how do you measure capacitance of a condenser microphone capsule?

[newmod]

Eh Guys,

I was just wondering how would you go about measuring the capacitance of a condenser capsule when its not in the circuit?

I know that the original Neumann U87i capsule was between 63pf and 68pf if I am not mistaken.

Thanks

 

[Bo Deadly]

A DMM that measures capacitance might be enough. If it's too small to measure accurately, you could parallel with another cap of a known value and subtract back.

You could put a cap as close to the same as the capsule in series with the capsule and a put known voltage across both. The supply voltage, voltage at the intersection and value of the other cap can be used to compute the capacitance of the capsule.

Although these methods are purely theorectical. In practice you might need 48V across the capsule, tolerances will greatly influence accuracy and of course if you're using a reference cap you have a chicking-or-the-egg sort of problem. You also have a significant issue of the probes having relatively comparable capacitance.

An AC-based method might be better but it's not clear how the fact that the capsule can vibrate will influence the measurement.

 

[RuudNL]

I use this one. Works for me.

 

[abbey road d enfer]

An AC-based method might be better but it's not clear how the fact that the capsule can vibrate will influence the measurement.

Typical capacitance variation at 94dBspl is about +/- 0.1 10e-3 or 100ppm. However measurements may involve voltages that are higher than that and result in capacitance varying by as much as 1% betweeen the positive and negative peak of the electrical stimulus*. No common AC measurement can reflect that accurately.
However, diaphragm movement results in production of acoustic energy, which is viewed as electrical loss by an LC bridge.

*Note that measuring with a capmeter or LC bridge uses an AC stimulus that results in sin² force exerted on the diaphragm, so always pulling the diaphragm to the backplate, when in normal operation the diaphragm moves to and fro. And of course, the bias voltage exerts a permanent pull on the diaphragm, that increases significantly the capacitance.

AFAIK, capsule mfgrs always specify the inert capacitance. I don't think the variation of capacitance between an inert one and a biased one is worth considering in practice.

 

[RuudNL]

In the past I used a different method to get an idea of the capacity of a microphone capsule:
I used a sine wave generator with an output of 1 Volt / 20 KHz.
Then I measured the signal coming out of the microphone capsule with an AC voltmeter with an input impedance of 1 M.ohm. I made a note of the voltage and replaced the microphone capsule with a fixed capacitor that would give the same voltage on the meter. Maybe not the most accurate method, but good enough to get a rough idea.
(The 1 Volt would never hurt the microphone capsule.)

 

[abbey road d enfer]

In the past I used a different method to get an idea of the capacity of a microphone capsule:
I used a sine wave generator with an output of 1 Volt / 20 KHz.

Supposedly with a resistor in series? In essence it's not too different than what a typical capmeter does.

Then I measured the signal coming out of the microphone capsule with an AC voltmeter with an input impedance of 1 M.ohm. I made a note of the voltage and replaced the microphone capsule with a fixed capacitor that would give the same voltage on the meter.
Maybe not the most accurate method, but good enough to get a rough idea.

Actually, the substitution method has the advantage of eliminating parasitics (probe and leads capacitance).

 

[RuudNL]

The capsule was between the generator and the meter.
So you got a C-R 'filter'. (With the 'R' being the input impedance of the meter.)
In fact what you are measuring is the reactance of the capsule at a certain frequency.

 

[newmod]

thanks

 

[GeorgeToledo]

I happened to stumble on this while checking something else.

Between the point about Neumann capsule capacitance values and the suggestion to test while the capsule is polarized to 48v, I want to add that I happened to read a historical article that says Neumann capsule measurements are traditionally at 60v. They also at some point would test change in capacitance under varying voltage, as an indicator of low end response.