High frequency high voltage generation

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

High frequency high voltage generation

Post by 3l »

Ok I know that a lot of members have been telsa coilers for a long time....
Would it be herresy to suggest using a vacuum tube driven
telsa coil as a high frequency high voltage source?
Thought about it long age but lost interest when I realized the difficulties of rectifing high freq high voltage. But it was revived for me when I realized a telsa coil would not have to be rectified
to run a linac.
Hence my interest.

I built one years ago with a two tube driver system that generated 500W at 425 khz with my dad's help in 1969.
But that was a reciepe type project.
I have long since lost the plans and the device.
I have recently run across plans for a higher frequency telsa coil that runs at 645 khz at 15 kv. It has a single tube for the oscillator. Humm ... would it be possible to build a higher frequency unit at higher voltage.

The website is:

http://www.pupman.com/msrp/vt_mr.html

From his basic plan one could modify his unit to provide a higher frequency and voltage by simply using tunable caps on the tube oscillator with more windings on the secondary.

Q1: what is the highest freq an amatuer could resonably expect?
Q2: Could a relaxation oscillator of say a thyratron could be used for higher imput voltages ?
Q3: Would you have to be another Telsa to bring it off?

I ran across a reference to a particle accelerator that made 5 mev deuterium ions......it was an oil immersed telsa coil under pressure with lo and behold ... a linac.
Any comments would be welcome.

Fusion is fun!
Larry Leins
Fusion Tech
User avatar
Carl Willis
Posts: 2841
Joined: Thu Jul 26, 2001 7:33 pm
Real name: Carl Willis
Location: Albuquerque, New Mexico, USA
Contact:

Re: High frequency high voltage generation

Post by Carl Willis »

Hi Larry,

The CW tesla coil has been used most directly for high-energy x-ray tubes. In this case, the secondary winding is located in the vacuum chamber and the "hot" end of the winding holds a target for electrons. Use for ion linacs would be rather difficult for several reasons, however. Ions simply are not fast enough to stay in phase with the accelerating potential. I'd be interested to see how that 5 MeV linac was configured.

I have worked with some tube tesla coils that make ~24" sparks: http://www.angelfire.com/electronic/cwi ... acoil.html

Upper frequency limits are set by the ability to make high-Q secondary coils. 10 MHz is really the top in my experience.

Yes, a thyratron of sufficient ratings could be used as a switch to replace the inefficient spark gap in conventional disruptive-type coils. Problem is, they're just hard to come by. Best results--as in biggest sparks--with the tube-driven coils are had with high plate voltages, pulse-rated tubes, and very high plate currents at low duty cycle. Thus, the more they start to look like a disruptive type of coil, the better they perform.

I think it would be fun and easy to make a small ~5 MHz tube-driven coil whose secondary was self-supporting and confined in a vacuum chamber. But such an apparatus would only be suited for producing electrons or x-rays. Could you let me know if you remember the details on that proton linac you referred to?

-Carl
Carl Willis
http://carlwillis.wordpress.com/
TEL: +1-505-412-3277
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: High frequency high voltage generation

Post by Richard Hull »

All stock tube driven coils are just colpitts and or hartley oscillators on steroids with whacked out coil sizes.

The limit in frquency is in the 50 mhz range, but the practical limit for super HV might be 100khz, preferably lower if your hv input to the oscillatory is limited.

The rules for high voltage output are:

Use very HV input to the oscillator (>10kv). You will need a very special tube. You can go to 5mhz easily with small, appropriate coils.

If you can't get a special tube, then use a super old tube like the 4-1000 and you stay under 3kv on the input and drop to 50khz in frequency where giant coils can get you the HV needed.

A hydrogen thyratron can be used in low duty cycle pulsed circuit where 35kv input voltages are possible (again with selected H2 thyratrons) and any frequency can be used. Here you are just replacing the gap in the regular, classic Tesla coil.

I have built them all, but again there is that ultra high voltage rectification problem off the Tesla coil. Unless, of course your linac will rectify in its normal use mode like an X-ray tube.

Vertually all Spellman supplies in the 0-50 kv range use a form of Tesla ascillator to generate the high voltages. Most are limited to a few mils though.

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

Re: High frequency high voltage generation

Post by 3l »

Correction:
OOps! that accelerator system was a voltage drop setup not a linac. It had charge balancing resitors with some far out snubber network that I mistook for a linac. The coil itself was a 5 MV coil with an accelerator tube in the terminal capacitor terminal.
I understand the Institue for Terestrial Magnetism used it for about five years before it went to a school ...Osu.
It sat in a 500 gallon oil tank. It ran at 40 psi of pressure in oil.
It had specially folded capacitors that needed high pressure oil to maintain them.
It was abandoned in early 58 due to 1/2 cycle losses. ie it had no rectification would only work on the positive cycle.

My plan was to build a 50 KV at 10 mhz system.
Although not a simple thing it seems doable.
I had planned a simple 1 tube system around a 6146 tube.
But your 883a plan blows it out of the water.
The linac I've planned is more properly a voltage drop accelerator
with a Telsa coil rather than a Van de Graph generator.
rather like the high school kid's but with a decent ion source.
1 ma or so

I plan to pre accelerate the deuterons to 15 kv at my ion source
before the linac.. If I can not get the linac down in size rfq is next.
The first drift tube will be huge -15 kv the tube is 1.8 meters long
at 675 khz) a tad bit long but I've got a yard). telsa coils would work in rfq accelerators.
Hey this is a dream compared to the Ullam Teller quagmire.

Fusion is fun!
Larry Leins
Fusion Tech
User avatar
Carl Willis
Posts: 2841
Joined: Thu Jul 26, 2001 7:33 pm
Real name: Carl Willis
Location: Albuquerque, New Mexico, USA
Contact:

Re: High frequency high voltage generation

Post by Carl Willis »

Hi Larry,

You'll find that the 6146 is an excellent, durable tube for small Tesla coils. Also on my website I describe a "plasma tweeter," which is one of the few other Tesla coil applications, that is exactly a 10 MHz, 6146 Tesla coil. Except it's not really a Tesla coil, more like a quarter-wave transformer--it has no primary and no large terminal capacitance. All you need is listed at http://www.angelfire.com/electronic/cwi ... eeter.html; just take the RF final part of the circuit on the left side of the page. I'd say it does at least 50 kV until its terminal arc loads it down. In vacuum you might see even more than 50 kV with low currents. I'll reiterate, however, that it will not work satisfactorily for accelerating ions unless they are already moving at nearly relativistic velocities or the coil frequency is much much lower.

About the RFQ: in my understanding, they need very high frequencies, being travelling wave devices (also I presume they are one helluva biotch to machine and probably require sofisticated beam transport computer code to design). But here's a theoretical idea: make four long helical tesla-style coils, with proper turns density as a function of length--which would be a challenging calculation--and arrange them as an RF quadrupole. Drive them in proper phase, and particles along the quadrupole axis will be accelerated by the alternating voltage antinodes along the coils. This way you might could use the RFQ idea with low frequencies. But this is a real reach of an idea fraught with all sorts of trouble.

Your 6146 idea I think will prove to be an excellent, portable source of x-rays should you chose to build it...and I don't beat ya to it (heh!) Just don't plan on doing protons or anything heavier with it. Let's say you have a 100 kVp tesla coil at 10 MHz, and your linac is 30 cm from HV terminal to ground. When the terminal reaches peak negative voltage, think about what happens if you start a proton from rest at the grounded end. It sees on the order of 300 kV / m depending on geometry. How far is it gonna go in a quarter period (25 nanoseconds), and more importantly, how much energy does it have after that? Doing the kinematics, it can be shown that the proton will go less than one centimeter before the terminal voltage reverses! At best then, it only gets a few keV before slowing down and getting turned around. The deuteron, being more massive, would not even be this well off. While you can up the voltage and make the linac even shorter, practical problems catch up with you rather quickly and you still don't get high energy particles for all the effort spent.

-Carl
Carl Willis
http://carlwillis.wordpress.com/
TEL: +1-505-412-3277
3l
Posts: 1866
Joined: Sat Jul 07, 2001 3:51 pm
Real name:

Re: High frequency high voltage generation

Post by 3l »

Linacs are funny critters in I need a low voltage such as 10-20 kev but at 40 to 50 Mhz....strange but true. I resolved the rectification problem tho. Surface currents defeat the use of vacuum tubes but if the proper sized diode was in the vacuum you could concievably use the voltage drop method. Still plugging at it. My other work is still proceeding.
Carl : I really like your tweeter...it's like I do projects.

Thanx guys for the input and will ask more questions later.

Larry Leins
Fusion Tech
grrr6
Posts: 217
Joined: Mon Sep 16, 2002 11:10 pm
Real name:

Re: High frequency high voltage generation

Post by grrr6 »

The only problem i see, is that when hooked up to the linac, the drift tubes will create a relatively large topload capacitance. This will want to drive the frequency way down, so if you want a super high frequency, you need a really really low inductance coil.

In the link you gave at the start of the thread you can see that the guys topload is tiny, a sphere only 1 inch in diameter. Im assuming your linac is going to be around that in diameter, but a whole lot longer, so that you'll have maybe 100 or so times the capacitance, just a rough estimate. This will lower the resonant frequency of the existing coil by a factor of 10, and since you already were aiming at increasing it by around a factor of 70, from 650 kHz to 40 MHz, it will be a feat. You'll need an increadibly miniscule secondary coil. And then, to keep the same voltage gain, youll need a miniscule primary coil.
User avatar
Carl Willis
Posts: 2841
Joined: Thu Jul 26, 2001 7:33 pm
Real name: Carl Willis
Location: Albuquerque, New Mexico, USA
Contact:

Re: High frequency high voltage generation

Post by Carl Willis »

Greg, you've definitely noted one of the big problems with linacs. At the frequencies that most of them operate, it becomes more sensible to use transmission-line resonators for high Q and high power operation. But--I'll mention it again--the drift-tube linac as we know it does not really become useful until the particles are already at a rather high energy, moving fast enough that they can stay in phase with the alternating potentials. Most RF drift-tube linacs are fed with electrostatically accelerated particles at a few hundred keV at least.

-Carl
Carl Willis
http://carlwillis.wordpress.com/
TEL: +1-505-412-3277
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: High frequency high voltage generation

Post by DaveC »

Not sure in all the fascinating detail here.. whether the objective was AC or DC at HV.
If it is DC, and a LINAC is the end result, why not build a horizontal 1/2 wave voltage multiplier connecting each stage to an accelerator dirft tube. If the vacuum is in the 10-6 Torr range, then 40 KV/mm fields are sustainable.and dimensions become quite modest. Depending on what kind of power output you need, one can probably consider up to 40 stages, before the droop gets impractically large.

For all solid istate elements, use the fast recovery (30 -50 ns) HV diodes, and run at 30 - 50 kHZ. Whole thing gets very compact. I have built several 125 kV oil insulated, 20 stage units that run at more than 100 watts output. You can hold them in the palm of your hand... including the drive transformer.

I think a .5 to 1 MEV LINAC with 30 kV stages could be built less than a yard long. The voltage is less of an issue, I think , than the basic safety here.

Dave Cooper
3l
Posts: 1866
Joined: Sat Jul 07, 2001 3:51 pm
Real name:

Re: High frequency high voltage generation

Post by 3l »

Great idea Dave!
My aim was high voltage Dc in one step.
But your plan is a very neat alternative.
It solves the light weight quest and the compact linac problem in one stop.
Brainstorming is a great asset on the forum

Thanx

Fusion is Fun!
Larry Leins
Fusion Tech
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: High frequency high voltage generation

Post by DaveC »

Most welcome... the idea exchange here is great!!

Dave Cooper
Post Reply

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