Author Topic: Testing capacitors in circuit and comparing results with old and new PCB's  (Read 1709 times)

0 Members and 1 Guest are viewing this topic.

Offline xga

  • Newbie
  • Posts: 1

I'm looking at purchasing an ESR meter (something like a Peak ESR70) to check the condition of a number of capacitors on some old PCB's, so that I can work out which ones need to be replaced. 

I understand that some ESR meters (like the ESR70) allow testing of capacitors in circuit and that testing in circuit is not always as accurate as measuring out of circuit as some caps can be in parallel on the PCB.  I wanted to know that if I have a brand new PCB with known good capacitors and an old PCB with potentially old, dried up capacitors, if I was to test in circuit on both the new and the old PCB's and then compare the results between the two, would this be accurate enough to tell me if a capacitor is actually bad or would the best method be to remove each individual capacitor and test them out of circuit?  I'd really prefer to not have to remove 60+ caps from the PCB to test, so was hoping that this comparison technique might work and save me a bucket load of time.

Thanks in advance for any advice you can provide.


Online mariush

  • Super Contributor
  • ***
  • Posts: 3882
  • Country: ro
  • .
Re: Testing capacitors in circuit and comparing results with old and new PCB's
« Reply #1 on: September 05, 2014, 01:07:34 am »
All ESR meters tell you that it is POSSIBLE to test capacitors in circuit, not that it's a good idea to test capacitors in circuit.  It's like saying "yes, you can drive your tesla on country roads/offroad, doesn't mean you should"

What they say by that phrase is that the test voltage is low enough that other components in the circuit won't be turned on or damaged by the test signal. It doesn't mean the results shown will be accurate.

Capacitors in parallel will fool any tester, the ESR value reported will not be correct, same for the capacitance.  Also, quite often large electrolytics are connected in parallel with ceramic capacitors, which have very low ESR and therefore will trick the esr meter into giving you a bad result. 
Bleeder resistors  (connected between the leads of a capacitor to discharge it when power is turned off) will also make the meter measure things incorrectly.

So no, best idea is to remove at least one lead from the pcb.

Now of course,  when you're more experienced you can learn to recognize how some capacitors are connected in a circuit and then you can know it's safe to measure some capacitors right away without being fooled.

If you want just to measure esr for a quick fix, search eBay for some cheap $20-30 esr meter. You don't need accurate results, just something to say "within reasonable values" or "outside reasonable values". What's reasonable, depends on the capacitor brand and series... for that you can read the information in datasheets.

I would not recommend Peak ESR70 simply because it's too expensive for what it does. It's just a more professional looking ebay esr meter. 

For just about 10-20$ more (from $135 on eBay), you can get the Uni-T UT612 LCR meter, which has more features and is a proper tool. You can find review and teardown of it here on this forum :

Mastech MS5308 is also a good choice, but more expensive.

ps. See here a thread with esr meters suggestions:
« Last Edit: September 05, 2014, 01:12:44 am by mariush »

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 10316
  • Country: us
  • DavidH
Re: Testing capacitors in circuit and comparing results with old and new PCB's
« Reply #2 on: September 05, 2014, 01:39:48 pm »
mariush covered most of what I would have said.

If you have a comparison board, then a two terminal curve tracer may be used to check capacitors visually in and out of circuit and might be more reliable.  The old Huntron Trackers were used this way for diagnostics.

With the right design, you can qualitatively see dielectric absorption, dissipation, and multiple circuit elements in series and parallel on an X-Y display.  This is one place where a cheap analog oscilloscope will work better than even an expensive digital storage oscilloscope.

Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo