how fast el. caps charge up?

[yeeshkul]

Can i somehow figure out how fast a 100uF cap charges up when i connect it straight to 9V battery (a filtering cap without any resistor in series to figure out the charging current)? There is something strange going on in my friend's stompbox that reeks of a bad cap and i don't want to change all of them.

[demonstar]

With an ideal voltage source the cap would charge instantaneously but in reality the battery has internal resistance so that limits the current. The equation that applies here if I remember correct is...

Q_t=Q₀(1-e^-(t/(RC)))

The simple rule to remember is that a cap is CONSIDERED fully charged when five time constants have elapsed. The time constant is the product of the capacitance and the resistance.

So let us say that in your battery the internal resistance is 25ohms (I' guessing here because I don't know what value it will be. I don't think it'd be any bigger, much more likely to be a lot smaller.).

T=RC=25x(100x10^-6) = 0.0025s.

5T= 0.0125s

SO you cap will be charged in around 1/80th of a second.

Hope that helps. 

[yeeshkul]

yes it does! thanks

[R.G.]

Just out of an urge to kibitz, how does the charge time help with the debug? This may be a debugging technique that I don't know yet.

My guess would be that the low and variable resistance of the 9V battery would make the test difficult to do properly and hard to read; I'd use an oscilloscope.

I have found it quite difficult to tell about caps in circuit without an ESR meter. This is a meter that pumps an AC current into a cap and measures the AC voltage that appears across the cap. The voltage is small, in the 100mV range so any paralleled semiconductors cannot be turned on by it and give false readings. The meter tells you how much series resistance the capacitor has internally, and this is almost a direct reading of the cap's health. Good cap = low ESR. High ESR = bad cap. My ESR meter was a kit from Dick Smith in Oz.

Some other things that I've done to debug filter cap problems:
- bridge the in-circuit cap with an external BFC (observing polarities); this lets the external cap take over filtering even if the in-circuit one has gone high impedance. If things improve markedly, the in-circuit cap is probably bad.
- lift one lead of the in-circuit cap. Now you can measure capacitance with a cap meter. But you can also put a largish resistor in series with that lead, say 100K to 1M. Apply voltage across the resistor+cap. If the cap charges up to the full voltage, its internal leakage is greater than 10X the series resistor; but whatever voltage it charges to, you can calculate the internal leakage resistance by reading the voltage and solving the voltage divider for the unknown resistor.
- Also with the large resistor in series you can now time the charge-up with some precision. The version of the timing equation that demonstar gave in the charge-balance form can be converted to voltages: V = Vbatt*(1-e^-(t/(RC)). You can now measure the voltage on the cap at certain time intervals and directly compute the capacitance by solving that equation for C. That gets messy if you're not into higher math.

The easy way to do this is to measure the time it takes for the cap to get to 63.2% of the total voltage. That time is equal to the product of the R you have in series and the capacitance, so you can solve for capacitance as C = t/R. The big R stretches things out so you can measure the time without an oscilloscope and also to you know the resistance for calculation.

[frank_p]

I have found it quite difficult to tell about caps in circuit without an ESR meter. This is a meter that pumps an AC current into a cap and measures the AC voltage that appears across the cap. The voltage is small, in the 100mV range so any paralleled semiconductors cannot be turned on by it and give false readings. The meter tells you how much series resistance the capacitor has internally, and this is almost a direct reading of the cap's health. Good cap = low ESR. High ESR = bad cap.


Where can we find "acceptables" values of ESRs for caps.  Are there ranges or tolerances we can compare to as to know if there is too much ESR in the caps we are testing.  I am thingking of a kind of chart (or something similar).

[R.G.]

Where can we find "acceptables" values of ESRs for caps.  Are there ranges or tolerances we can compare to as to know if there is too much ESR in the caps we're testing.  I am thingking of a kind of chart (or something similar).

Look at the front of the meter in this buildit article: http://members.ozemail.com.au/~bobpar/k7214.pdf.

To read about ESR meters, start here: http://members.ozemail.com.au/~bobpar/esrmeter.htm

An analog meter is here: http://midwestdevices.com/pdfs/Capmanual8.pdf  It's worth reading for the additional debugging info.

[frank_p]

Wow ! Cool.  Thanks RG. 
I have a big bunch of old electrolytics to test now.  I've bought them when I didn't know electrolytic caps had a "limited lifetime". 
Thanks again !

[yeeshkul]

Hehe, good stuff, thanks I don't have any method for a charging time helping with debugging, i was wandering how a damaged cap changes its behaviour in general.