Resistor voltage ratings

[Paul P]

I'm now shopping for resistors and have come up with a question :

Am I correct in believing that the "maximum working voltage" specification
for a resistor refers to the voltage drop across it, and not the actual
voltage applied with respect to ground ? I ask this because most 1/2 watt
resistors are rated at 300-350v (or less) and resistors in amps can have
voltages a lot higher than that applied. (I did find some military resistors
at Mousers that could handle 500v but you have to buy a minumum of 200
pieces.)

I don't see any places in my schematics where the voltage drop across a
single resistor would be over 300V.

[Ray Ivers]

Paul,

Did you post about this recently at the old Ampage site? Maybe it was someone else... at any rate, the voltage rating is across the resistor, not to ground (unless one end of the resistor is grounded, of course). A 350V resistor could be suspended in mid air - nowhere near a ground - and have 3,500 volts applied across it, and it would flash right over.

Do you have some resistors that are in intimate contact with the chassis or some other grounded component, and you're worried they'll arc?

Here's a link to Digi-Key's metal film resistors - minimum quantity 5, the 2W are rated for 500V and the 3W for 750V, and they're small - let's see if the 'Insert Picture' button actually works:

http://dkc3.digikey.com/PDF/T062/1294.pdf

If not, the link is http://dkc3.digikey.com/PDF/T062/1294.pdf


Ray

[Paul P]

Did you post about this recently at the old Ampage site?
Not me. I've mostly moved here now. Lots of room :)

Do you have some resistors that are in intimate contact with the chassis
or some other grounded component, and you're worried they'll arc?

No, I'm just getting ready to place a couple of orders to stock
up on amp parts and came across the maximum voltage specification
which I'd never thought about before (I have some limited experience
in solid state and digital stuff - voltage was never an issue.)

Just wanted to make sure before I bought a few hundred resistors.

Thanks.

[Carl / Zwengel Amps]

Yup, you've got it pretty much correct. The voltage rating is ACROSS the resistor. By the way it works the same way for coupling capacitors! I think Enzo brought this up in passing over on the old Ampage but it's worth repeating....

The way they manufacture metal and carbon films is that they cut a spiral in a coating of metal or carbon film whch creates something of a faux wirewound resistor. And at higher voltages these siprals can arc over and causes the resistor to fail.

Carbon comps are a little different in that they use a slurry of carbon based material that is formed into a resistive paste and compressed to make the body of the resisor. There's really nothing to arc over here which is why you generally have a higher rating on carbon comp's than films. However these fail as a result of the voltage potential "punching through" this material in a similar way that voltage can punch through insulation on wires. Also this material can dry up over time and get overly brittle and develop microscopic fractures in the body. This causes a momentary discontinuity and is why older comps tend to snap, crackle, and pop.

As an aside, if your margins are tight you really should account for the AC component as well. For instance, if you have a resistor hanging off a 300 volt node and it has a 100 volt AC signal riding on it, the AC is going to increase and decreas the value of that 300 volt node by about 140v at the peaks and troughs. So in this particular case, you're fine with a 350v rated resistor in steady state but during operation the resistor would fail! In most cases it's not a problem but it's still something to keep in mind when you're selecting components.



-Carl