10kV here and 10kV there ...

Seal everything and fill with transformer oil...or epoxy if you don't intend it to be serviced. Epoxy has the advantage in not leaking out. Be sure of the ground on the case.

Likely not practical, but does the job quite well. Once past the debug phase, might be worth serious consideration. Alters thermal management, sometimes in a good way, sometimes in a bad way.

For air-filled designs: Avoid sharp edges, points, that sort of thing. Make liberal use of high-voltage heat shrink - the stuff is worthwhile. RTV Silicone, while not pretty, can also be a very useful material. Paper phenolic is a remarkably robust material with good mechanical properties.

Just Say: did some time in the plasma lab, and the laser lab.

Do these helpful repliers to your post not know who you are?

Well, put it in a 4U or 40U chassis. If safety's a poser, then to hell with space 'constraints'.

John how does one email you these days?

Not much worse than an old color TV. I have an old Bertam 30 KV power supply about that size, all air insulated, ancient and dusty, works fine.

Got any pix? Schematics? Did you lose many fets getting the stacked ramp thing to work?

Some sort of streak tubish spectrometer thing?

John

Thanks!

Yep, old color-TV transformers and connections are an inspiration.

I'm working on that.

Not any so far. Fingers crossed.

Hmm. Interesting idea.

On Mon, 5 Feb 2007 01:40:26 -0000, "john jardine" Gave us:

There is no safe airspace design for that voltage. A small, cheaper design, and encapsulation would be the most reliable way to go. Oil bath is good too, but a lot more messy and a lot more controlled, actually.

Rubber (read rubber, silicone, etc.)is cheap (comparatively), and low on the MSDS hazard list... very low. Transformer oil is a major PITA, sealed up case or not.

What you can do is make a driver board which is not encapsulated that pumps a separate, encapsulated multiplier/storage module (high temp cheap COTS plastic case) which IS encapsulated that has the HV output lead(s) coming from it (3 for this design). No messes there. Very low maintenance compared to oil. The driver stage can be center tapped, dual, or whatever you want. You only need to store as much energy as you need to consume, so no need for massive storage banks of caps as that can be unsafe. So you need to do a requirements analysis of what currents you will be using these voltages at.

Bertan also used RTV, but they allways forgot to evacuate the moisture when filling. They had many QA issues, probably the reason they sold out to Spellman.

Cheers

Uh, MP, 500KV in air is almost routine at scientific labs. The power lines over your head peak at >190KV in open weather. I've worked in a lab that had 5MV supplies (not near my work area, though, and there was lots of space).

Yeah, Win's severely restricted space is a consideration (one I'm glad I don't have to deal with, and one I wouldn't presume to offer advice for), but it has also been done many times. I really don't know whether it was done with air, though. I never saw the innards of those things.

Good luck, Win.

John Perry

Avoid sharp edges too! If you must have HV on PCB's then make cut-outs around the HV sections so that anything that wants to flash over has at least to creep up a field gradient. Remember that dirt will accumulate on surfaces to aid in the formation of a breakdown.

The relevant electrical safety standards for "secondary high voltage equipment - haven't got the number, since it is at home" are I.M.E.X. extremely useful to consult: Full of practical advice and useful guidelines for layout and spacing (Surprise, that!!)

Yes, right. Creep distance is the shortest distance along the surface. Apparently 10kV to 20kV per inch is a good minimum creep distance, although I've often seen less.

I'd like to read that document.

Can we not assume that the tubes of GE clear RTV from the hardware store are free from troublesome moisture?

Is there any way you could generate the highest voltages directly on the apparatus that needs them and have only the lower voltages and the control gear in the rack?

Someone asking a not-every-day question ?

Rene

Win, I was rather sucessful in keeping highvoltage stuff far away from me. However recently we opened a TWT before sending it back to the manufacturer for repair. The shipment and the repair are both costly and time consuming.

The operating voltage of the TWT is in the order of 15kV. We found the high voltage divider which was potted in some yucky stuff, to have disintegrated this yucky stuff. Whatever this yucky stuff may have been before became liquid from corona dischanges and leaked out. Could have been rubber, silicone, even polyester or epoxy. IMO, the makers assumed to be save from whatever by potting it and this was not the case. So, I'd lower the max field strength of a setup even when it is potted into a solid or liquid.

Rene

No. It contains acetic acid and will corrode stuff. There are RTV formulations for potting

Regards,

Boris Mohar

Got Knock? - see: Viatrack Printed Circuit Designs (among other things)

void _-void-_ in the obvious place

creep

I'm currently working on a high voltage diff probe and have heard that the standards we have to apply impose around 7mm/kV. Ouch^N!

Fortunately we only have to go to 4kV.

equipment -

spacing

On Sun, 04 Feb 2007 15:54:39 -0800, Winfield Hill wrote: ...

Corona dope:

Cheers! Rich