Questions about the practical construction of a Linac

For the design and construction details of ion guns, necessary for more advanced designs and lower vacuums.
User avatar
Noah C Hoppis
Posts: 56
Joined: Sat Dec 22, 2012 9:05 pm
Real name: Noah C Hoppis

Questions about the practical construction of a Linac

Post by Noah C Hoppis » Sat Dec 22, 2012 9:43 pm

I have recently become interested in Linacs and I have only found sparse information on the construction of a Linac, so I pose the question, how would one go about constructing a DC Linac on a relatively low budget? I get the concept of a Linac but the actual ingredients of a Linac have eluded me. I've decided on a Linac that is DC driven because I have no idea how to get my hands on a good RF supply. I also want to accelerate protons from striped hydrogen (would that even be practical?). I am not sure how to go about the actual acceleration cavity. Would it have many drift tubes or only an anode & cathode? Please help and don't just say 'you don't know what your getting into'. Thanks!

PS sorry if this is in the wrong category (I'm new)
"No missile ever flew before 10 pm"

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: Questions about the practical construction of a Linac

Post by Carl Willis » Sun Dec 23, 2012 12:01 am

Hi Noah,

I have several years of career experience building and operating RF linacs, and a couple projects at home that are just ion sources coupled to potential-drop beamlines (I suppose this qualifies as a linac in some primitive sense).

The linear geometry requires more attention to charged-particle optics than a fusor, and this translates into more mechanical complexity during construction. I think it takes a good familiarity with standard vacuum fittings, feedthrough technology, and light machining to make a decent shot at it. Glass and ceramics, solders and brazes, and elastomer seals often need to be custom-worked to realize one's vision.

A linac typically consists of a beamline containing the accelerating structure at high vacuum, and an ion source that operates at a much higher pressure. The pressure difference between the source and the accelerating structure is maintained by differential pumping. Unlike a fusor, a fast vacuum system is really paramount for a linac that operates as I have described. Good vacuum engineering is crucial.

There are some accelerator projects described in the old Scientific American "Amateur Scientist" column that offer perspective on the simplest kinds of potential-drop accelerators, powered by Van de Graaff generators. These projects are worth looking at, but keep in mind that vacuum technology has changed radically (and for the better) since those ideas were tested. Also, some ideas that work acceptably with the low-power, low-current Van de Graaff designs are going to be inadequate for driving whole milliamps of current using modern switched Cockroft-Walton supplies. Gradient shaping by the use of external wires on a glass beamline is a prime example.

I hope that input is helpful.

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

User avatar
Noah C Hoppis
Posts: 56
Joined: Sat Dec 22, 2012 9:05 pm
Real name: Noah C Hoppis

Re: Questions about the practical construction of a Linac

Post by Noah C Hoppis » Sun Dec 23, 2012 12:33 am

I suspected that a very hard vacuum was required
Would something like this (https://www.youtube.com/watch?v=PNJGqjB ... Q&index=16) be possible / in the realm of amateur feasibility? Also what would its capabilities be limited to if it was driven by a VM and supplied with protons? Thanks!
"No missile ever flew before 10 pm"

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: Questions about the practical construction of a Linac

Post by Carl Willis » Mon Dec 24, 2012 5:21 pm

The project shown in that video is a high-quality amateur project that follows the design conventions of the old Amateur Scientist column.

It is hard to differentially pump all but the tiniest ion sources with this arrangement, but that's probably just as well since the beamline cannot handle more than a few microamps and the power supply (a Van de Graaff) can't supply more than a few microamps without a debilitating drop in terminal voltage. You might consider some alternative arrangements that put the ion source at ground potential. For one thing, this opens the door to use some powerful ion source designs that require RF or auxiliary DC voltages. It also makes an easier job of differentially pumping the beamline right near the ion source aperture, and of supplying gas to the ion source. If ion source power supplies and gas supplies have to be floated at high potential because the ion source is located at the "hot" end of the beamline, then you have some real challenging engineering to figure out. The downside to putting the source at the grounded end is that the target is then at HV; for certain experiments that can be a nuisance.

When using powerful multiplier-based supplies that can source many mA of current at high potential, some other issues need to be considered. Backstreaming electrons and insulator charging are some problems. The beam emitted from the ion source needs to be well-defined and should not be allowed line-of-sight to any insulating surface. Secondary electrons created bv beam impingement in the accelerator structure or at the target should be minimized by the beam optics design, and then blocked from traveling upstream in the structure through the use of suppression voltages and magnetic traps. It is easy to put magnetic and electric traps at the target and at the ion source, and this is the minimum requirement for keeping the electron current under control. Inadequate control of electron current results in breakdown, damage to the ion source, and horrendous x-ray yields. Finally, you might think about supplying the focusing electrodes (drift tubes) directly from various points on the multiplier stack. Unlike the VdG designs, you can't reasonably get away with a corona divider when running with serious current.

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

User avatar
Noah C Hoppis
Posts: 56
Joined: Sat Dec 22, 2012 9:05 pm
Real name: Noah C Hoppis

Re: Questions about the practical construction of a Linac

Post by Noah C Hoppis » Tue Dec 25, 2012 1:26 am

I was wondering about that! I knew a VDG wouldn't supply enough current and the drift tube geometry would adversely affect te ability of the diff pump to remove all the gas. you mentioned the design not working with a VM, was it simply the lack of supply to each drift tube that wouldn't work, or the whole thing failing utterly due to geometry? I've also heard that a value of ~1G ohm between each drift tube would be sufficient, correct? The comment on electrons back flowing, I was thinking of putting a ring magnet from a Magnetron out of a microwave around the Ion source to stop electrons from running rampant, would that work or would more be required? (sorry for all the questions!)

Thanks!
"No missile ever flew before 10 pm"

Daniel Firth
Posts: 37
Joined: Thu Apr 05, 2012 2:31 am
Real name:

Re: Questions about the practical construction of a Linac

Post by Daniel Firth » Wed Dec 26, 2012 2:30 am

I've been planning an electrostatic LINAC as well. Here's everything I've managed to dig up:

1959 Amateur Scientist Article, http://xa.yimg.com/kq/groups/10603698/7 ... rotons.pdf

1971 Amateur Scientist Article, http://xa.yimg.com/kq/groups/10603698/1 ... erator.pdf

Website detailing John Cockcroft and Ernest Walton's nuclear transmutation experiment:
http://homepage.eircom.net/~louiseboyla ... walton.htm

Fred Niell's site: http://niell.org/linac.html

Segment from "Stephen Hawking's Universe" featuring Fred Niell and his LINAC.
http://www.youtube.com/watch?v=6dnEchmSYzk

Andrew Seltzman's accelerator: http://www.rtftechnologies.org/physics/linac.htm

Particle Accelerators, by M. Stanley Livingston: http://archive.org/details/ParticleAccelerators

Introduction To Experimental Physics, by Wiliam B. Fretter: http://archive.org/details/introductiontoex032616mbp

Ions, Electrons, And Ionizing Radiations, by James Crowther: http://archive.org/details/ionselectronsand030780mbp

John Cockcroft and Ernest Walton published their findings by submitting them to the Royal Society. These articles were then published
in a journal known as "Proceedings of the Royal Society."

Experiments with High Velocity Positive Ions
http://rspa.royalsocietypublishing.org/ ... l.pdf+html

Experiments with High Velocity Positive Ions. (I). Further Developments in the Method of Obtaining High Velocity Posistive Ions
http://rspa.royalsocietypublishing.org/ ... l.pdf+html

Experiments with High Velocity Positive Ions. (II). The Disintegration of Elements by High Velocity Protons
http://rspa.royalsocietypublishing.org/ ... l.pdf+html

Experiments with High Velocity Positive Ions. (III). The Disintegration of Lithium, Boron, and Carbon by Heavy Hydrogen Ions.
http://rspa.royalsocietypublishing.org/ ... l.pdf+html

Experiments with High Velocity Positive Ions. (IV). The Production of Induced Radioactivity by High Velocity Protons and Diplons
http://rspa.royalsocietypublishing.org/ ... 814df7cc22

Experiments with High Velocity Positive Ions. (V). Further Experiments of the Disintegration of Boron
http://rspa.royalsocietypublishing.org/ ... 814df7cc22

Experiments with High Velocity Positive Ions. (VI). The Disintegration of Carbon, Nitrogen, and Oxygen by Deuterons
http://rspa.royalsocietypublishing.org/ ... 814df7cc22

User avatar
Richard Hull
Moderator
Posts: 11936
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Questions about the practical construction of a Linac

Post by Richard Hull » Wed Dec 26, 2012 8:41 pm

It is important to remember that any glass beam line or any metal beamline that works at over 60kv will produce "horrendous" and deadly x-radiation at the target end. Carl mentioned this near the end of his last post.

A linac is a cool device, but as Carl also notes, it requires a special diligence in assembly and operation that the fusor does not. A well constructed linac can do a lot of beam-on-target fusion and make lots of neutrons, but with an attendant x-radiation that must be dealt with in a manner the average fusioneer never has to worry about.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

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: Questions about the practical construction of a Linac

Post by Carl Willis » Wed Dec 26, 2012 9:14 pm

Using a resistive divider or a corona divider is limited to applications where the beam current is very small and the power supply of very low power. If you use a resistive divider to supply focusing voltages and some breakdown or beam mis-steering or blowup occurs (a very routine occurrence), it is possible for the full supply voltage to appear across one or more gaps. The resistors typically cannot survive this unless the HVPS is low-powered, and even if they survive, they may be unable to sufficiently regulate the focusing voltages to get control of the beam. Finally, resistors just waste energy. The lower the divider resistances (and the lower the beam current), the better the regulation of the voltages supplied to the focusing electrodes, but at the expense of power wasted as heat. In a multiplier, you often have access to all points along the cascade and can supply the focusing voltages directly without having to run current down an external divider.

-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: Questions about the practical construction of a Linac

Post by Doug Coulter » Thu Dec 27, 2012 7:19 pm

I have to second this - X rays are real and are a hazard, though I wouldn't declare the amount and energy of the ones a "regular" fusor creates as harmless, myself - which is why I spent all that time and money on lead and lead glass, along with detectors to ensure my personal safety. Even with these measures, at my op position the X rays are roughly 10x the background count - on a simple pancake geiger tube that isn't particularly low energy X ray sensitive compared to say, a plastic or NaI scintillator.

In fact, back in the day when color TV's got to 35-40kv ultor voltages, there was quite the hue and cry about the copious emitted "low energy" X rays, and the industry went to leaded glass face-plates - even with lower beam currents than fusors use. Whether that was self-panicking or a real issue is left to those wiser than myself.

I will also note that there are limits due to the Z of whatever things are hitting - the K lines only go up so far for a given Z, almost no matter the input energy. This was brought home to me in spades when I experimentally coated my fusor tank walls with Pd - while it doubled my neutron output and Q, it also much more than doubled the X ray output due to the high Z of the Pd. Changing this to Ti (which also holds D) cut the X rays way down, leaving the neutrons and Q still higher than plain stainless steel walls.

In fact, with the Ti coating, X rays were less than with the plain stainless steel - even with higher neutron output.
Why guess when you can know? Measure!

User avatar
Noah C Hoppis
Posts: 56
Joined: Sat Dec 22, 2012 9:05 pm
Real name: Noah C Hoppis

Re: Questions about the practical construction of a Linac

Post by Noah C Hoppis » Fri Dec 28, 2012 3:50 am

I expected some X-rays but that sounds pretty gnarly, I guess all that Pb shot will come in handy...

Carl,
I guess this implies corona rings will not be required? So a multiplier of about 200kV distributed down 8 drift tubes should provide proper reaction of an Lithium target with protons?
"No missile ever flew before 10 pm"

Post Reply