Potential problems with series capacitors?

[d3x0r]

I tried to do a search for information, and am hoping someone can help me figure if this is a real problem...
if I have 4 capacitors in series, is it possible to end up starving the middle capacitors of charge carries?


If I build the system with pre-charged capacitors and mate them (+ -)(- +)(+ -)(+ -)  (just so the ends are + and -) is the overall capacitance still 1/(c1+c2+c3+c4)?  Or because there's a + + junction (and a - - junction) will this affect the overall capacitance?


Or if I charge the whole network from (0 0)(0 0)(0 0)(0 0) to 12 volts so it's (0 3)(3 6)(6 9)(9 12) volts with a 3V drop across each one... if I attach a ground (accidentally or incidentally) to the middle will it affect the overall capacitance if it ends up being (0 3)(3 0)(0 6)(6 12)  as the voltages/drops in the circuit?


If it does affect the capcitance, how can I avoid having an issue?  (very large ohm parallel resistors?)

[MarkE]

Your formula is not quite right.  The series capacitance is the reciprocal of the sum of the reciprocals of each capacitance.  C_SERIES = 1/(1/C₁ + 1/C₂ + ... 1/C_n)

The main issue of connecting capacitors in series is how to avoid blowing them up due to disproportionate charge distribution and leakage.  They can be individually precharged with whatever charge you like.  The problems will only crop up when you connect the two ends of the string together through an impedance.  Balancing resistors are one solution that costs continuous power.  Zener diodes offer some protection at generally lower power.

[d3x0r]

Your formula is not quite right.  The series capacitance is the reciprocal of the sum of the reciprocals of each capacitance.  C_SERIES = 1/(1/C₁ + 1/C₂ + ... 1/C_n)

The main issue of connecting capacitors in series is how to avoid blowing them up due to disproportionate charge distribution and leakage.  They can be individually precharged with whatever charge you like.  The problems will only crop up when you connect the two ends of the string together through an impedance.  Balancing resistors are one solution that costs continuous power.  Zener diodes offer some protection at generally lower power.

(ya just scribbling formula if I ran the math I would have realized)
Okay so it IS an issue.
So these are 2.7V caps, if I put a ... 2.5V zener across each one, it could maintain a equilibrium better... I keep forgetting zeners as a useful part....
hmm... and maybe a shotkey the other way? (for low voltage drop? Or are they just fast?)

[MarkE]

You want something more like this.

[Farmhand]

I discovered the same issue with my supercapacitors, I wanted to put two in series to get a 5.4 volt capacitor but soon realized
that I would need to limit the voltage on each in order to avoid over volting one due to unbalanced charge.
(isn't volting a word ?) Seems not but i'll use it.

I could not think of any way to correct the issue without consuming power except to use a voltage source (battery) across each one.

Which of course is just silly unless the battery is a low output battery (home made cells) and I allow the supercaps to charge from
them until maximum voltage is achieved then use the caps until the voltage is too low then repeat. A circuit could do that
automatically for me and harness the low output of the home made cells via the capacitors intermittently.

Do 5.5 v double layer super caps have some kind of internal balancing ?

..