3
\$\begingroup\$

I need to block a DC voltage on a microphone line, and a bit of research shows that MLCC is a bad idea because they can mechanically ring from vibration.

Question:

  • What type(s) of capacitor could be used on a microphone-level line to block DC?
  • How should it be sized?
\$\endgroup\$
4
  • 2
    \$\begingroup\$ Do you really need to avoid a microphonic cap in a microphone? \$\endgroup\$
    – tobalt
    Commented Jul 12 at 5:55
  • \$\begingroup\$ @tobalt It's not for a microphone, but for feeding audio from a device instead of a mic. \$\endgroup\$
    – Justme
    Commented Jul 12 at 6:35
  • 1
    \$\begingroup\$ @tobalt I would say so, yes - you want to have full control over where the audio signals are generated - generally in the region around the mic capsule where it's sensitive. Not everywhere else. \$\endgroup\$
    – MrGerber
    Commented Jul 12 at 7:51
  • 1
    \$\begingroup\$ @tobalt, updated link with the related question that Justme referred to. \$\endgroup\$
    – KJ7LNW
    Commented Jul 12 at 18:44

2 Answers 2

3
\$\begingroup\$

You need to know the frequency range you want and input impedance.

If this related to the radio of your previous question, then the following is true:

The mic input has 8 VDC bias supply and the unit is calibrated to transmit known RF power with 30mV 1.5 kHz sine wave applied to the mic input.

And for the internal operation, the mic DC bias is removed internally with C2123 which is a 2.2 uF capacitor and the input impedance to the op-amp is R2124 which is 3.3 kohms. The impedance to supply is more complex but for simplicty let's just consider the 1k bias resitor R2125.

That makes the input impedance 767 ohms, which is close to the rated 600 ohms in the manual.

The existing capacitor and resistor forms a high pass RC filter with cutoff of 22 Hz.

So, assuming you need to pass the same 22 Hz into a 600 ohm impedance, you would need a 12 uF capacitor.

But as always with RC ciruits, two RC filters feeding each other will alter the total cutoff freqency.

A common rule of thumb to avoid phase shifts is to just use 10x lower cutoff which means 10x larger capacitor. However a 120uF capacitor would be a bit excessive for this purpose, as the internal circuit anyway has the 22 Hz filter.

So that approximately defines the range of any normal audio circuit, but for a lo-fi solution such as this where low frequencies are not important for speec, you can put even smaller capacitor.

The internal capacitor is ceramic Murata GRM series anyway, and microphonics from vibration has not been a huge issue for years.

But since the capacitor needs to be maybe 10 to 22 uF, try a basic electrolytic capacitor, with at least rating above 8V with a good marging, like 16V to 50V range. The cap needs to handle 8V on positive terminal even if your mixer outputs the most negative DC voltage it possibly can if it has develops an issue.

\$\endgroup\$
2
3
\$\begingroup\$

Film capacitors are not microphonic and do not vary their value with applied volts, in other words they are linear. Polystyrene capacitors are good and expensive but I have not needed to use them. Polypropylene capacitors are also very good and not too expensive.

The capacitor value depends on the low cut frequency and the system impedance. So if 3 dB at 16Hz is fine then 100 nF 100 kΩ. 1 uF 10 kΩ. 10 µF 1 kΩ. If your microphone is used for vocals the low cut frequency could be increased to reduce low frequency background noise.

\$\endgroup\$
1
  • \$\begingroup\$ What is the kΩ rating you list with each? Is that impedance at 16Hz? \$\endgroup\$
    – KJ7LNW
    Commented Jul 12 at 20:13

Not the answer you're looking for? Browse other questions tagged or ask your own question.