I have a bunch of vintage caps I pulled from an organ and I need to test them for DC leakage at their operating voltages. I have two groups of caps, 200V and 400V rated caps and I've created a power supply that generates both of those voltages (well, actually about 10V shy of each).

My power supply takes line level voltage (coming through a sort of isolation transformer: two 12V transformers with the secondaries connected together) and runs it through a two stage greinacher voltage multiplier to create my 200 and 400V. I have two RC filters with 100uF in the first node and 47uF in the second with 4K resistors in between. This has smoothed the ripple to about 60mV on the 400V line (I've actually only been operating it at about 70% with the variac, which is giving me about a 40mV ripple, so I'm extrapolating out)

The problems is, that ripple goes right through the caps and I think it's making my DMM think there's a DC voltage of about 1/2 the ripple on the other side. Or is that really the DC voltage? I'm just trying to determine the suitability of the caps for reuse in tube amps. These vintage caps (Good All and Ajax) are pretty highly sought after (especially the Ajax caps) for their tone in guitar amps, as long as they're not leaking like a screen door. I've got hundreds of them, so I'd like to test and sell some of them.

 

put a resistor in series with the cap, and mesure the dc across it. as the cap charges up, there will be a dropping voltage, the lower it is when the cap is charged, the lower the leakage. resistor around 10k or so.l

 

http://www.angelfire.com/electronic/funwithtubes/Testing_caps.html

 

Thank you both for your responses. Kermit, that appears to have all the details I need. Thanks so much!