Caps leaking DC

[C Moore]

I thought it was possible to just lift the end of a couplinbg cap where it was undesireable to have DC, and check that lifted lead to ground. But from RG's sight I have "copied and pasted" instructions (in quotes) I was not aware of. So I gues I have two questions:
1. I need to use a 1M resistaor when using my DMM?
2. Is there a difference between chassis ground and "signal" ground?
Thank You

"It is possible to unsolder the grid end of the coupling capacitor, turn the amp on and measure the DC voltage from the unconnected end of the capacitor to ground with an analog voltmeter. If you only have a Digital Multi Meter (DMM), connect a 1M ohm resistor from the free end of the capacitor to signal ground, then measure the voltage across the resistor. If this voltage is even 1V, the capacitor should be replaced; the capacitor is leaky. This is a definitive test, but you do have to have the amp open and powered on, and so it is a more hazardous test. "

[Torquil]

I would say that you do need the 1Mohm resistor. Otherwise the open cap end will reach a high voltage no matter how good it is (unless the "leakage resistance" of your cap is infinity, which is impossible) Also, the answer would otherwise depend on the input impedance of your voltage meter.

Say your previous stage plate voltage is 200V. If you get 1V across the 1Mohm resistor, the leakage current is approx.

I = 1V / 1Mohm = 1 uA

Since the voltage drop across the cap is approx. 199V this means that its DC resistance is approx.

R = 199V / 1uA = 199Mohm

What "RG" (I don't know who that is) is saying is that this is not enough. So if you want to follow RG's recommendation, the voltage across the 1Mohm resistor should be lower than 1V.

I think you can use either "signal ground" or "chassis ground" in this test. The difference should be negligible in this test unless there is something wrong with your amp.

[jjman]

I've used my digital (Fluke) on about 40 caps or so and don't find that the DC drifts up during the test (on good caps.) Most caps test under a volt with 200-400vdc on the other end. I had one (old astron) that had about 90vdc and 5 more astrons with well over a volt. One blue drop from a Fender with a few volts. All brown turd Fenders well under a volt.

[C Moore]

Thanks for the replys.....
RG's instruction was specific to a coupling cap on the control grid of a tube. Does his method apply to all/most instances where a guy working on a guitar amp wants to check for DC coming through a cap? That is to say, to check leakage on any coupling cap, do I ground the lifted end of the cap with a 1M resistor?
Thank You

[tubeswell]

Using RG's method, you are looking to measure a voltage drop across the 1M resistor-to-ground that you are hooking up to the 'free' end of the cap (i.e.: if there is any DC present on the end of the coupling cap it will cause a voltage drop across the 1M). But that other way is still fine for checking you have DC on a cap which shouldn't be there. (See Fig. 4-4 on P 57 of the Darr Book)