Balancing resistors for series capacitors

[Shoegazed]

Maybe a basic question, but Is it possible to read balancing resistors for series capacitors “in circuit” or is it necessary to pull them to get an accurate reading with a multimeter?

Thanks for your time!

[martin manning]

Best to lift one end, then measure.

[GAStan]

If you are measuring the resistance of the resistor then isolate at least one side, otherwise the cap will throw off the meter reading.

If you are checking voltages leave the circuit complete.

[Shoegazed]

Thanks so much!

[Stevem]

Why are you in need of testing them, because of a precived problem with the amp?

[R.G.]

The devil makes me want to say that if you need to be measuring them, and really ought to unsolder one end, why not unsolder >both< ends and put in new, tested, known-good balancing resistors? Resistors are free compared to the cost of your time.

[martin manning]

Resistors are free compared to the cost of your time.

Unless you don't have the right value available...

[R.G.]

Resistors are free compared to the cost of your time.

Unless you don't have the right value available...

That's the thing - it's really, really hard to say that one value is right, with a solid rationale for it. Electro cap leakages are ill-defined, to say the least. The manufacturers will only say that it's less than thus-and-such after some aging processesm, etc., implying that it might be zero or nearly so, and that the cap is "bad", needing replacement if it goes over. And guitar amp folks are notorious for re-forming electro caps because that lets them keep the original parts, however leaky the re-forming process, often done in place, happens to be.

My approach (which may and probably is not perfect!) is to decide how much power I'm willing to waste in balancing resistors, then pick the value that the voltage and that amount of dissipation implies. As a rough guide, on a 500V supply kept split with 250V across each balancing resistor, a 220K balancing resistor would dissipate V^2/220K = 0.284W. A half-watt resistor would be OK.

Depending on how leaky a cap I wanted to be sure of holding in check, I might go with 110K 1W at just about 1/2W wasted on the resistors, or 470K if I thought the caps would neither/never get down to leaking at a rate of 4.7M. This implies my rule of thumb that a swamping resistor needs to be a factor of ten from the real resistor I'm trying to swamp.

The “right” value is then some resistor that you have two of that you can afford to spend the heat and electricity on, and which are 1/10th or so of the lowest leakage resistor in a cap that you think you’ll get. So the right value is easier to come by. “Original” balancing resistors are not necessarily the perfect values.

[nuke]

If it is a Fender Blues/HotRod series, just replace them. No need to test them.