Measuring ultra high voltages with DMM

[Alexontherocks]

Dear all. I have a few questions i haven't found answers to.

I am working on filtered 2kv anode voltages and have achieved filtering with series capacitors. Everything works and the caps are balanced. The problem is i have a 500v max dc dmm. I have used it across one of the five series capacitors to estimate (x5) the overall voltage. So i get 400v and estimate that the whole bank sees 2kv across it.

On to the questions:
1 can this method be applied to an oscilloscope?
2 what is the reliability for this kind of method forripple and voltage values?

Thanks

Inviato dal mio GT-I5800 usando Tapatalk

[Mooly]

You can't really use a scope because normally the scope probe ground would apply a "short" to chassis ground across the caps. That is if both scope and equipment are grounded.

To have the scope "floating" is dangerous and can still give unexpected results due to capacitive/inductive coupling.

Safest method is a proper high voltage probe. You can just use series resistors (but you need several so as not to exceed the voltage rating of the resistors).

Somerthing like this. You don't need the frequency compensation for looking at low frequency ripple. The resistor R1 in this example MUST be either a high voltage resistor rated for what you need or made up of several lower values in series.

That would allow you to measure the ripple on the 2KV rail. 2KV isn't ultra high

How to make a 100X oscilloscope probe - a great resource for How To's from Wikia

[sofaspud]

Here is a good reference for high-voltage multimeter probe design. I'm not sure how well enclosures are addressed; I'd probably put it in something like a foot long piece of 3/4" PVC pipe.

[miklos]

high voltage probe | eBay

[miklos]

Or make your own HV probe with the help of these:
high voltage resistor | eBay

[miklos]

You can't divide HV with series connected capacitors, you have to use resistors.

[Alexontherocks]

Thanks everyone. 2kv is quite high for an active device like an amplifier but i get the point. As for the capacitors these all have balance resistors. Results measuring across one cap and multiplying gives just about the exact result from psudII.



Inviato dal mio GT-I5800 usando Tapatalk

[KatieandDad]

Just be careful with NOISE.

I built a HV probe for my oscilloscope, OK it was to work at 40KV but it was useless.

The noise picked up by the chain of high value resistors made it unworkable.

[c2cthomas]

You might want to check around for local HAM radio freaks as a lot of them use the HV stuff in their radio transmitters and thus need measuring gear.

[LafeEric]

I once made a 165KV voltage divider using a bunch of high ohm resistors and one precision resistor.

It was all just ohm's law. I figured out the resistance needed for 30uA @ 165KV and found a high ohm 5 watt 5% resistor to build my ladder. I think the max voltage on each was about 1200V. I measured each individual resistor as I added it and kept track on a spreadsheet. There were a bunch, I forget just how many right now, but because my ladder resistors weren't that special the cost wasn't that much.

The last resistor, which was the one actually measured across, was .1 % @ 2 watt, if I remember correctly. I figured out my overall precision was about 2%, which was acceptable. The bottom of the probe was grounded, which allowed O-scope measurement. Solder joints were all carefully smoothed, and the whole thing was dipped multiple times in corona dope, then mounted in a plexi tube about two feet high.