Impedance question

[Dr.EM]

Hi. I dont know too much about impedance, but I was basically wondering, what are the implications of using say a 3kohm device with perhaps a 600ohm input. Other than degraded sound quality, are there any more serious issues involved with this?

 

[Nigel Goodwin]

Hi. I dont know too much about impedance, but I was basically wondering, what are the implications of using say a 3kohm device with perhaps a 600ohm input. Other than degraded sound quality, are there any more serious issues involved with this?

You need to be more specific, there are two totally different methods of impedance matching used:

1) Power transfer - this requires both input and output impedance to be identical, and gives the maximum power transfer (by simple ohms law and watts calculations). Used for the output in RF transmitters - by definition the output dissipates the same amount of power as the load receives.

2) Signal transfer - this requires a low output impedance, and a high input impedance - this gives maximum signal transfer, but not maximum power. Used for general audio connections (or most signals?), for example the output impedance of an audio amplifier is probably under 0.1 ohms, and the speaker load 8 ohms. This provides high power in the load, and low dissipation in the amplifier.

As you mention 600 ohms, presumably you're talking about microphone inputs?.

The first thing you need to think about is what the actual impedances are?, a 600 ohm microphone input is not likely to be 600 ohms, most likely it will be higher?. Likewise, your 3K output impedance may well be lower than that?.

You only need simple ohms law to work out that the two together form a potential divider, with 3K at the top, and 600 ohms at the bottom - giving considerable attenuation to the signal. But as a low impedance input has more gain, it won't be as low as you might first imagine.

Best idea is to try it and see what happens, if you're trying to feed a high level signal (out of a CD player or preamp for example) into a 600 ohm input, you should add a series resistor - which will provide a better imedance match, and attenuate the signal down to the correct sort of level.

Many years ago the first Philips cassette recorders only had microphone inputs, to make a line input lead you added a 1.5MOhm resistor in series inside the DIN plug.

 

[Dr.EM]

Yep, this is probably the signal transfer type, and yes, its for mics.

I am already using it, and I think it sounds ok. Basically, the mic input on my mixer reccomends 600ohms, but is actually 3k. The mic I made and am using with it is 2.2k, so I suppose its ok to use. But, say, if the input were ACTUALLY only 600ohms, or less? You say that would attenuate the signal, which I guess would result in more noise, but will it damage anything? The mixer, or mic?

 

[Nigel Goodwin]

Yep, this is probably the signal transfer type, and yes, its for mics.

I am already using it, and I think it sounds ok. Basically, the mic input on my mixer reccomends 600ohms, but is actually 3k. The mic I made and am using with it is 2.2k, so I suppose its ok to use. But, say, if the input were ACTUALLY only 600ohms, or less? You say that would attenuate the signal, which I guess would result in more noise, but will it damage anything? The mixer, or mic?

No, it won't damage anything - it's simpe to understand the principle - imagine the microphone lead has a resistor in series with it (it's source impedance), and imagine the mixer input has a resistor from the input down to ground (it's input impedance). These two resistors form a potential divider, which attenuates the signal.

When you say 'mic I made', what sort of mic is it?, and have you measured it's output impedance?.

BTW, 3K input impedance for a 600 ohm microphone would be the sort of value I would expect. If your mic is actually 2.2K (which sounds suspiciously like a resistor?) the attenuation will give slightly more than half the signal.

 

[Dr.EM]

Thanks for that

The mic I have made is an electret condeser type, and yep, its the resistors value that is 2.2k.

http://www.uneeda-audio.com/pzm/pzm-ch.htm

^first circuit on there, minus the 10k, powered by 3v.

I also wondered, when you switch it on, it makes a fairly loud pop. Would that "surge" be able to damage anything? It goes through that capacitor, so I assume it cannot be DC, and thus it is safe, but just want to be sure

 

[Nigel Goodwin]

Thanks for that

The mic I have made is an electret condeser type, and yep, its the resistors value that is 2.2k.

http://www.uneeda-audio.com/pzm/pzm-ch.htm

^first circuit on there, minus the 10k, powered by 3v.

I also wondered, when you switch it on, it makes a fairly loud pop. Would that "surge" be able to damage anything? It goes through that capacitor, so I assume it cannot be DC, and thus it is safe, but just want to be sure

It won't do any harm, but it's best to keep the volume low when you turn it on.

The output impedance of the mic isn't the value of that resistor - it's probably considerably lower than that - you would need to measure it to find out (which you might not have the capability of doing).

 

[Dr.EM]

Ok, thanks for your help

I don't have a multimeter :shock: , I suppose I should really get one...

 

[Nigel Goodwin]

Ok, thanks for your help

I don't have a multimeter :shock: , I suppose I should really get one...

Sorry, but you can't measure impedance with a multimeter - but you really should get one!.

 

[audioguru]

Hi Nigel,
I have always thought that the drain of a junction FET is a high-impedance current sink, eg a constant-current diode. Therefore the output impedance of the mic with a 2.2K drain resistor is very close to 2.2K.

2.2K seems like a very low value for a low-current (0.5mA?) FET's drain resistor, reducing its gain. It is needed when using a battery of only 3V or even 1.5V, and can go a fairly long distance before cable capacitance rolls-off its high frequencies. The FET will have a lot more gain with a 10K drain resistor and a 9V battery for use with short cables and a 10K or preferably more termination resistance. But nothing much will be gained when it is loaded down by the 3K amplifier input.

 

[Nigel Goodwin]

Hi Nigel,
I have always thought that the drain of a junction FET is a high-impedance current sink, eg a constant-current diode. Therefore the output impedance of the mic with a 2.2K drain resistor is very close to 2.2K.

I'd rather measure it and find out :lol:

 

[audioguru]

The output impedance of the mic isn't the value of that resistor - it's probably considerably lower than that

Hi Nigel,
You are absolutely correct, sort of.
I just finished testing the electret mic from a cell phone. It is interesting that I found the little mic to be thermal, and it is trying to be a high-impedance current sink but not very well with a low supply voltage.

I used a 10K drain resistor for its FET:
1) I measured the DC voltage across the 10K resistor at various supply voltages:

Supply..Measurement
..2V......1.9V cold, dropping to 1.84V in 10 seconds.
..4V......3.8V cold, dropping to 3.36V in 20 seconds.
..8V......7.2V cold, dropping to 4.35V in 30 seconds.

The little mic took the time listed above to thermally stabilize. I checked the resistor by measuring it's resistance before and after putting 8V across it for a while, and after heating it with my soldering iron. It didn't change its value.

2) I fed 100mV, 1KHz through 10K in series with 1uF into the mic's output (the junction of the FET's drain and the 10K drain resistor):

Supply..Measurement..Calculated Mic's Impedance
..open...50,0mV..........10K (check)
..2V......8.9mV............1087 ohms
..4V......19.5mV..........2422 ohms
..8V......23.8mV..........3123 ohms.

 

[Nigel Goodwin]

2) I fed 100mV, 1KHz through 10K in series with 1uF into the mic's output (the junction of the FET's drain and the 10K drain resistor):

Supply..Measurement..Calculated Mic's Impedance
..open...50,0mV..........10K (check)
..2V......8.9mV............1087 ohms
..4V......19.5mV..........2422 ohms
..8V......23.8mV..........3123 ohms.

Interesting technique!.

I would have fed a speaker with a 1KHz tone and monitored the output of the mic with a scope (as I don't have an AC milli-voltmeter). Then loaded the output with a resistor and measured the change in output, and calculate from that. Or, same method but with a variable resistor as a load - adjust the variable resistor for half the signal, then take it out and measure it's resistance - saves doing any sums :lol:

If you've still got the stuff setup, I'd be interested to hear if it would give the same results as your ('rather sneaky' method) :lol:

 

[audioguru]

I would have fed a speaker with a 1KHz tone and monitored the output of the mic with a scope

I thought of doing it that way, but couldn't find any Crazy Glue to make a fixed distance. Clamps and screws? No thanks. As it was, I had the mic in a blanket to stop it from hearing me swearing at it when it was warming up. Hey, it's winter and freezing outside. I should try...... nah.
Maybe that's why "on the street" radio and TV reporters have electret mics with 3-wires so the FET is a source-follower with lots of feedback for stability.
He, he.

Or, same method but with a variable resistor as a load - adjust the variable resistor for half the signal, then take it out and measure it's resistance

My spare pots are all intermittent cheap Chinese ones. My good ones are all installed in projects. Besides, my old slide-rule still works pretty well for calculations.

See, there are many ways to skin a rather sneaky electro-cat! :lol: