Longest Power Supply search?

[mobious]

What's the longest period of time anybody has ever experienced searching for the right power supply? I've been hitting E-bay and Lab-X every single day for the last 3 weeks. To say the power supply with the right combination is hard to come by is an understatement.

I've got plenty of NST's and MOT's laying around for my Tesla coils but I'm not going to be happy with just a demo fusor.

[Steven Sesselmann]

Glen, I know what you mean, I ended up buying a brand new one from Gamma HV ($7,000), they upgraded their website after I bought that

... check out.

http://www.surplussales.com/PowerSuppli ... S-8-2.html

and

http://www.gammahighvoltage.com/

Steven

http://www.beeresearch.com.au

[Mike Beauford]

I've looked for months and gave up. I ended up deciding to build my own. It's actually been very rewarding with all the stuff I've been learning (and never really wanted to know about, clamping, overshoot, zero voltage switching, etc...). I have to say if you build your own, the hardest thing will be winding the transformer. Everything else is somewhat straight forward. At least that's what I'm saying right now.

[UG!]

think its been about 2 years now every now and again i buy something to make my own. currently have the transformer core, transformer wire and huge heatsink. kapton sheet and power semiconductors still to go

Oliver

[MarkS]

I looked for a little over 8 weeks, and two days ago bought a -30kV glassman off ebay which is guaranteed to work, but I have to do the testing myself within 7 days of arrival. I think in order of importance it goes vacuum > HV > radiation safety > D2 gas > radiation measurement.

And if you're willing to pay $800 plus shipping from the UK,there's a glassman -40kV @ 15mA for sale in Ebays "buy it now"

[Starfire]

Glen - a simple way is to use a X-Ray transformer and variac on the primary to control - it is easy to fit a rectifier system on the secondary circuit, if you do not require precession voltage and current control. Fusors will work just as well on a simple supply.

It will cut down on the expense and probably provide a more powerful system.

[mobious]

What about the Microwave inverter that's catching a lot of hype lately? Would a CW multiplier use up too much power? 300mA.....that's a lot of room to play.

[MarkS]

Heres a microwave schematic, the neat thing is the timing and power mechanism are built in. I have the transformer and everything, i just pulled the magnetron.

[Carl Willis]

I think Glen was referring to the high-frequency inverter power supplies for certain brands of oven (Panasonic in particular) that operate at ~30 kHz.

They're usable as Cockroft-Walton stack drivers, with some caveats. First, they do not produce anything resembling a sine wave. Consequently, you have to be careful to understand the waveform whose voltage you are multiplying and be careful not to stress the CW components. Second, you have to be able to understand how the driver board interfaces with the timing components in the other parts of the oven circuit so that you can "hot wire" the inverter to turn on and stay on at full power. With some ovens this is easy to figure out, but apparently with some others it is not so easy. I have posted on this process before I believe.

The inverter circuit will shut down if it senses a fault (and has to be unplugged and plugged back in to reset). Annoying, but it does protect the parts.

-Carl

[mobious]

Are they suitable for fusion?

[MarkS]

If it can drive a CW multiplier it should be fine for fusion, the end result is high DC voltage and low amperage. The schematic I put up before isn't current limited, but does have a fuse. I couldn't tell you about the panasonic ones though.

[Richard Hull]

Carl sort of answered your question in his post related to.." will it do fusion"?

The answer was qualified well by Carl.

As is and even in slightly modified form the answer is a strong NO.

As a basis for a modified setup and driver with selected components that are very fast (expensive), YES, this might form the basis of a decent lower order fusion supply.

The HF transformer is one limiting factor and slow components often readily avaialble to the experimenter is another.

There is no simple quick solution to obtaining fusion power supply systems, especially over 30kv.

every one wants a switcher arrangement as it is very efficient, very compact and easily controlled with little or no stored energy in the filter components.

Richard Hull

[mobious]

Thank you all..... searching for the stuff is part of the hobby I suppose. I just blew my 15/60 NST tonight while trying to squeeze that last little inch of streamer length out of my Tesla Coil. I'm still getting voltage from one HV bushing so I'm hoping its just a carbon arc through the potting to ground. If baking the potting fixes the other HV bushing I'll just go ahead and turn it into a demo fusor power supply. I've got a DRSSTC driver on the bench that I'm working on now also, so I'll be hanging coiling up for a bit. It's a learning experience of its' own.

I think this project is going to cure my instant gratification tendency. Not to mention I just got my wisdom teeth pulled today.........one word....Percocet. Never knew a wall could make me laugh so hard.

[Carl Willis]

>I just blew my 15/60 NST tonight while trying to squeeze that last little inch [...]

>just got my wisdom teeth pulled today [...] one word....Percocet. Never knew a wall could make me laugh so hard.

Glen-- condolences on the loss of your wisdom teeth and your 60 mA transformer in the same day. But something caught my attention about your account here: you should not be 'coiling under the influence of narcotics like large doses of Percocet. In retrospect, that probably makes a lot of sense. At the time, it might have seemed like a fine thing to be doing. But that's the insidious, deceptive game that a lot of drugs--both legal and otherwise--will play with a person's head. Sorry to be the purveyor of hackneyed moralizing. Just be careful.

On to a more technical point: If you are successful at solid-state Tesla coils, you can easily convert that experience toward the construction of HF switching power supplies for fusors. Much of the same circuitry shows up in my multi-kW C-W stack driver: PWM chip, IGBT H-bridge, resonant load leg. The stack requires, as Richard pointed out, high-frequency-rated diodes. Still, you can save huge amounts of scrounging time and monetary expense by homebrewing this supply. No need to sack Panasonic microwave inverters and use them in flaky C-W stack drivers if you can do much better yourself. Jon R. and I have been using homebrew SLR drivers and I think we both like their performance. My level of experience with switchmode power supplies is not on par with some other hobbyists like Steve Ward, and yet I've managed to last most of a year now without smoking a single transistor.

-Carl

[mobious]

lol, no no no, sorry, percocet taken at bedtime! Ibuprofen, while coiling. Thanks for pointing this out. You never know with some people.

This is my first attempt at anything solid state. I wondered if the driver board could be used in as a power supply so you answered that question before I even considered asking. I have only a basic understanding of electronics, which is one reason I chose to build the driver board. Now that you mention it, if you have a schematic for your fusor supply that I could study I would love to see it.

[Jon Rosenstiel]

Glen,
Just a little ways back in this forum you'll find a post by Carl describing the SLR topology full-bridge driver he built…

viewtopic.php?f=11&t=4569#p27407

Carl's informative post (and his link to Steve Ward's site) is what got me started down this road. I didn't know the difference between a half-bridge and a full-bridge, and barely knew what an IGBT was. I spent some time Googling, did some experimenting, and was on my way to power supply nirvana. My words of wisdom..."If ya' can read a schematic ya' can build it".

Jon Rosenstiel

[mobious]

Reading the schematic is easy...... figuring out how your going to arrange all those pretty little components to fit together in a nice clean manageable fashion is another story.

Thanks for the link. I did a fair amount of searching on the subject before hand. But, the subjects don't always convey what exactly your searching for.

Of course, each path is not without it's own obstacles. I could build the driver no problem (in time) but obtaining the stack is probably just as hard as finding the perfect Glassman or Spellman supply on eBay.

Unless of course the stack isn't required for fusion. But, in my understandings so far it's not only power you need for fusion it's stored energy that is needed to cope with the dynamic power requirements of the fusor. I would assume the stack would not provide that unless said driver is more sophisticated than I thought.

[Jon Rosenstiel]

Glen McGowan wrote:
> Reading the schematic is easy...... figuring out how your going to arrange all those pretty little components to fit together in a nice clean manageable fashion is another story.

I had the best of intentions of making mine pretty, but unfortunately, it didn't turn out that way... but it did work. I just made sure it was buried it deep inside of the supply cabinet so as to not attract anyone's gaze.

In addition to the driver you need a ferrite cored step-up transformer and the CW multiplier stack… so it is rather complicated. But, if one has the time it is a great project, and a great learning experience. (I’m retired, so time I got)!

Don’t forget about the lowly, iron-cored x-ray power supply. They’re much, much simpler, albeit heavier, and usually larger, than either the Glassman or the Spellman supplies.

No, it’s just the opposite; stored energy (as in energy stored in a capacitor or inductor) is the fusor’s enemy. That’s the reason we need ballast resistors.
My experience (thanks to Carl W.) has been that the higher (up to a point, at least) one makes the ballast resistance the easier the fusor is to operate. High ballast resistance (100 Kohm to 1 Mohm) means there will be a fairly large voltage drop across the resistor, (and quite a bit of power dissipated also). And a large voltage drop means that if you plan on running your fusor at 40 kV your supply will need to have an output considerably higher than 40 kV. (As usual, a bunch of trade-offs). Another thing I’ve found (at least when it comes to running a fusor) power supplies aren’t “happy” running at their max output. There seems to be a need for some “headroom”… somewhere around 20 ~ 25%. But then again, this is just my experience… your mileage may vary.

Jon Rosenstiel

[mobious]

There's got to be a nice compact unwieldy solid state solution out there (That I can build). I just want something that spits out -30kvdc @ 20 or so mA (Variable of course) to power my fusor.

My interest in the fusor and tesla coils is the physics not the electronics. I really have no interest in electronics really other than it makes for some interesting physics. Unfortunately, I can't play physics until I learn the electroncs aspect.

Oh well, you can't have your cake and eat it too I suppose.

[mobious]

I know folks have posted about these rinky-dinky transformers from this site before and been shot down everytime. But, I've never seen this one posted. Seems the voltage and current are in the ballpark at least. The only issue is the polarity is without a doubt positive. It doesn't look like there would be too much to it in the way of changing to negative output. Then again my x-ray vision as never been all that dependable, Thoughts?

[MarkS]

I built the ion gun from amazing1 from the kit. I doubt that the diodes are potted in that, and there is no stack, thats just a driver, given if it is just a driver its only 4 diodes in the bridge? Not to hard of a reversal job.

[Wilfried Heil]

This thing has an RF output and no rectifying diodes in it. Maximal power is spec'd at 10 kV and 12.5 mA, which would be ok for a demo fusor, but not enough for real fusion.

You can build something similar inexpensively with a Royer oscillator (check Google) and a TV high voltage transformer. If you try that, be sure to get an old transformer with no diodes in it, since these would be wired for positive polarity and can't be reversed.