Three wire remote microphone

[kathmandu]

Hello,

I need to connect some electret microphons to a CCTV system and I only have (already in place) some one pair STP cables.

The microphones came with a low-noise preamps thus the output is around 1V p-p signal.

I've tried to connect one of the pair's wire to Vcc (12V), the other one to mic output and the ground (shield) to power supply ground but I got a lot of noise. I have a power supply stabilizer and filter on the mic side hence the mic should have quite a clean 5V supply voltage.

I suspect the ground wire (20m long) to be the main culprit, as the parasitic voltage across it is actually summed up with the mic output.

Is there any solution to avoid this? Do I need to run a separate ground wire for the mic output? What about the ground loop, in this particular case?

Thanks in advance for any suggestion.

 

[Nik_2213]

May be a silly question, but do the CCTV cameras have provision for 'Audio In' ??

 

[kathmandu]

Unfortunately, not. Anyway, the microphone and camera locations are not the same (though they are closer to each other than the DVR).

 

[betwixt]

If I understand, you have a single shielded twisted pair so there are three wires in total. It would make more sense to eliminate power supply noise before it enters the cable rather than at the microphone end. Doing it that way would keep noise away from signal along the cable length. Try either regulating the supply at the recorder end or filtering the 12V at the recorder end to keep the whole thing as clean as possible.

If that fails, the other option is 'phantom' power. Not as difficult as it sounds, you carry supply on both the twisted wires and extract it at the microphone end and you return the audio on the same two wires but as in-phase and inverted signals. Back at the recorder, you invert one of the signals and add it to the other. Now you get twice the audio you want and anything picked up along the cable length, which will be equal on the twisted wires, is canceled out.

Brian.

 

[kathmandu]

Thanks, Brian.. that's exactly what I have (a single pair audio-like STP cable).

The 12V supply (DVR side) is also regulated and filtered but it's shared among the three (for now) remote microphones, hence I further filtered and regulated it (stepped down to 5V) on every microphone side.

I've heard about phantom power before but I thought it was only applicable to ballanced output dynamic microphones.

If I understand what you have suggested, the third wire (shield) might not be used at all (or just connected to the ground at one end - the DVR side - for screening purpose only)?

Does it help if I'll have a sepatate twisted pair for the output? How to avoid the ground loop (the output signal ground wire being used as the returning DC ground path, too)?

 

[betwixt]

Phantom power was indeed the normal way of powering remote microphones using a center tapped transformer to produce the out of phase signals but you can do exactly the same thing using simple amplifier circuits these days.

You need all three wires, the shield is still used as the ground return for the DC but both the twisted wires carry DC to the microphone and the signal back from it. You mimic the balanced output using a simple amplifier circuit that has two outputs, one 'normal' and the other the same signal but inverted. when you combine them at the DVR (with another simple circuit) you get this:

Mic - non-inverting amp -----[noise on the cable] ----- Amp+ input
Mic - inverting amp -----------[noise on the cable] ----- Amp- input
Mic GND -----------------------[noise on the cable] ----- Amp GND

As the wires are twisted, they will pick up the same (or almost the same) interference on both wires. In other words the noise is the same on both wires but the wanted signal are upside down with respect to each other. When you combine them you get this:

(Mic+) - (Mic-) = 2* Mic signal (double the wanted audio)
noise - noise = no noise (they cancel out)

Example: https://cschema.blogspot.com/2014/10/balanced-microphone-amplifier-diagram.html
At the mic end, use a resistor from each of the twisted pair wires to combine the DC on them, filter it to the shield with a capacitor and use it to power the mic/mic amp.

Brian.

 

[kathmandu]

Ok, so I will only need an inverter (active) circuit on the mic side? On the DVR side, may I use a passive summing circuit (resistors)?

Btw, could one more twisted pair solve the problem (with no auxiliary circuits needed)? How to avoid the ground loop having two separate ground wires coming from the mic?

 

[betwixt]

Not quite right, you need to invert it on one of the twisted wires at the mic end then re-invert that signal and add it to the other signal at the DVR end. It isn't as complicated as it sounds, the DVR end is just a differential amplifier and almost all op-amps have differential inputs.

What you are trying to do is maximize the signal and minimize the noise. Because the noise is picked up equally on the twisted wires (that's why they are twisted!), when you invert one of the wires and add it to the other you cancel it out. By inverting the audio at source on one of the wires, the same cancellation circuit actually doubles the audio level.

Brian.

 

[kathmandu]

Now it's perfectly clear, many thanks for your detailed explanation. I'm going to build a quick circuit to run a full test on one microphone channel. I'll let you know about the results (or if I'll need further support!).