17-01-2024, 11:41 AM
(This post was last modified: 17-01-2024, 11:42 AM by Mike Watterson.)
The insulating dielectric is aluminium oxide. The liquid is conductive. At the start there is no aluminium oxide and a small current causes oxidisation on the positive or anode. That's very thin and a good insulator so the capacitance is high.
With age the oxide degrades and becomes porous, so reforming is a voltage at a limited current till there is less than a microamp.
If you put AC or a reverse voltage the oxide dissolves in the electrolyte which decomposes producing gas and the part explodes. If it was current limited DC the capacitor might reverse polarise.
These seem good:
Wikipedia Aluminium Electrolytic Capacitor
and
Wikipedia Electrolytic Capacitor
However it doesn't explain bi-polar electrolytics, which are rare outside of loudspeaker crossover networks. They seem to develop high ESR so I often replace them with larger metpoly types.
I've never understood how they work and didn't care to find out. I suspect they can't be reformed if the oxide is created at manufacturing time by an anode layer between two cathodes, so there is in a sense a third wire not brought out. Two series electrolytics back to back with diodes is likely longer lived.
I had a quick search and I don't find any sane info on unpolarised/bipolar electrolytic caps than what I imagined.
If the foil for the anode is etched rather than plain you get a larger surface area and more capacitance. That might not be a good idea for higher voltage parts.
They don't work well as a diode, and not for long at higher current without exploding.
With age the oxide degrades and becomes porous, so reforming is a voltage at a limited current till there is less than a microamp.
If you put AC or a reverse voltage the oxide dissolves in the electrolyte which decomposes producing gas and the part explodes. If it was current limited DC the capacitor might reverse polarise.
These seem good:
Wikipedia Aluminium Electrolytic Capacitor
and
Wikipedia Electrolytic Capacitor
However it doesn't explain bi-polar electrolytics, which are rare outside of loudspeaker crossover networks. They seem to develop high ESR so I often replace them with larger metpoly types.
I've never understood how they work and didn't care to find out. I suspect they can't be reformed if the oxide is created at manufacturing time by an anode layer between two cathodes, so there is in a sense a third wire not brought out. Two series electrolytics back to back with diodes is likely longer lived.
I had a quick search and I don't find any sane info on unpolarised/bipolar electrolytic caps than what I imagined.
If the foil for the anode is etched rather than plain you get a larger surface area and more capacitance. That might not be a good idea for higher voltage parts.
They don't work well as a diode, and not for long at higher current without exploding.
