Electronics > Beginners
Inexpensive Capacitor Leakage Tester
<< < (3/5) > >>
Old Printer:

--- Quote from: jsi on February 19, 2018, 05:30:39 pm ---
--- Quote from: Old Printer on February 19, 2018, 03:45:20 pm ---I have a couple Fender tube amps from about 1972 so this interests me. Are you considering replace original caps with new old stock and therefore want to test them? I know good quality large caps can be expensive, but for the work involved that would all I would consider using and would trust the mfgs QC. I would be interested to learn if that is a naive thing to do.

--- End quote ---

I bought the replacement caps from justradios.com  He's got just about everything you could want for these old tube amps.  I never considered putting new old stock (NOS) in my projects.  Lots of stories about them being bad too.  The explanation that seemed to make the best sense to me was it's like replacing the tires on a classic car.  Nobody wants to drive around on leaky old tires, and the same thing goes for old capacitors.  YMMV, but IMO a cool working radio is far more fun than a pretty box that doesn't do anything.

--- End quote ---

What I meant is I would trust QC on good brand caps from current manufacture (i.e. Sprague from DigiKey). NOS maybe fine for tubes, but old caps are old caps, in circuit or in the bag.
Zero999:
It's easy to measure the leakage of a capacitor, with a multimeter and stopwatch.

Charge the capacitor, to the appropriate voltage.

Leave it to stand, with nothing connected to the terminals, for a known amount of time.

Set the multimeter to read the maximum voltage, connect it to the capacitor and record the reading.

Calculate the lost charge and work out the equivalent parallel resistance/leakage current.
jsi:

--- Quote from: Hero999 on February 20, 2018, 01:56:28 pm ---It's easy to measure the leakage of a capacitor, with a multimeter and stopwatch.

Charge the capacitor, to the appropriate voltage.

Leave it to stand, with nothing connected to the terminals, for a known amount of time.

Set the multimeter to read the maximum voltage, connect it to the capacitor and record the reading.

Calculate the lost charge and work out the equivalent parallel resistance/leakage current.

--- End quote ---
The procedure you describe sounds like a discharge test which in my learner's mind is different than a leakage test.  I need to test for DC leakage through the capacitor.  I'll leave the distinction between the two for the more experienced people here. 

What I'm going to do is haunt ebay and Craigslist untill I find a suitable tester.  I'll buy it, use it for two projects and then sell it.  If I pay the right price to start with, I'll be able to sell it for what I paid. So, when all is said and done the tester is free. 
PA4TIM:
http://www.pa4tim.nl/?p=1385 my page about reforming and leakage measurement. At the end of the page some simple schematics from the ones I builded and use (a lot)
There are two different leakage principles. The first one is stated in the manual from the electrolytic caps. Stated for 1 and/or 3 minutes. I have never done those tests so I do not know the right set up for that. I have a WK bridge that has a leakage function but the manual does not explain how to interpret  the result and do the test. It is done with a low voltage.  I think it is some kind of coulomb measurement. After one minute the cap must be charged to the max voltage (3V ? ) from the tester. The charging current then will be zero. If not it is leaking. This is a test for normal formed caps.

This is not the leakage you worry about in old tube gear that is long not used. That has to do with the leakage of the degraded oxidelayer on the working voltage. So a "defect" but often "repairable"  You test that at the working voltage so you know for sure the oxide layer can withstand that. Just for safety. This way you can find defects in the isolation. If have seen caps working perfect at some voltage (far beneath the working voltage)  but  if you increase the voltage, nothing happens other them more leakage current. Almost like it is a zener. (until it heats up to much and then poefffff )
daedalux:
Use a regular 250V 500V 1000V insulation tester in the range it fits. Have one, is very useful for lots of stuff as you need high voltage safely lots of times for testing for shorts. We trust our regular multimeters too much and they inject just 0,3V in resistance mode.
If you don't go on anything that may remotely be dangerous like real power electronics a BM500 insulation tester may work for you.
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod