Competition: High Voltage Power Supply

[DerStrom8]

Hello everyone!

A member has asked me to set up a competition for a project he's working on. Before I introduce the competition, however, since this project deals with high voltage and dangerous currents, I feel I must post the obligatory warning.

*****WARNING*****

THIS PROJECT INVOLVES VOLTAGES IN THE TENS OF KV RANGE, WITH CURRENT LEVELS THAT COULD CAUSE HEART FIBRILLATION AND ULTIMATELY DEATH. PLEASE DO NOT PARTICIPATE IN THIS COMPETITION IF YOU HAVE LITTLE OR NO EXPERIENCE WITH HIGH VOLTAGE UNLESS YOU HAVE A KNOWLEDGEABLE PERSON OR PERSONS TO HELP YOU. THIS IS FOR YOUR OWN SAFETY.

Now that that's out of the way, let's begin. The following are the criteria for the project:

UPDATE: PLEASE VIEW POST #92 FOR UPDATED CRITERIA


Ideal design should:

--Drive a "flyback" transformer from a Zenith TV, like the one shown here, though other transformers will be considered:



--Supply a stable, yet variable voltage output of roughly 0-20 kilovolts, with enough current to sustain an arc rather than a series of small sparks. Adjustable current is preferable. The closer the current comes to 60mA, the better.
--Must be able to run for 24 hours straight without damaging components or creating excessive heat
--Fit on a 4x4 inch perf board, not including the flyback transformer or any external power supplies

1. Try to keep the cost under $150.00us. Includes Shipping.
3. Focus on how you can sustain an arc 24hrs at 20k without burning things up. That's the question. If it's a Flyback design, how far would you push it? Only 1ma? So be it. If it's a NST, again, how far would you push it? The same question goes for the tripler John suggested. Also, what would be the sacrifices in order to get 20kv? These are the questions for the contestants to answer for themselves.
4. NST's (solid state type, due to their small physical size and weight) and Tripler's are allowed, provided the cost remain under $150.

These specifications are not firm, so feel free to let your mind run. Designs will not be judged based only on how much of it matches the specs. Instead, they will be judged on how well they operate, efficiency (use of space, etc), and overall design.

Bonus points will be given for original designs (not pulled off google), and for designs which allow for switching between battery and mains.

Entries should be submitted no later than March 1st. The community will vote on the entries to determine the best 5, based on theory and overall design. Finally, I have been selected to build these 5 circuits and choose the best design according to the afore-mentioned guidelines.

The winner will receive a $75 prize from the member hosting the competition.

EDIT: Prize has been increased to $100 to anybody who can take the given information and create a design that shows ingenuity and cleverness.

If you have any questions, feel free to PM me and I will do my best to answer them for you.

This project is a great way to expand your horizons, gain experience in the field of electronic design, and to release your inner creativity! Have fun!

Best wishes, and may the best design win!

Regards,
Matt

 

[KeepItSimpleStupid]

OK, Matt:

Don't have time for the project, but you do or should provide some basic info for the flyback transformer.

Are you also defining regulation? Line, load
How about ripple?

 

[DerStrom8]

OK, Matt:

Don't have time for the project, but you do or should provide some basic info for the flyback transformer.

Are you also defining regulation? Line, load
How about ripple?

The project is not my own, but I will send a PM to the real host (who has asked to remain anonymous) and see what he says. Thanks for your patience!

Regards

 

[tcmtech]

I already see some potential issues.

60 Ma at 20 KV is 1200 watts which is way past what a typical flyback transformer from a CRT type TV or monitor will handle.

Also getting a full drive system for that level of wattage to fit on a 4" x 4" circuit board will be tough as well unless heat sinks are not included.

 

[DerStrom8]

I already see some potential issues.

60 Ma at 20 KV is 1200 watts which is way past what a typical flyback transformer from a CRT type TV or monitor will handle.

Also getting a full drive system for that level of wattage to fit on a 4" x 4" circuit board will be tough as well unless heat sinks are not included.

He mentioned that the 60mA is not firm. He was hoping for at least 15mA, but prefers as close to 60 as possible. 15mA at 20kv is still 300 watts, but I'm thinking the flyback can handle it.

As for the size of the board, it is not firm either.

 

[DerStrom8]

OK, Matt:

Don't have time for the project, but you do or should provide some basic info for the flyback transformer.

Are you also defining regulation? Line, load
How about ripple?

Okay KISS, I've talked to the host, and unfortunately, since this thing was pulled out of an old TV, many of the specs are unknown. He did have a photo of the label on the transformer, though I am not sure how much help it will be to you. I've posted the photo below.

I believe this will just be an arc generator, so the output would be a near short. A high power resistor might help things from overheating (hint hint).

I forgot to mention that if the circuit is designed to run on mains (which, if running for 24 hours straight, probably should be), it will have to be on 120vac.



@TCM, regarding the heat sink issue--

The host has suggested placing the whole thing on a small board with standoffs to set it up a bit. Setting it on a board should also give a little more room for heat sinks to be attached and for the larger power electronics.

Perhaps the 4" x 4" spec should just be ignored, though smaller designs would be preferable. Keep the final product as small as possible, and the judging will not be too strict

Regards,
Matt

 

[Nigel Goodwin]

He mentioned that the 60mA is not firm. He was hoping for at least 15mA, but prefers as close to 60 as possible. 15mA at 20kv is still 300 watts, but I'm thinking the flyback can handle it.

Maximum design current is only 1mA - I would have thought 15mA massively too high for it.

 

[DerStrom8]

Maximum design current is only 1mA - I would have thought 15mA massively too high for it.

Where do you get that? Were you able to find a datasheet?

I have often seen them supplying much more than that. 1mA would not be able to create a sustainable arc--only a series of small sparks. I'm sure I've seen them put out 10mA at the very least.

This model of flyback is also one of the best. It's very sturdy and rugged, and can often be over-driven without any problems. ZVS drivers can often supply around 100mA from the flyback, and the transformers used are the same type as this one.

 

[Nigel Goodwin]

Where do you get that? Were you able to find a datasheet?

Manufacturers don't provide datasheets - 1mA is the designed maximum EHT current for a TV - and the LOPTX runs hot in TV's, and is a major source of failure.

 

[DerStrom8]

Manufacturers don't provide datasheets - 1mA is the designed maximum EHT current for a TV - and the LOPTX runs hot in TV's, and is a major source of failure.

Interesting. Never knew that

The host said he measured the current and it read 3mA, but I still think you could get 10-15mA out of it, if not more, since it's not being limited to use in a TV. But again, the current output is not set at a firm value. The higher the current, the better, though all designs will be considered without bias.

 

[Nigel Goodwin]

The host said he measured the current and it read 3mA, but I still think you could get 10-15mA out of it, if not more, since it's not being limited to use in a TV.

So if you buy a car with a maximum speed of 100mph, you'd expect to be able to get 500+mph from it?

You might be able to get short term higher current pulses from it, but sustained current would just cook it.

 

[Ian Rogers]

As this is a competition... Why don't we suggest alternatives rather than shoot down the requirements... Designing a flyback transformer that WOULD do the job would be a hell of a lot more positive....

If this competition succeeds, it opens the possibility of others having designs ( proven and built ) that they can use in the real world.

You have all seen the "Electric discharge" video. So we know the competition is viable....

 

[DerStrom8]

As this is a competition... Why don't we suggest alternatives rather than shoot down the requirements... Designing a flyback transformer that WOULD do the job would be a hell of a lot more positive....

If this competition succeeds, it opens the possibility of others having designs ( proven and built ) that they can use in the real world.

You have all seen the "Electric discharge" video. So we know the competition is viable....

Thank you Ian, that's a good point. I'll keep in contact with the host to see what he thinks.

 

[tcmtech]

If it was me and I wanted to get to 20 KV at 60 ma (assuming its a DC output) I would start with four off the shelf 7.5 KV 30 Ma neon sign transformers connected in parallel and use a simple voltage doubler circuit on their combined outputs.

As far as battery backup goes a good stout UPS unit in front of them would do the job just fine.

 

[DerStrom8]

If it was me and I wanted to get to 20 KV at 60 ma (assuming its a DC output) I would start with four off the shelf 7.5 KV 30 Ma neon sign transformers connected in parallel and use a simple voltage doubler circuit on their combined outputs.

As far as battery backup goes a good stout UPS unit in front of them would do the job just fine.

Voltage doublers change the arcs into sparks. The host is really looking for a solid arc. I believe the transformer has a built-in rectification diode, as a high voltage AC output isn't usable in CRT applications.

I think the point of the competition is for the host to get the best out of what he has. He knows he'll have to make compromises, which is why a lot of the specs aren't firm. They're more like a set of guidelines that show the perfect project. The closer you can get to them, the better, though it's obvious it won't be perfect. I'm rewording the first post to make this clear.

Regards,
Matt

 

[DerStrom8]

The host says he's open to anything. The design does not have to be limited to the flyback he originally described. He has suggested the possibility of toroid transformers as well, since they have some advantages over the flyback. Again, he's open to anything. He would however like for the whole thing to be made from off-the-shelf parts. Avoiding specialty components would be preferable

Regards

 

[tvtech]

old tvtech is watching this thread. Maybe I can learn something too

Regards,
Me

 

[tcmtech]

What exactly is he wanting this device to do?

knowing the application makes a world of difference.

If its just purely for HV arc work then getting 20 KV at 60 Ma is just a matter of pairing up two 15 KV 30 Ma neon sign transformers and then over driving them with a slightly higher input voltage and frequency.

Taking a normal 120 VAC 60 Hz input and running it with 160 VAC 80 Hz input would easily get him the 20 KV he wants without severely over driving the transformers. Toss a set of HV rectifiers on the output and he is back to having DC again.

 

[DerStrom8]

I think the reason he wants to avoid using NSTs is because they tend to be too bulky and heavy. He's looking for something smaller, preferably with some smart functions (he wants it to have variable voltage/current, etc). I'll let you know what he says about it when he reads your reply.

 

[tcmtech]

To get variable current and voltage he is going to have to use a inverter drive of some sort and to get to 20 KV at 1200 watts continuous duty this thing is most likely going to need a custom wound transformer regardless of what frequency it's running at and none of this is going to be overly small or cheap.

As far as variable voltage and current goes with NST's that's easy. Just use two variacs. One is the voltage control and the other is wired in series with the NTC transformers primaries to work as a variable inductance type current limiter.

Standard old school Tesla coil driver circuit tech.