To the Ineffable All,
I believe there are some facts being ignored in this thread.
1) First, manufactured NP (non-polarized) are different than two regular capacitors connected back to back. This is stated in
https://en.wikipedia.org/wiki/Electrolytic_capacitor under the section "Polarity", which reads "Special capacitors designed for AC operation are available, usually referred to as "non-polarized" or "NP" types. In these, full-thickness oxide layers are formed on
both the aluminum foil strips prior to assembly. On the alternate halves of the AC cycles, one of the foil strips acts as a blocking diode, preventing reverse current from damaging the electrolyte of the other one." That tells you that the manufacturing process is different between the two types.
The same section quoted above also says, "Most electrolytic capacitors are polarized and require one of the electrodes to be positive relative to the other; they may catastrophically fail if voltage is reversed. This is because a reverse-bias voltage above 1 to 1.5 V will destroy the center layer of dielectric material via electrochemical reduction (see
redox reactions). Following the loss of the dielectric material, the capacitor will
short circuit, and with sufficient short circuit current, the electrolyte will rapidly heat up and either leak or cause the capacitor to burst, often in a spectacularly dramatic fashion."
So, that tells me that any scheme to connect
ordinary caps in a back-to-back reverse configuration had better make sure the voltage does not exceed the above (1 to 1.5 volts), and then only for a short time, or there will be trouble.
2) The model of a capacitor shows a Zener diode which represents the capacitor working voltage, and the small capacitor reverse voltage (1 to 1.5 volts) allowed for a short time. It must be remembered that this is not a real diode, but only a model that sort of describes the cap's performance characteristic. The model does not tell you that you cannot put a reverse working voltage across the capacitor. So if you use something like that, either don't exceed the 1 to 1.5 volts limit, or insert
real diodes in parallel or in series with the caps. Just remember, without diodes, all the forward current existing in one capacitor must also exist in the other series capacitor in the reverse direction.
3) Although the model of a back to back reverse connection will show two caps in series. In reality only one one capacitor at a time is functioning as a capacitor, so the capacitance of the unit will be the value of whatever each cap is by itself. It will not be one-half of the capacitance.
Ratch