Help! Microphone impedance matching?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have a dynamic microphone with 5 Ohm impedance and a radio designed to use an external electret microphone with 150 Ohm impedance. I attempted to use the two together after matching the impedances with a transformer (pretty close to a 1:30 impedance ratio), but I could barely hear anything when transmitting to another radio. ???

The transformer has an primary impedance of 16 Ohms and a secondary of 500 ohms, which is a close ratio to the microphone/radio impedances (5/150).

So does a transformer used for impedance matching only need to have the right turns ratio, or do the impedances of the actual windings need to match the source/load impedances as well?

Any way to do this without winding a custom transformer? :confused:

Oh yeah, here's the crux: It has to be passive...
 
Those are winding resistances, I suppose. They are effectively in series with the source and load, respectively. Not too bad for the load side, but at the source side you have 5 ohms vs. 16 ohms. IOW you're already losing about 12 dB there, as only 5 out of 21 ohms are actually contributing signal.

Then if you really have a 1:30 turns ratio, this means 21 ohms (5 + 16) get stepped up to 19 kOhms... and connected to a 1 kOhm input. Poof, another 26 dB gone at the output side.

And who knows what kind of low-frequency response the respective inductances are good for.

In short, you need a beefier transformer. Maybe you can use one for 70V / 100V systems. Primary winding resistance <<5 ohms (like 1 ohm or smaller), inductance probably in the hundreds of mH or higher. Ideal turns ratio maybe 6..10.

Oh, and the 150 ohms of that electret mic probably was the output impedance of the built-in source follower. BTW, electrets will always be more sensitive than dynamics... easily 10..20 dB.
Well, I never came across a dynamic mic with a 5 Ohm impedance.
Neither did I, but Google says they seem to be common in aviation.
 
Last edited:
wow, I'm impressed, sgrossklass. you did your homework, the microphone is for an aviation headset, which I am trying to integrate with a portable radio. I can't change the microphone (or it will no longer work with the helicopter intercom system) and definitely can't modify the radio, but I can do whatever I want in-between.

I measured the DC resistance on the transformer; 2.2 Ohm primary, 45 Ohm secondary. supposedly the transformer has an primary impedance of 500 Ohms, and secondary of 16 Ohms, for am impedance ratio of about 1:30. So the turns ratio should be about 5.3:1, right?

Is the relationship between turns and DC resistance proportional? if so, the transformer is definitely not what it says it is (45 Ohms/2.2 Ohms = 20:1, not 30:1).

So do I understand correctly that I either need a transformer with fewer windings on both sides (lower inductance?) or a transformer wound with heavier gauge wire to reduce the DC resistance?
 
I know that's possible to do this effectively. I have an adapter on my desk designed to solve exactly this problem. Unfortunately, the adapter is extremely expensive and is not able to be opened up and examined. I tried, whatever was in there was fully potted in plastic and got destroyed in the process. I tried testing both sides of the adapter with a multimeter and I know that there is some sort of inductive component as well as a capacitor in there.

I only have 1 other of these adapters, and can't afford to destroy it. But I REALLY want to know what's in there...

I was convinced it was just a transformer similar to the one I tried, but thats not the case...

Thoughts?
 
I measured the DC resistance on the transformer; 2.2 Ohm primary, 45 Ohm secondary. supposedly the transformer has an primary impedance of 500 Ohms, and secondary of 16 Ohms, for am impedance ratio of about 1:30. So the turns ratio should be about 5.3:1, right?
Yup.

This thing is about a factor of 3 away from what I'd like impedance-wise, but it should still work passably (though it may be a little lean in the lower registers). Hmm.

Wait a microsecond... you have used a suitably sized coupling capacitor between xfmr and mic input, haven't you? We don't want to drive the poor xfmr into saturation, though I can't quite imagine a bias supply for electrets would manage that so easily.

I also assume you had the xmfr wired up the right way around (lower impedance side at the mic).

Do you have any sensitivity spec for the mic (mv/Pa or dBV/Pa)? Run-of-the-mill small electret capsules tend to be in the -50 dBV/Pa territory, better ones a little over -40.
OK, I found something for an M-87/AIC:
34dB to 40.98 re. 1mV AT 1KHz WITH AN INPUT OF 28 DYNES/CM²
Grr, I hate unit conversions. So that's like -34 dBV/Pa, it appears. Odd. I'd try that one with just a coupling cap.

Aircraft audio seems to be a nice mess. Oh my.
Is the relationship between turns and DC resistance proportional?
Not necessarily - that depends on wire gauge.
So do I understand correctly that I either need a transformer with fewer windings on both sides (lower inductance?) or a transformer wound with heavier gauge wire to reduce the DC resistance?
Actually you'd need higher inductance and lower resistance, i.e. more turns with thicker wire or a more substantial core.
 
The commercial adapter does not have batteries, it is entirely passive. There are actually many different versions of the same thing available online. It seems to be a common problem.

Im sure I have wired the transformer correctly (I did try it both ways actually). I also tried 2 of them in series, but then it really didn't work.

I'm sorry, a what? Coupling capacitor?

I don't have any more specs on the transformer, only what's on the online data sheet (it's part # 42tl026):
http://www.mouser.com/catalog/specsheets/XC-600127.pdf

I don't know if this helps, but the microphone is almost impossible to hear unless I literally put it in my mouth and scream. Then it sounds okay...

You guys are awesome. This is way more fun/educational than hassling the manufacturers...
 
The commercial adaptor may have an active buffer stage inside, powered from the radio. An electret mic does, typically a JFET with open drain connected to the output signal line. The radio provides maybe 5 volts DC through a 2K2-ish resistor, all down the same wire.

Can you measure DC voltage at the radio's input jack (with transformer removed, of course)?

All good fortune,
Chris
 
I'm sorry, a what? Coupling capacitor?
A coupling capacitor is a capacitor used for coupling between stages, which means nothing more than pass AC and block DC. Something like a >=22µF, >=16 V electrolytic (+ to output, - to xmfr) should work here.
I don't have any more specs on the transformer, only what's on the online data sheet (it's part # 42tl026):
http://www.mouser.com/catalog/specsheets/XC-600127.pdf
001 or 004 would seem like the ideal choice. Anyway, it seems the thing has center taps on both sides (pins 2 and 5). Have you tried using those instead of the whole winding? Which combination gives highest output?

This transformer is pretty cute. I wouldn't be surprised if an electret bias supply could, in fact, drive it into saturation. So do try the coupling cap first.
The commercial adaptor may have an active buffer stage inside, powered from the radio.
That's a very good idea. A few surface mount components would also fit the definition of "small and easily destroyed".
 
So, I measured the DC voltage across all combinations of the 4 wires from the radio. I measured constant 5 VDC across one of the mic wires and one of the speaker wires. So there is a potential to use an active circuit here (I was very surprised buy this!). However, after further testing I discovered that the adapter still works without the speaker wires connected, and therefore it works without the help of the 5 VDC!?

sgrossklass, I also tried the 001 transformer, which I was able to salvage from the speaker side of another adapter that I have for a different radio. They sound very much the same, very difficult to hear unless you scream into the mic.

I have tried all possible combinations of wiring the transformer, It seems to work about the same whether wired to the center-tap or not.

I will try the capacitor, although I'm not exactly clear were it goes. Is it wired in series or parallel to the transformer, and on which side? Is series on the radio side best?

Thanks again for the help. My mind is boggled, although it seems like we might be getting somewhere.
 
The ordinary consumer electronics version of electret mics uses what's called "plug-in power", a single wire phantom supply on the same wire as the signal. That's the 5 volts you're seeing. Electrets have orders of magnitude larger signal than a dynamic mic, and that's what your radio is waiting for. I'd probably try an n-channel FET buffer with its drain connected as the plug-in power would be. Can't hurt!

All good fortune,
Chris
 
I will try the capacitor, although I'm not exactly clear were it goes. Is it wired in series or parallel to the transformer, and on which side? Is series on the radio side best?
In series, and at the radio side since that's the higher-impedance one and allows for a much smaller-value cap.
I measured constant 5 VDC across one of the mic wires and one of the speaker wires.
But not across both mic wires? Having a bad connection there would explain quite a bit. Are you sure about the pinout?

Then again, you say the commercial adapter still works? :confused:
 
okay. so there is 15 VDC across the mic wires while transmitting. Bummer. It's probably active then?

Tried the transformer with the coupling capacitor. No luck, sounds about the same. I want to try the FET buffer. How on earth do you isolate the DC voltage, use it for the FET, and then feed the output back into the radio? I guess you "de-couple" and then "re-couple"?

Any one have or know of a schematic for a simple N-channel FET buffer?
 
okay. so there is 15 VDC across the mic wires while transmitting. Bummer. It's probably active then?
Ah. It makes sense to power the mic during tx only.
Any one have or know of a schematic for a simple N-channel FET buffer?
Try something like this.
You'll only need R1, R2 and Q1, the rest (presumably) is inside the radio. R1 does not need to be 3 meg, 1 meg is just fine. Similarly, R2 was chosen for a low Idss, low gm part (the J201). As you want a healthy amount of gain, try something like a 2N3819, MPF102 or J309/J310, with R2 in the low-mid 100s of ohms. Take ESD precautions when working with the FET, which is easily destroyed by static electricity.

Actually it's not a buffer but an inverting amplifier. For a buffer (source follower), you'd need a 3-wire connection.
 
Can I just do this? Or is there some special way to couple the output from the JFET back into the DC supply from the radio?

This seems too simple...
 

Attachments

  • PREAMPJPEG.jpg
    PREAMPJPEG.jpg
    3.5 KB · Views: 113
I'm to do something very similar to this thread. I provide sound for a local air show and I have asked if i can provide conversation between the announcer and the pilot.
I know I can do this by used of a scanner connected to the pa mixer but the announcer would have to hold two mics to his mouth in order to do this (one for the radio and one for the pa). I would like to do this by feeding the dyanmic mic signal into a handheld radio using a ptt switch for the annoucer to key the xmitter. --Any suggestions ???
 
hi impedance microphone

Hi you all.
I have found an old dynamic microphone with 5kohm output, so I need to match the microphone's impedance to my microphone preamp in order to record with the best quality possible.
My knowledge about electronics is very basic, so... i really need some help because I would like to build a impedance matcher for this microphone.

What do I need?
How can I do it?
Can you help me?

Thanks and greetings!!
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.