Announcement

Collapse
No announcement yet.

Faulty brand new filter caps

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Faulty brand new filter caps

    So I purchased and installed 6 new filter caps in my 2103. After the recap I got bad ghosting, which went away when I installed one of the old caps in parallel. So I concluded that one or more of the new caps must be bad. I contacted the vendor and he wants proof that they are faulty. I don't have a high voltage power supply to run a test on them and the best I could come up with is to run a 9v battery in series with my DMM and the cap. I didn't expect to see anything but both of the first 2 caps that I tested passed DC voltage! I haven't bothered to test the others. Curiously one of them managed to generate a 16v reading on my DMM from a 9v battery. Why is this?

    Click image for larger version

Name:	faulty cap 1.jpg
Views:	1
Size:	4.25 MB
ID:	876686Click image for larger version

Name:	faulty cap 2.jpg
Views:	1
Size:	4.44 MB
ID:	876687

  • #2
    I'd say you are measuring some cap voltage.
    It's weird, because it WAS working fine.....

    Comment


    • #3
      I drained the caps first, and desoldered them from ground to isolate them. Thanks for the quick response BTW.

      Comment


      • #4
        I also meant to say: Does anyone disagree that a cap passing DC means it's no good? I can't see how it could be even slightly conductive if it's working properly? I tried the same thing with some new signal caps I have kicking about and - as expected - they behaved like an open circuit and conducted 0v.

        Comment


        • #5
          Originally posted by greengriff View Post
          I also meant to say: Does anyone disagree that a cap passing DC means it's no good? I can't see how it could be even slightly conductive if it's working properly? I tried the same thing with some new signal caps I have kicking about and - as expected - they behaved like an open circuit and conducted 0v.
          To see if a cap passes DC (no good) you best use your Ohmmeter (2M range or higher). Disconnect the cap from everything, discharge and connect your red meter lead to the positive cap terminal and the black lead to the negative terminal. Reading should be out-of-range ( "1." on cheap meters or "OL"). Wrong polarity might give a lower reading but that's irrelevant as ecaps are not supposed to work with wrong polarity. Any polarized ecap will show some DC leakage with wrong polarity.

          A cap that passes this test might still be bad, because
          - considerable leakage develops only at a higher voltage
          - its capacitance is too low
          - it has an ESR (equivalent series resistance) >>1R.
          Last edited by Helmholtz; 05-21-2020, 08:45 PM.
          - Own Opinions Only -

          Comment


          • #6
            Thanks. Oddly on the first cap I tested the positive one side of the can gave me a '1' (I've got a cheap meter lol) and the other side gave me a minus number! This was on the 2M setting. What does that mean?

            Also in reference to my earlier post, is running the cap in series with my DMM and a battery a good way to test for DC leakage?

            Comment


            • #7
              and the other side gave me a minus number! This was on the 2M setting. What does that mean?
              Not clear what other side means. If you reversed meter leads you put wrong polarity voltage on the (still charged) cap. Wrong polarity voltage can damage the cap.

              Also in reference to my earlier post, is running the cap in series with my DMM and a battery a good way to test for DC leakage?
              It's more or less the same as using the Ohm range, which runs the internal battery in series with the cap.
              - Own Opinions Only -

              Comment


              • #8
                Originally posted by Helmholtz View Post
                Not clear what other side means.
                My lack of clarity sorry - what I mean by 'the other side' is the other positive. The can cap has 2 positive terminals (one marked red and the other yellow). From red to negative gave me the 'out of range' reading, from yellow to negative gave me the minus number. Cheers!

                Comment


                • #9
                  A minus reading on an ohm meter means there is voltage in the circuit. There is no such thing as negative resistance. If you measure a cap with a meter, it will usually start relatively low and charge up to a very high resistance or just open reading. Now quickly reverse your probes and you will find a negative resistance reading, due to the voltage your ohm meter left behind.
                  Education is what you're left with after you have forgotten what you have learned.

                  Comment


                  • #10
                    from yellow to negative gave me the minus number.
                    You'll get a negative reading (reverse current) if the cap is still charged to a higher voltage than your (meter) battery. Always make sure that the cap is completely discharged.
                    Last edited by Helmholtz; 05-21-2020, 10:07 PM.
                    - Own Opinions Only -

                    Comment


                    • #11
                      Originally posted by Enzo View Post
                      A minus reading on an ohm meter means there is voltage in the circuit. There is no such thing as negative resistance. If you measure a cap with a meter, it will usually start relatively low and charge up to a very high resistance or just open reading. Now quickly reverse your probes and you will find a negative resistance reading, due to the voltage your ohm meter left behind.
                      That makes perfect sense, thank you.

                      Comment


                      • #12
                        If they are new old stock, they probably need reforming.
                        To do that connect a 1M resistor in series with the positive of the capacitor, both positives in your case. Place a 200volt or thereabouts across the capacitor in series with your resistor. Wait 10 minutes and read the voltage across the resistor. The voltage reading will be mV = uA leakage.
                        Once almost zero on the meter is achieved, run it up at the full working voltage.
                        If almost zero leak is achieved, the capacitor is restored.
                        If no leakage reading could be obtained, whilst charging, then the electrolytic has dried out and no good.
                        This process can take an hour or two.
                        Attached Files
                        Support for Fender, Laney, Marshall, Mesa, VOX and many more. https://jonsnell.co.uk
                        If you can't fix it, I probably can.

                        Comment


                        • #13
                          The voltage reading will be mV = uA leakage.
                          I think you meant 1V across the 1M corresponds to 1ľA.

                          According to ecap manufacturers' (Epcos, Nichicon) recommendations, the reforming current can be up to 5mA and the recommended series resistor is 1k.(With 1k, 1mV corresponds to 1ľA.)

                          If full rated voltage is applied, this method directly shows the leakage current. Typical good values are in the 50ľA range (corresponding to a leakage resistance of around 10M at 500V). While a filter cap leakage current of up to maybe 500ľA would probably go unnoticed in an amp, a leakage current above 1mA after at least one hour of reforming must be considered bad.


                          Lethal voltages, stay safe!!

                          BTW, it's much more likely that the ghosting was produced by low capacitance or high ESR rather than high leakage. It would need a very large leakage to support ghosting.
                          Last edited by Helmholtz; 05-22-2020, 03:50 PM.
                          - Own Opinions Only -

                          Comment


                          • #14
                            I don't think any check with a 9v battery is going to give a verdict on whether the cap is good or bad. Any diagnosis needs to be made at the working voltage. Are they NOS?

                            Comment


                            • #15
                              I don't think any check with a 9v battery is going to give a verdict on whether the cap is good or bad.
                              Well if a measurement with 9V gives a leakage resistance below 500k, the cap is definitely bad.
                              - Own Opinions Only -

                              Comment

                              Working...
                              X