Testing a Capacitor with Voltage for Leakage

Frank, I have a few General Radio bridges in the shop for component measurement, but only the GR 1633-A Incremental Inductance Bridge operates at working voltage. I do have a jig that I can use with an external variable supply to measure caps too large for the 950B by measuring their time constant (and I can also measure leakage with it). For normal antique radio cap testing, though, it is the 950B.

Dale- The 950 is one of my "Go to" machines.
Bought it originally as a kit back in the 60's from Zalytron. I was so disappointed to receive it as an assembled unit.

 I have quite an assortment of the eye types. Some nice and some the flaky in/out of circuit types.

I save my mil RCL bridge for special jobs and my Sencore LC unit- which I home will stay working- bought it from a ripoff company and got taken really badly on it.

Hoping to get time to finish restoration on one of the old eye type SOLAR cap checkers. Nice old wooden cabinet!

FrankB wrote:

So far, the best I have found is the variable output voltage eye type capacitor testers. You can crank them up to the caps max and watch for the eye to close or open. (Note: On electrolytics it may take some time for the cap to properly test if it has to reform).

+1. I have been using an EICO 950B for about 60 years. My DER EE DE-5000 LCR meter is great for, and much more convenient for, many tests, which is why it is in my field kit. For caps in vacuum tube equipment that I often have in the shop, though, you need the type of leakage test that only something like the EICO can do.

Love the Heathkit one. Here's a couple of videos on them I saved for myself.

Jeff Tranter
[You must be registered and logged in to see this link.]

Mr. Carlson's Lab
[You must be registered and logged in to see this link.]

All of us have seen the "nifty hand held" capacitor testers.

 Good for testing capacitance and that's about it.

Why??

 Because they do not apply very much voltage to the Device Under Test (DUT).

Capacitors can test perfectly on the hand held testers, but under higher voltages they can break down and short/leak.

The oldest method of testing a capacitor was to charge it and put a pair of head phones across it and listen for a "click" denoting the discharge of the capacitor
The louder the "click", the greater the capacitance. It took a lot of experience to determine if a cap. was any good or not by the loudness of the "click".

Then folks tested them by putting a multimeter/VTVM across the capacitor and looking for resistance/ leakage. All well and good, but again, minimal voltage across the capacitor, again not real accurate, unless the DUT was leaky at low voltage or dead shorted.

Then someone came up with the brilliant idea of hooking one end of a cap. to a power supply and measuring the amount of voltage leaking through it to ground. Great! Excellent test.
Drawback was you needed to have sources of DCV that came close to matching the cap's voltage rating.

I have seen caps that tested perfectly OK at 200V then shorted out at 250V and up with a rated voltage of 600V.

So far, the best I have found is the variable output voltage eye type capacitor testers. You can crank them up to the caps max and watch for the eye to close or open. (Note: On electrolytics it may take some time for the cap to properly test if it has to reform).

They made a couple of models of "In circuit/ out of circuit"  eye type testers. All they did was you had to hook them across the cap, and hit the test button.  Yes, if the cap was being tested out of circuit and was very leaky or shorted, they would indicate bad. Again the applied voltage issue. In circuit it was a crap shoot. Parallel devices gave false readings. Useless for accurate in circuit testing.

Personally, I like the old eye type testers that Eico, Heathkit, Solar, and a few others made. If you are ever fortunate to find a U.S. military RCL test bridge- grab it right then. Few and far between, and great instruments.

The eye tube types are really simple to fix and check calibration on. Plus they have a hidden bonus feature that few know about!

    Since they have built in current limiting, you can place a Zener diode across the test terminals with a voltmeter across the Zener and slowly crank up the voltage on the eye type tester.
 When you hit the voltage of the Zener, the voltmeter will show a fixed voltage if the Zener is good. If the voltage does not show a fixed point of voltage on the meter as you increase the voltage, then one of two things: One The voltage is not high enough to hit the Zener point, or Two: The diode is bad. You can look up the voltage rating of the Zener to verify its voltage- and beware, sometimes the Zener is mismarked, or has changed its characteristics. I have even seen the wrong Zener put in at the factory! How the unit passed QC is a mystery....

 Pretty nifty trick for testing those packs of unmarked, assorted Zeners you bought a long time ago, huh? This will NOT tell you the wattage rating of the Zener, just the voltage.

I used this on a stereo decades back and found the Zener HAD changed it's "knee" point. Replacing with the correct voltage and wattage Zener restored the set to proper operation.