high voltage safety

Here in Europe that's down to voltage & insulation

30mA. Don't ask where the data's from.

You've not really made clear where the safety question is. Are you having HV loads that plug into this HV supply? What area exact;y is the concern?

NT

On Thu, 14 May 2015 16:50:42 -0700, John Larkin Gave us:

Use a switcher and C-W multiplier stage. A flyback is bad.

if it is continuous worry about current and stored energy. If it is a controlled pulse, not much need, unless it is medical and human contact is expected or a possibility. 1kV isn't much as far as HV supplies go.

You failed to mention just what the expected current is, and the load type is also a factor.

I made 15kV supplies which get inserted into the mouth, and they clamped at 11uA. Several uA below the medical human contact spec.

The supply you want can be purchased off the shelf.

flyback based supplies are typically far less efficient.

On Thu, 14 May 2015 16:58:11 -0700 (PDT), snipped-for-privacy@gmail.com Gave us:

You ain't real bright.

That is across the arms through the skin.

Fibrillation can occur at 2mA with direct blood contact.

Open heart direct contact defib paddles are around 2mA pulses, whereas through the chest wall is far higher.

Human contact HV medical devices get controlled at the microsecond shut-off range and at the uA level.

Wow, I have no idea. I've had ~5-10kV across my finger @~uA... kinda like a bee sting, maybe? That felt a lot different than 240 AC, which kicked my ass across a room... once.

(measure twice and cut once... perhaps I should cut that down. George's rule is measure once. :^)

George H.

I meant flyback with a transformer, not a simple boost. And a few multiplier stages would make sense. Something like one of these

That would work, if it's not real expensive. What's the circuit topology? Given my low power, efficiency doesn't matter. [1]

I got some quotes from the usual suspects, and they wanted $200-300 for a HV brick. That seems like a lot.

Besides, I haven't done any HV in a while and it might be fun.

I did this, low power at about 1 KV, a long time ago...

[1] a competitor proposed using Apex HV opamps on huge heat sinks. About 40 watts static power dissipation per channel. Big rackmount box with fans.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

I guess the more serious issue will be the joules of energy storage that might get out if a fet fails or something.

Well, the customer wants me to address safety issues. I guess we don't want prying fingers to touch HV (even inside the box, with interlocks and stuff) so a HV rated connector on both ends (my pulse driver, the optics assembly) might be prudent.

This is essentially a charity job, to help the guys who make the electro-optical thing. I can't get involved in CE standards and test labs and stuff. All I can afford to do is make it reasonably safe and ship it to the end-user to take over from there. One HV CE compliance consultant would double the budget (unless I can get him for burgers and beer.)

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

On Thu, 14 May 2015 17:33:19 -0700, John Larkin Gave us:

Probably a couple of 3904s driving a miniature transformer into a two or three stage C-W. No flyback needed. Use ALL of the waveform.

The transformer secondary turns count determined final output voltage as they are (were) not regulated. (Those in the link may be.)

They initialize and produce virtually instantly, and adding an HV cap increases stored energy when and where needed, and a nice carbon comp bleeder clamps it back down after power is removed.

My 15kV unit was far more complex and used a switching chip (set)

One could fit about 6 of them into a 100mm cigarette pack. The multiplier section was about one inch square by 5/16 thick (potted box). All SMD that one.

What topology is that? Got a schematic?

--

John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Bleeders, 60 second discharge? And nice big Danger High Voltage Labels.

Alden

Cheers

Cool. Part of Amphenol now.

--

John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Sounds like something that Bosch, Delphi, or Denso handles every day, in sets of four, six, or eight...

Alden? (Found in Sam's Laser FAQ.) Part of Amphenol. They say some of their stuff is good for up to 40 kV DC. 3 circuits max.

Anderson? Their Powerpole stuff is only good for up to 600 V though, they say. It tends to get used on high-current DC things, like big batteries. You can stack some of them up to get as many circuits as you want.

Standard disclaimers apply: I don't get money or other consideration from any companies mentioned.

Matt Roberds

On Thu, 14 May 2015 19:38:01 -0700, John Larkin Gave us:

I posted one a few years back and you pissed and moaned about it being on a black background, instead of actually examining the circuit.

Since it is in my archives, any future posting of it would similarly be the same.

Pretty sure it went into a.b.s.e.

If you can keep the output voltage below 1000 Vac or 1500 Vdc, in Europe, it is covered under the Low Voltage Directive (LVD). Many other countries use the same IEC based limits between low voltage and high voltage.

Since you intend to take out the voltage from the equipment, keeping the output within the low voltage area simplifies things a lot. Having devices with outputs in the high voltage area may require extra regulations for building, installing and using such devices.

I get it. No schematic. No surprise.

--

John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

What were these power supplies for?

I recall it being unreadable. And that it wasn't actually your design. If it had been yours, you could have reposted a better image. You poached it from somewhere.

Either of my daughters would send you away wimpering. Heck, my granddaughter could almost do that.

Be brave: post a high voltage supply schematic of yours that we can discuss. Or be a wuss and don't.

--

John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

On Fri, 15 May 2015 15:18:25 +0000 (UTC), Cydrome Leader Gave us:

Before the e-cig craze (which is merely piezo atomization, btw), we developed a supply which creates a "perfect smoke". I spent nearly three years on the development.

It was/is used in an inhaler which immediately showed the need for one fourth the amount of the medicinal media to get the same job done, as it gets the material to the actual site, as opposed to merely spraying micro-droplets onto the back of the throat.

It was curing some forms of TB, and had greater efficacy in cancer drug applications as well.

Den fredag den 15. maj 2015 kl. 19.00.55 UTC+2 skrev DecadentLinuxUserNumeroUno:

no it's not, it's a heater and a wick

-Lasse

On Fri, 15 May 2015 10:10:09 -0700 (PDT), Lasse Langwadt Christensen Gave us:

Perhaps YOURS is. Most are as I described.