Capacitors & 1/RC: Dangerous Charge Retained?

[SUDOnym]

I heard a point made that capacitors can still retain a dangerous amount of charge in an electronic circuit (inside your computer, say) even when everything is unplugged...this is because a capacitor never fully discharges..
However, I'm not sure if the comment above was made in jest, that is to say, can a capacitor inside a computer that's been unplugged for say a few minutes really contain a "dangerous" amount of charge?

 

[timthereaper]

I would tend to say no, simply because the the time constant RC is a time at which the value is about 63% less than the peak value. Either the capacitance or the resistance must be huge for that time constant to be upwards of 100 seconds. To have a dangerous amount of charge, the capacitor would have to be larger than any I've seen in a computer. A stiffening capacitor for a subwoofer amplifier would, but those things have a capacitance in the 1-5 farad range. But, then again, I'm no electrical/computer engineer.

 

[Nik_2213]

Unless there's a 'bleeder' arrangement, there could be high voltages retained on parts of the switching power supply...

There's another 'gotcha', IIRC: Electrolytic capacitors that have held a high DC charge --smoothing capacitors ?-- may 'soak' some of it into their dielectric and recover it after discharge. This is probably more relevant to 'classic' valve amps, valve-like circuits built with MOSFETs and anything driving electrostatic speakers...

 

[phinds]

I agree with timthereaper. I did have an EE prof once who swore he found an old capacitor out lying in a field and it still had a several-volt charge. We all decided later that we should have asked him if he had ever found an inductor lying around that still had a current flowing through it.

A capacitor with no current bleed CAN hold a charge for a very long time (this was really the prof's point) but outside the power supply I'm sure the capacitors (1) will bleed and (2) don't even start off with a dangerous charge. In the power supply, absent bleeding, it would be possible, as Nik said.

 

[AlephZero]

I can assure you that large capacitors inside power supples CAN store a "dangerous" amount of both charge, even at fairly low voltages (say 50V). I once vaporised the end of a small screwdriver when shorting a power supply cap "just to make sure it was properly discharged" before working on the circuit.

Also, large electrolytic capacitors can appear to "spontaneously recharge" themselves, because charge is "absorbed" into the dielectric and is gradually released over a long period of time.

If a capacitor has been fully charged for sevral hours or days, is then "fully discharged" by shorting the terminals for a few seconds, and the short circuit is then removed, it is quite likely that a voltmeter would measure a few volts across it 24 hours later.

Of course if you left the voltmeter connected for the whole 24 hour period, the meter itself would be providing a discharge path for the capacitor, so you would not see this effect.