HV PSU - why isolate

For lab use, an isolation transformer is cheap. It's also cheap insurance against unforeseen faults and the sorts of silly things one sometimes does in the lab while preoccupied with some other debugging problem....

Cheers

Phil Hobbs

like wires come adrift? Other than that, no matter what goes on in the psu the resistor chain keeps the output touch safe. I'll enclose the R chains and output terminals so nothing stray can touch the output end, or naer it.

I'm wondering whether its you or me that's missing something here. The unisolated output is entirely touch safe on both poles, you can stick your tongue on it or stick it across your head and no harm happens. This remains true whatever one connects the other pole to in a moment of foolishness. One would have to connect one pole to an external supply of over 1kV to reach the ratings of the R chain, and the other output terminal would still be completely touch safe.

The current limit Rs are a chain of Rs on each pole, so are safe against a whole series of componenet failures. Even if 60% of the parts fail short, its still all within ratings and touch safe.

I cant see any scenario in which isolation offers any increase in safety, in this particular case. And 85uA is a nonissue with RCDs/ GFCIs.

NT

A sufficiently large resistance will make it safe to touch, assuming that the resistance is actually there when and where you touch it. It depends on your construction practices--handwired protos are flakier than PCBs, and e.g. a resistor string arranged in a loop is vulnerable to shorts. If you do it right, it'll be safe enough--as long as you include the effects of things like wristwatches and loops of fine solder drooping down into the circuitry while you're working on it.

The sort of stupid stuff that happens sometimes when you're cutting corners because you're in a hurry to do something else.

Cheers

Phil Hobbs

Right. I'll put a series of ?4 barriers at the end of the case with the R chains going thru them. Each barrier will have holes offset from the next etc. The barriers will be glued down for vibration resistance.

NT

How about the fine carbon dust that gets attracted to high voltage devices (such as your resistor string)?

I would just ask is there anyone that could sue me if something (anything) goes wrong? If no, then go for it.

tm

How will that protect you if a resistor fails? Some go down in value, or arch intenally, but it's you who will be risking someone's life, not me.

Not a good idea generaly but it depends on where the supply gets its power and what it's feeding. Problems arise over earthing.

ue,

Each pole has a chain of 5 Rs in series, all of which are operated at about 1/3 rated V.

Re risk to life, it wouldnt be any safer if isolated in this case. Not your typical psu scenario.

NT

at

ends

=A0If

ts

Its something that will see few hours of use, its in a sealed case, and V is +- 340v which is no higher than mains (peak on 240v).

Thats a no in this case. Cheers.

NT

Its recified mains

Its a hand held breakdown tester, no earthing will happen.

Thanks, NT

Then you've never seen a whole string of resistors fail, youngster.

rom

value,

ot

How would isolation make it significantly safer in that scenario?

NT

th

from

in value,

not

er.

Please correct me if I'm wrong, but from what I can see,

- isolated or not it would deliver a similar voltage on its own.

- if isolated it could also be connected an external potential and rise higher than if unisolated.

NT

Silly conversation. If you string a couple of cheap 1M resistors from the source to each accessible contact, there will be no hazard.

John

et?

with

set from

wn in value,

fe, not

at

Not

gster.

Yes, as long as the Rs dont all fail s/c. I'm ok with it, and I dont see isolation adding any safety in this instance. Even if they do, the input end current limiting will result in a fairly safe situaiton. It would take a series of faults for a shock to happen, making it safer than a normal isolated low v psu.

NT