Hi! I am trying to make a box that will be inserted between a wireless microphone receiver and a mixer which will unmute the audio once a strong and persistent enough signal is detected. I came up with this circuit but I want to be sure that it can work before I build it. I also want to make sure I understood correctly some things about the mic/receiver signal.

I chose the mosfet design because relays would introduce pops. I chose to short the audio signals in order to keep the (possible?) DC offset and so to avoid clicks (and to avoid passing the signal through silicon). Is there a DC bias expected on XLR lines (coming from the receiver)?

I am mostly worried about the amplifier circuit: I removed the usual 0.1uF decoupling capacitor on the input because 1) I am afraid it will "load" the line(alter the audio?) 2) the signal is expected to be centered around 0 anyway(right?).

From what I read I understood that: the audio signal is indeed centered around 0V but all the references about -40db or -60db refer to the absolute value of the signal aka p-p. Is it so?

 

Interesting, I've never seen anyone use a MC for a Noise-Gate.

I'm guessing this proposed Circuit is Binary, ( on/off ),
this, even when perfectly balanced, will create a very obnoxious "Thump" when Switching.

A better solution to the FETs is called an "Opto-FET",
which uses an LED to turn on the FET, keeping it completely isolated from the Audio.

Any means of "Switching" must be very smooth to be unobtrusive.
Usually all that is required is around ~40db of attenuation.

If you're interested I'll show You how to do this with a single Op-Amp.
.
.
.

 

Thanks for the noise gate reference. I had no idea such devices exist! I think the uc offers a bit more flexibility in detecting human voices (what I intend to use it for).

I also thought to use some pwm to smooth that transition (I just realized I forgot the gate resistors on the mosfets). Would this work?

 

IRLZ44? How much signal were you expecting from the microphone?

Have a look at THAT.
https://thatcorp.com/datashts/dn100.pdf

If you want to use an op-amp, then you can switch with a JFET on the inverting input. There's a section in Doug Self's Small Signal Audio book on it.

If you want to use a microprocessor, then attenuate the signal with a LM1972
https://www.ti.com/lit/ds/symlink/l...l=https%3A%2F%2Fwww.ti.com%2Fproduct%2FLM1972

If you use PWM, you will need to make sure that you filter it out of the output signal.

 

It really doesn't need to be all that complex ...........
A Band-Pass-Filter, ( 1khz ),
into a Precision-Full-Wave-Rectifier,
into a Peak & Hold Capacitor with a ~5-second taper,
into a Current-Amplifier for the LED in the Opto-FET,
is much more simple than it sounds,
and will provide excellent "transparent" performance.

The only unknown is whether the Input will be Balanced-Mic-Level, or Balanced-Line-Level.

Or, You can just buy a very nice Controller from Beringer
with all the Bells and Whistles for around a ~$100.oo Bucks.
.
.
.

 

IIRC, a lot of things just used a regular small signal transistor to mute the outputs. My first Philips CD player had two transistors per channel, presumably because one didn't mute it enough. (Page 25 of the CD150 service manual from Electrotanya).

 

I was thinking of using transistors but won't the CE saturation voltage cause issues? It is much higher than mic level audio. Even DN100 has 0.15V of Vce sat.

 

I was thinking of using transistors but won't the CE saturation voltage cause issues? It is much higher than mic level audio. Even DN100 has 0.15V of Vce sat.

What current is that 0.15V measured at?
I've used bipolar transistors to attenuate audio signals and they do work, even when the collector voltage is negative of the emitter. JFETs are better.

 

An LM358 is never used for audio:
1) It produces crossover distortion.
2) It is very noisy (hissss).
3) It has poor audio high frequency response and will cut all frequencies above 1kHz with the high gain of your circuit.

 

Is it possible the wireless mic already has a noise gate or "squelch" built-in? If you know the model number, the manual may be online.

 

Why not switch a low pass filter between your mosfets, this will give you a pop free solution.

Please use the usual 0.1uF decoupling capacitor on the input from the sense line as it creates a high pass filter eliminating the low frequency rumble that can give you false switching.