1) Suppose you want to verify a prediction about the
equivalent capacitance of two or more capacitors
connected in one way or another. This is easy when
a method for measuring a capacitance Cx is available.
Here is one simple method.
2) You must have a reference capacitor whose C is known.
Connect it to a dc source of known voltage, such as
U1=12 V, to fully charge it. Then disconnect it from
the source.
3) Connect the terminals of your fully charged C to the
terminals of the Cx box and measure the new voltage, U2.
The voltmeter must have very high impedance, so high
that U2 remains practically constant for at least 5 seconds.
Pasco electrostatic voltmeters are ideal for the purpose.
4) Convince yourself that the conservation of Qleads to
Cx=C*(U1-U2)/U2. Use this formula to calculate Cx.
Be aware that C should be of the same order of magnitude
as Cx. It is a good idea to have several reverence
capacitors. Use a larger C when U2 is too small, use a
smaller C when (U1-U2) is too small.
5) Do not wait too long before measuring U2, even with
a nearly ideal voltmeter, when Cx consists of capacitors
connected in series. Why is this important? Because the
implied assumption of "no effect from the leakage of
charge" may no longer be valid. This topic was discussed
here in 1994 (?) In most cases several seconds is not too
long but several hours is too long. Several minutes may
be too long for some electrolytic capacitors.
Ludwik Kowalski