Capacitor nominal values versus measured

[negativ3]

Hi all,

During repair of a bt keyboard recently, I noticed the BMS circuit used an electrolytic on the input side. It did not show any signs of degradation at all but just out of curiosity, I removed it and measured it. Labelled capacitance was 470uF @ 10V but it only measured 20.8uF so I replaced it.

Was this a sensible move, and was it a lucky catch?

The BMS chip is a TP5000 and the keyboard uses USB to charge.

Thanks
Andy

[Aheld]

Hi, can you tell us a little bit more how you measured this value?
Br. Axel

[Swake]

Aheld has a good question, but I'll assume you used a multimeter to measure the cap and that you measured the new cap to compare.
Does it charge now? Because if that keyboard is working again you likely found the issue replacing that cap.

Lucky catch or 'electronics intuition'. Call it what you want, the fact is you tried to repair the keyboard with the knowledge you had and that made you take a decision. Turns out that you at least found a dirty cap.

[negativ3]

Hi guys, I used an LCR meter.

The keyboard is less than three years old and there was no sign of a dodgy cap and except for curiosity of the only electrolytic, I guess I got lucky.

[MathWizard]

I'm not sure the real reason why, but my LCR meter, the higher the frequency I test at, the lower it reports the capacitance. So that doesn't sound too weird. I'm sure I've measured 100uF caps at 100kHz, and the meter, however it calculates it, would say 10-20uF, or something like that for larger caps. But at low freq, allowing it to charge, like with a DMM, they would read within tolerance, which is often like +/-20%.

So I usually ignore it, and just go by the ESR at 100kHz, and dissipation Factor at 1kHz (or 120Hz for that I can't remember)

[wraper]

With LCR meter you should primarily measure ESR or resistance/impedance (at AC current) because capacitance often looks fine but capacitor is already dead. Also measuring in resistance mode usually works fine in circuit.

[Aheld]

There is a very good application note from Wuerth Electronics, related to capa value and frequency and also which capacitor for what application and why. I‘m not sure if it is also available in english.

So, if you say measurement frequency at 100kHz then you wouldn‘t get the expected 100uF. Reduce frequency to 100Hz then it will show you 100uF.
The relation will be listed in datasheets of capacitors.
For me the behavior of the elko looks fine.