particle accelerators / General Science and Electronics / Forums

Forums

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

« Previous topic | Next topic »

particle accelerators

1 2 3 4 next

Move Thread

LAN_403

father dest

Thu Aug 08 2019, 06:37PM

Registered Member #54795 Joined: Mon Apr 06 2015, 11:55AM
Location:
Posts: 28

there is wonderful forum about fusors on the net -

maybe somebody of u guys know something as wonderful, but about electron accelerators instead? especially like microtrons and rf linacs, with theory/construction/results sections. or not so wonderful, but still good read.
am wondering what will be easier to build at home, for about 2-5 MeV or so, 0.1 ma average beam current.

jpsmith123

Sat Aug 10 2019, 11:53PM

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

I've not built anything (yet at least), but I did look into it a little bit.

An RF linear accelerator or a microtron would be a real challenge and the magnetron and associated power supply would probably be very costly.

I came to the conclusion that a voltage-scaled-up version of the "multibeam accelerator" described by G.V. Dolbilov et al. might be one approach (there are a few other possibilities, IMO).

This design eliminates the problem of a big and bulky vacuum feedthrough insulator and associated problems.

This device apparently uses a multi-turn primary driven with 500 volt pulses at 18 khz or something like that. But to scale up the voltage without making it too big, I think you'd need to drive the primary at a much higher voltage with a very short pulse (sub-microsecond), and use a lower turns ratio.

I think you could operate a graphite cathode in an explosive emission mode so you wouldn't need the big mosaic cathode, the downside would be it would need more maintenance.

father dest wrote ...

there is wonderful forum about fusors on the net -

father dest

Sun Aug 11 2019, 06:50AM

Registered Member #54795 Joined: Mon Apr 06 2015, 11:55AM
Location:
Posts: 28

thnx for the info. need time too look into all this more carefully, but i found other work of this dude, like

is this a tesla coil?

with sectioned bifilar secondary?

as for the old ex tesla coiler things are starting to get very interesting for me from this point

jpsmith123

Sun Aug 11 2019, 04:27PM

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

I would call the "Multibeam Accelerator" that I linked to in my first comment a type of Tesla coil based accelerator since it has an external primary winding and a resonant secondary.

(BTW I've often wondered what material is used for the outer housing since it would have to be non-conducting, mechanically strong and vacuum compatible. But I could never find a detailed description of this device).

Also, I've seen a description of another Tesla coil based electron accelerator built in Russia called the RIUS-5 which could produce a 5 MV electron beam, and for which the secondary voltage ws said to reach 7 or 8 MV or something like that. But that machine is big and bulky.

Anyway, it seems the device described in the patent you referred to is not a Tesla coil but something akin to a helical resonator; or the more I think about it, a type of quarter wave transformer.

Since the device (if I understand correctly) does not use a single layer helical coil but a bunch of discrete presumably multilayer coils connected in series, he doesn't describe it as a "helical" anything but uses the phrase "coaxial delay line."

This would be the low frequency equivalent of a quarter wave transformer made out of coaxial cable, IMO, which is driven at the resonant frequency by a generator at one end, and is loaded at the HV end by an electron beam diode.

I've never built anything like this so I can't speak from experience, but I am guessing it would be difficult to reach a multi-megavolt output using this approach in anything of a practical size.

With the transistors available today I think you could operate at a frequency significantly higher than the patent suggests, so you could probably use a single layer helical coil instead of many discrete multi-layer coils in series. And I imagine you'd want to design and build it to get the Q as high as possible.

The main problem I see is the risk of HV breakdown in trying to put a few megavolts across a single accelerating gap with a relatively low frequency CW drive - in anything of a practical size.

I think you'd have to build a prototype to see what kind of voltage you can get before breakdown, and then you could experiment with things like magnetic insulation.

In the link below is an example of a Tesla transformer that's used to generate 2.5 MV pulses. Although the paper doesn't go into very much detail, I think the transformer may be capable of going significantly higher than that. The transformer is roughly the same size as that of the "multibeam accelerator" yet it produces a much higher voltage. IMO this is because it is driven with fast HV pulses using a single turn primary.

father dest

Sun Aug 11 2019, 09:36PM

Registered Member #54795 Joined: Mon Apr 06 2015, 11:55AM
Location:
Posts: 28

jpsmith123 wrote ...

Also, I've seen a description of another Tesla coil based electron accelerator built in Russia called the RIUS-5 which could produce a 5 MV electron beam, and for which the secondary voltage ws said to reach 7 or 8 MV or something like that. But that machine is big and bulky.

this is rius-5

Ir11

a 40ns low repetition rate multi kiloampere device - not what i want. on the other hand this is relus-5

Relus2 20160901 1840114434

rf driven linac with a very compact accelerating srtucture, like this one

Linac Structure

coz u don't need an insulation for full output voltage/energy - that's what i want. and that's why i think all those "tesla" like devices are bs for home builders. and that multi beam device from your first post too - how is it different? it's for multi kw tasks basically, hence more than one beam, but i need 5 times more energy - not the average power. so microtrons/other cyclical systems are the way to go.

i've found even more interesting paper from the same author, at first glance atleast

never saw such a concept before. there is an english version

jpsmith123

Mon Aug 12 2019, 02:47AM

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

"coz u don't need an insulation for full output voltage/energy - that's what i want. and that's why i think all those "tesla" like devices are bs for home builders"

There's actually nothing "bullshit" about it.

"and that multi beam device from your first post too - how is it different? it's for multi kw tasks basically, hence more than one beam, but i need 5 times more energy"

Seriously? The reason I mentioned it is because of the principle involved, not because that particular design would meet your requirements. A field emission cathode situated opposite an anode foil is connected directly to a secondary coil, and the whole thing is under vacuum, thus avoiding a massive and generally problematic vacuum feedthrough insulator. It's relatively simple and cheap, and can theoretically be scaled up in voltage - without getting too much bigger physically - by running a single turn primary driven with fast HV pulses.

" - not the average power. so microtrons/other cyclical systems are the way to go."

Do you know what's involved with a microtron? I happen to have the book by Kapitza and Melekhin and I can tell you it would be involved. It's the way to go if you have many thousands of dollars to spend, a machine shop, all kinds of lab equipment, lots of free time, and you really know what you're doing. And considering the magnet and the power supply for the magnetron, it will not be very small.

Rather than do that - if you've got the resources - you'd probably be better off copying a design for a small PWT linac or a traveling wave linac from a published paper (e.g, thesis) somewhere.

father dest

Mon Aug 12 2019, 07:08AM

Registered Member #54795 Joined: Mon Apr 06 2015, 11:55AM
Location:
Posts: 28

"coz u don't need an insulation for full output voltage/energy - that's what i want. and that's why i think all those "tesla" like devices are bs for home builders"
>There's actually nothing "bullshit" about it.

i mean that it's not very practical to try to insulate voltages in the mv range at home.

>The reason I mentioned it is because of the principle involved
>avoiding a massive and generally problematic vacuum feedthrough insulator

i don't see any problems in getting huge voltages - tesla or not, there's many ways to accomplish that. and what insulator you talking about - why you need it? any example?

" - not the average power. so microtrons/other cyclical systems are the way to go."
>Do you know what's involved with a microtron? I happen to have the book by Kapitza and Melekhin

that book is WAY dated today - they build microtrons on neodymium magnets these days, with iron core weight about 20 kilos only.
and to tune pwt or any other rf linac you must know what you are doing anyway. and know very well.

Sulaiman

Mon Aug 12 2019, 08:04AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

@father dest
the diagram that you showed above looks like a 'pulse forming network' (google it)
I helped to commision a 55 MeV linac, the pfn gave about 1 MV pulses to the anode of a klystron,
(actually four parallel units of 20 MW peak each), the rf was used to accelerate the electrons.

For hobby use a cyclotron may be easier to construct

jpsmith123

Mon Aug 12 2019, 12:02PM

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

"i mean that it's not very practical to try to insulate voltages in the mv range at home"

Nor is it very "practical" to build an RF linear accelerator at home for that matter. That's apparently why you don't see people doing it on youtube. You have a choice between having to deal with HV insulation issues or having to deal with cost and complexity. Pick one.

"i don't see any problems in getting huge voltages - tesla or not, there's many ways to accomplish that. and what insulator you talking about - why you need it? any example?"

Okay, so you haven't done enough research. The problem isn't "getting huge voltages" per se, the problem is creating a semi-controllable "huge voltage" and using it to form a beam in a vacuum system in a system of practical size. Look around at some industrial multimegavolt systems. They're very large.

I suggested fast pulsed HV and vacuum insulation because this way there is no big insulator with problematic "triple points" no multipactor issue and the fast pulses greatly increase the breakdown voltages and keep the size down.

"that book is WAY dated today"

The principles haven't changed; you'll still need a big magnet.

"and to tune pwt or any other rf linac you must know what you are doing anyway. and know very well."

Well of course you must. Why are you asking here if you're an expert?

"- they build microtrons on neodymium magnets these days, with iron core weight about 20 kilos only."

References in support, please.

father dest

Mon Aug 12 2019, 01:00PM

Registered Member #54795 Joined: Mon Apr 06 2015, 11:55AM
Location:
Posts: 28

"i mean that it's not very practical to try to insulate voltages in the mv range at home"
>Nor is it very "practical" to build an RF linear accelerator at home for that matter. That's apparently why you don't see people doing it on youtube. You have a choice between having to deal with HV insulation issues or having to deal with cost and complexity. Pick one.

so my pick is number three then - a microtron, or some other type of cyclic accelerator.

"that book is WAY dated today"
>The principles haven't changed

yea, but there is a ton of new ones, which were unheard of at the time of that book.

>you'll still need a big magnet.

i don't - if i am using those new principles. so as u've said above - "you haven't done enough research". read the last paper of G.V. Dolbilov from my posts for example:

wrote ...
...key parameters of the electron accelerator on energy E = 10MeV
The maximal size of a magnetic induction of Ð¡-shaped cores is equal B0max = mÎ³v / eR0 = 0.067 T
...magnetic induction Ðž-shaped cores does not exceed B = 0.1T
To realize such accelerating system the minimum of 10 ferromagnetic cores with the sizes 125 x 80 x 10 mm is required.

"- they build microtrons on neodymium magnets these days, with iron core weight about 20 kilos only."
>References in support, please.

1 2 3 4 next

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