Connect with us

Testing capacitors

Discussion in 'Troubleshooting and Repair' started by shady, Feb 12, 2017.

  1. shady

    shady

    43
    5
    Aug 3, 2013
    The method I use to check for faulty capacitor is to check for ESR, impedance or Dissipation factor /loss tangent value according to the datasheet, and if can't find a datasheet then i'll check the capacitor based on the value given by the meter, like the MESR100 does. However, a lot of datasheets have leakage test and I'm just wondering how important is it to test for leakage if you've already checked above values?
     
  2. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,213
    2,695
    Jan 21, 2010
    If the capacitor has a specified leakage and it's in a circuit where this is important, then if you suspect a problem with leakage, yes it's worth checking it.

    Did you say you check for the actual capacitance?
     
  3. shady

    shady

    43
    5
    Aug 3, 2013

    LOL, i forgot to mention i also check for capacitance, the first thing I do.

    This means now that I need to find a meter of some sort to test for current leakage, maybe something like a Applent AT680?
     
  4. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    648
    May 8, 2012
    Yes, my old heathkit LCR bridge provided leakage tests with a rotary switch that stepped the test voltage up to 600VDC.

    Actually, leakage isn't difficult to test. All it requires is a dc voltage source near but below the cap's max rating, current limiting resistor and an Ammeter capable of reading uA. That's just about all modern DMM's.

    Chris
     
  5. Arouse1973

    Arouse1973 Adam

    5,164
    1,078
    Dec 18, 2013
    How are you testing the capacitance out of interest. Do you take into account the DC bias issue with MLCCs?
    Thanks
    Adam
     
  6. shady

    shady

    43
    5
    Aug 3, 2013
    I have LCR/ESR meter - Keysight U1733C to check the capacitance.
     
    Arouse1973 likes this.
  7. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,213
    2,695
    Jan 21, 2010
    Here's one way to test leakage. In place of an electrometer you can probably use a multimeter, but you'll be limited to testing capacitors having fairly significant leakage.
     
  8. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    648
    May 8, 2012
    Steve, that pdf is confusing. On the first page it discusses FiG 1 schematic and then points the reader to FIG 2 that includes a Diode (Ds). The text further discusses the purpose of Ds but it doesn't exist? :confused:

    Chris
     
  9. shady

    shady

    43
    5
    Aug 3, 2013
    Pheww, thought it was only me LOL
     
  10. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    648
    May 8, 2012
    Nope, we're either both going blind or have absolutely no reading comprehension. :D

    Chris
     
    Arouse1973 likes this.
  11. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,213
    2,695
    Jan 21, 2010
    Yeah I noticed that after I read beyond the first part. :-(
     
  12. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,213
    2,695
    Jan 21, 2010
    See page 4-14 of this for the conceptual diagram. It doesn't include the switch that is required for practical purposes, and if you follow the instructions for discharge, it is clear that the diode is between the switch and the electrometer (unless you use a low voltage zener diode). In any case, the switch needs to be in the shielded box because it is effectively connected to the input of the electrometer.

    If you're using a standard multimeter, the noise limiting action of the resistor and diode are not going to be of practical importance because you're measuring such large currents (fractions of a uA)
     
  13. JRUBIN

    JRUBIN

    95
    15
    Jul 17, 2015
    My Heathkit IT-11 uses an older method called Power Factor rather than ESR.....

    For actual ESR measurements Ive restored both a Genrad 1650 and a Heathkit IB-2A and put them toe to toe to see how they performed. Obviously the Genrad is a whole different class, and wins in eqase of use, but the heathkit didnt do too bad.




    Ultimately for DC cap leakage, my IT-11 is most often used, and for the most specific measurements, a uA meter is run in-line of the leak-down test

     
  14. shady

    shady

    43
    5
    Aug 3, 2013
    Been looking for one of these for ages, not easy to find.


     
  15. JRUBIN

    JRUBIN

    95
    15
    Jul 17, 2015
    The IT-11 Was also somewhat short lived and replaced by a similar model that did not allow the switch to remain in the test position because it was too dangerous so they made the new one spring loaded. That means you could not use it to reform capacitors after the upgrade. Granted, I've shown movie after movie why trying to reform capacitors that are 50 or 60 years old is a bad idea anyway, but it was a short lived feature.
     
  16. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    648
    May 8, 2012
    I believe the Heathkit IT-11 termed it Power Factor because that measurement was intended for electro's only. It was a measurement of combined ESR and leakage that manifests itself as real power dissipated by the cap, which also reduced it's effective capacitance.

    I still have mine but it needs new electro's for the high voltage supply. ;)

    Chris
     
  17. JRUBIN

    JRUBIN

    95
    15
    Jul 17, 2015
    You would be hard pressed to find one of my restoration videos that I did not use that device as part of the project
     
  18. JRUBIN

    JRUBIN

    95
    15
    Jul 17, 2015

    http://www.ebay.com/itm/Vintage-Hea...496974?hash=item25e037180e:g:a50AAOSwGIBZszWK
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-