heated with a capacitative discharge? Or perhaps I could use a tungsten thermionic cathode in pulsed mode? (And/Or try to contain the resulting electron cloud with an electric field?)

Flash X-ray sources most often have a point-and-plane configuration. The cathode is a metal rod tapered down to a needle point to maximize charge density, while the anode is a metal film, foil, or sheet.

This paper describes what must be the simplest flash X-ray source one can imagine:

S. Ahmad, S. Hussain, H. U. Khan, M. Zakaullah,, and A. Waheed Study of Cu-K Emission from Compact Diodes Powered by Capacitor
Discharge CHINESE JOURNAL OF PHYSICS VOL. 43, NO. 1-I FEBRUARY 2005

which you can read in full for free here:

There has been a lot of work in recent years on using carbon nanotubes at the cathode, since these represent a far finer point than any machined metal rod can ever achieve.

previous 1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob