How to Safely Generate High Voltage for Capacitor Testing?

[Averagesupernova]

The plan is to design the power source such that it is so current limited and has safety features ensuring the voltage does not change even if the cap breaks down.

Do you have any idea how that will work and why another poster said it makes no sense? You are in WAYYYYYYYYY over your head.

 

[Averagesupernova]

You plug one of these into your DVM:
View attachment 349683

I don't remember the probe's input resistance, but if it is $10M\Omega$ then it will draw $350\mu A ~at~ 3500V$


As @DaveE says, that is not how a current-limited power supply works. An arc looks like a short circuit to that power supply, which causes the output voltage to drop out. Otherwise, the supply is not "current limited"...

Used to use one of those working on color TVs. Some chassis's were made so it was difficult to slip that probe under the silicone cap that seals against the picture tube. Larger color TVs had around 30 KV on the second anode under that silicone cap and slipping the probe in under would sometimes cause an arc off to the chassis.

 

[DaveE]

Used to use one of those working on color TVs. Some chassis's were made so it was difficult to slip that probe under the silicone cap that seals against the picture tube. Larger color TVs had around 30 KV on the second anode under that silicone cap and slipping the probe in under would sometimes cause an arc off to the chassis.

Yes, those are very clumsy to use in practice on smaller equipment. I sometimes would unscrew the tip and connect a HV wire to connect to the circuit, but of course that was a semi-permanent set up that had to be done in advance with some care. A lot of that stuff is designed for industrial installations where there's more space.

A small hand-held HV probe is kind of an oxymoron.

 

[DaveE]

its voltage drops but not that should not force the power supply to collapse.

What? Then where is the "extra" voltage drop? Do you have a ballast resistor or such?

If you don't want arcs, why not read about and design an appropriate insulation system for the voltage you will apply, with appropriate design margins, and test to verify without failure at an appropriately higher test voltage?

You can buy or rent insulation testers (hi-pot, megger, etc.), and you can buy low power HV PSs. Building your own isn't really that cheap or easy, particularly if you don't know what your doing.

I think it's time for me to leave this circus. In general, home built projects to apply dangerous voltages to large areas (like 100cm x 100cm plates?) isn't encouraged on PF forums. You need to describe exactly what you are doing, why, and some specific safety practices that will convince me that no one will get hurt for me to continue. It's is clear to me that you haven't researched this enough to continue, and perhaps you don't have the technical background required.

 

[bob012345]

Do you have any idea how that will work and why another poster said it makes no sense? You are in WAYYYYYYYYY over your head.

I'm a non-expert asking for advice on a highly technical and dangerous subject. By definition I'm in way over my head.

 

[berkeman]

Where I experiment has safety people and HV experts so everything I do will be under some kind of supervision with lots of advice and input.

Can you say more about this? If you have experts there, why are you asking us instead of them? I usually recommend that folks find a local mentor to help them with any dangerous type projects, but it sounds like you already have them, no?

 

[bob012345]

Can you say more about this? If you have experts there, why are you asking us instead of them? I usually recommend that folks find a local mentor to help them with any dangerous type projects, but it sounds like you already have them, no?

Well, it's a Makerspace with an electronics lab and a science lab among other things. I'm sure there are HV power supplies there like in our Fusor project for example. We may be able to put thin dielectric films on wafers also. But it's not a corporation and the people who know things are not always there to ask. But I have and will ask them. I'm not going to do something dangerous unsupervised. Why am I asking you all at PF? Because I joined PF to ask questions and this thread is part of my information gathering process. Here, I'm just gathering thoughts on my idea of a possible safer power supply for a specific limited experiment.

 

[berkeman]

Do you understand our comments about how current limiters work on power supplies now?

 

[bob012345]

Do you understand our comments about how current limiters work on power supplies now?

I think I do. Thanks.

 

[berkeman]

Well, it's a Makerspace with an electronics lab and a science lab among other things. I'm sure there are HV power supplies there like in our Fusor project for example.

What kind of high voltage power supplies do they have there? If they have a standard HV power supply with an adjustable current limit, you may just be able to use it and appropriate safety measures to do your HV capacitor breakdown tests. No need to try to cobble together your own supply and have do deal with a bunch of testing on it (which is where some of the danger comes from).

 

[bob012345]

So, if

What kind of high voltage power supplies do they have there? If they have a standard HV power supply with an adjustable current limit, you may just be able to use it and appropriate safety measures to do your HV capacitor breakdown tests. No need to try to cobble together your own supply and have do deal with a bunch of testing on it (which is where some of the danger comes from).

I'm not certain how high they go but I can check. I know we have many DC power supplies but not sure about HV power supplies. But thanks for the reminder to check.

 

[berkeman]

How did you get to the 3500Vdc number? Did you use the Paschen Curve for 1atm dry air and a conductor spacing of 1cm? What is the breakdown field strength of dry air in V/m?

 

[bob012345]

How did you get to the 3500Vdc number? Did you use the Paschen Curve for 1atm dry air and a conductor spacing of 1cm? What is the breakdown field strength of dry air in V/m?

No, not the Paschen curve. Breakdown in air with a 1cm gap might need 30,000V but I'm not going to do that since my ultimate goal is encapsulated thin film caps. The initial experiment will be in air. The breakdown of dry air is 3 million V/m. Initially, the cap is in air with ~1cm gap. The max field will be 10X less than breakdown. Ultimately, the cap will be sealed with no air, with a dielectric of thin kapton or teflon films of 1-2 mills or 25-50 microns. Then the 3500V limit is workable and safer. In fact it might only need to be a few hundred volts depending on dielectric strength of the films. Mouser has thin film HV caps in the kV range.

https://www.mouser.com/c/?marcom=182171537

 

[Nik_2213]

Several such projects, along with their 'obvious', 'less obvious' and oft-subtle but potentially lethal 'gotchas' were featured in former incarnation of 'Scientific American', the one with the 'Amateur Scientist' department.

To put it politely, even if you do not build those now-aged designs, their precautions are 'cautionary'...

Google etc search found searchable PDF compilation on CD-ROM available for $15~~$20...

 

[Vanadium 50]

I'm a non-expert asking for advice on a highly technical and dangerous subject. By definition I'm in way over my head.

You should then reread the PF Rules on this.

As far as I can tell, you don't want to build a C-meter. You want to build an arc flash generator. Have you seen what arc flashes do to people. It is not pretty. Life altering stuff. And sometimes "altering" means "ending".

 

[bob012345]

You should then reread the PF Rules on this.

As far as I can tell, you don't want to build a C-meter. You want to build an arc flash generator. Have you seen what arc flashes do to people. It is not pretty. Life altering stuff. And sometimes "altering" means "ending".

No, that’s not at all what I want to do.

 

[Averagesupernova]

No, that’s not at all what I want to do.

Have you considered the possibility that it is what you WILL do even though it is not your intent?
-
Engineering involves the train of thought that includes the things that could happen even though those things are not the goal.

 

[Averagesupernova]

@bob012345 your approach to this is the equivalent to trying to fly a plane without knowing how to fly and your reasoning is that you'll read a book on how to fly a plane once it's in the air. You don't realize this as you are unable to see it. But it's evidenced by your post #27 showing that your are in over your head. Nowhere have I seen you ask what is wrong with your line of thinking in post #27. It tells me you don't want to do the work of doing this project safely.

 

[berkeman]

Thread closed for Moderation.

 

[berkeman]

So this thread is done now. Bob -- you're only getting so much push-back because probably every one of us pushing back has shocked ourselves and/or started fires in the past, even though we could be called "experts" on stuff like this.

Please do pursue the Maker space, and seek out good local Mentors (but beware of posers). It is good to learn under Mentor supervision. But please also set realistic goals. Wanting to test and learn about capacitors and having that morph into "I want to see what arcing capacitors are like" is not a good theme.

And if you really want to test capacitors to learn more about their breakdown voltages, you should use a standard instrument called a "HiPot Tester", or design your own with exact adherence to the safety standards used in their design. I've used these for decades in my EE work.