cockcroft walton supply?

This forum is for specialized infomation important to the construction and safe operation of the high voltage electrical supplies and related circuitry needed for fusor operation.
Post Reply
colinheath
Posts: 78
Joined: Sat Oct 19, 2002 2:48 pm
Real name:

cockcroft walton supply?

Post by colinheath »

hi all,
im back into the fusors now and have sold my demo unit to fund a nuclear unit. my first biggest hurdle is the power supply as x ray transformers in uk are very hard to come by so im thinking of building a cockcroft walton to run it
im thinking of going for the full wave style to give more current and think i should be able to get 40-60kv with approx 6-8ma
will this be enough current ?
cheers
colin
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: cockcroft walton supply?

Post by Richard Hull »

I would think at least 5 ma would be needed for decent fusion work in a fusor. 10 ma would be an ideal level for most work.

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
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
colinheath
Posts: 78
Joined: Sat Oct 19, 2002 2:48 pm
Real name:

Re: cockcroft walton supply?

Post by colinheath »

hi richard,
thanks for the reply. ill try and design a unit to provide this although i only have 1.5nf caps at 25kv dc rating so im not sure that 1.5nf will give the current i need
im not sure how to work out the current output of a c w so i gues ill just build it and go from there
cheers
colin
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: cockcroft walton supply?

Post by Richard Hull »

Most multiplier systems, the net filter capacity is the sum of a string of capacitors. Thus if you have a 4 stage multiplier and have four, 1.8 nf in series, then you have an effective filter of about 0.45nf at the rated output voltage. Now figure on a 10ma load on the expected voltage. This means that at 60kv with 10ma draw, the effective DC ohmic load is 60,000/.01 or 6 megohms. The Tc for that would be 6x10e6 X 4.5 X 10e-10 or 2.7x10e-3 seconds. As the period of a 60hz power cycle is ~17ms you are far too short in period.

You should have an effective total capacity of at least 3 Tc of the power frequency to hold the business up. Thus, about 50ms Tc would be needed. The ideal would be a net capacity of 50x10e-3/6x10e6 or ~8X10-9 farad or 8 nanofarads. In a four ladder capacitor string that would make each cap have to be 32 nanofarads. (.032ufd)

All the foregoing is rough, off the top of the head stuff, but is within the order of magnitude of what would be required in a half wave quadrupler.

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
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
JohnCuthbert
Posts: 339
Joined: Mon Aug 19, 2002 4:30 pm
Real name:

Re: cockcroft walton supply?

Post by JohnCuthbert »

There is a formula for the voltage drop under load of multipliers at
http://home.earthlink.net/%7ejimlux/hv/cw1.htm
(assuming I have transcribed it propperly)
Unfortunately the current seems to drop very fast with small capacitors
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: cockcroft walton supply?

Post by DaveC »

Cockroft -Walton multipliers need to run at higher than mains freqs for a couple of reasons - to keep capacitors to reasonable levels, to minimize voltage regulation and ripple, and to minimize the stored energy at HV. The latter is for personnel safety.

Ceramic HV caps up to 15 kV and 2 nF, are relatively inexpensive. There are several sources. HVC out of Wisconsin, Murata (about the best qualtity, but looong lead times), Johannsen in San Diego..and Cornell Dublier, which is touting a "reconstituted mica" capacitor.

If you can find an old Xray HVPS, you can probably get some of the parts you need there.


For a few hundred watts, small ferrite core transformers are convenient. They typically need to operate at 15 to 50 kHz for best results. You may have to consider winding your own primary and secondary.. This is not as hard as it might seem, but inter layer insulation is a must.

HV diodes are the other remaining item of importance. On shore sources are not usually inexpensive or fast in delivery.

One thing that can be misleading is the formulae for voltage drop and ripple as a function of the number of multiplier stages.
The output voltage is equal to the number of diode/cap.pairs or stages times the peak input voltage.. typically a square wave if High freq. design. The formulae do not consider that a large number of large cap - lower voltage diodes - will have identical forward drops of the smaller number of high voltage diodes/cap stages. Remember HV diodes are built from stacks of 10-20 low voltage diode chips in series. The same is true for the HV ceramic caps. Some are built up as stacks of lower voltage, higher capacity units. Others are just a monolithic ceramic piece.

Another limitation is the reverse recovery time of lower voltage or standard diodes. This will limit you to lower frequencies, and may rule out the ferrite transformer core.

All in all, trade-offs abound.

Dave Cooper
kbonin
Posts: 123
Joined: Sat Jan 04, 2003 9:54 pm
Real name:
Contact:

Re: cockcroft walton supply?

Post by kbonin »

Are there any good online references available for design tricks, optimizations, equations, etc. ad nauseum for small through robust designs? I've seen many interesting tidbits, such as Glassman's "technology" page, any other good recommendations for anyone thinking about building something custom here? I'm interested in all the elements - PWM, coil winding, current limiting, etc.

I've been scouring the archives here, and looking around the web, but other than the one thread Larry started recently and a few I found from it I haven't seen any really good design info sources for cockcroft walton designs.

Thanks!
3l
Posts: 1866
Joined: Sat Jul 07, 2001 3:51 pm
Real name:

Re: cockcroft walton supply?

Post by 3l »

There are NO net sources except the ones I posted.
This stuff had it's hey day in 1920's.
Richard gave the crux of it in the following post:
2003-17-03 14:09 Re: cockcroft walton supply? (Richard Hull)

I will have to post the Glasstone's Nuclear Engineering Guide to this stuff.
It's from 1958!
Multipliers are moldy old tech ..Any 1930's accelerator book
will have it ...(pre cyclotron).
This stuff lives in old books only.
Josten's high voltage page is another.
http://www.kronjaeger.com/hv/
Any old timey experimental nuclear physics book will have it.

Happy Fusoring!
Larry Leins
Fusor Tech
kbonin
Posts: 123
Joined: Sat Jan 04, 2003 9:54 pm
Real name:
Contact:

Re: cockcroft walton supply?

Post by kbonin »

Thanks! I greatly appreciate any info you can pass on or post, my library is conspicuously absent of Cockcroft-Walton tech, and after looking at where I am and where I want to go, I think I want to play in that area. Thanks for the kw-tubes ref too!
Post Reply

Return to “High Voltage - Fusor Input Power (& FAQs)”