iPhone 5S microphone input impendance and swithching to external source.

[bezen4uk]

Hi guys,
Today I spent some time to investigate microphone input of my iPhone 5S.
I made simple DI for smartphone, and I wanted to find best way to interface it with my phone.

So, the microphone input has bias 2.75V through resistor 2.21kOhm (actual measurement is 2280 Ohm, measured with 3 different loads).
Internal microphone is switched off when the DC current of mic circuit is more than 85uA (30kOhm load).
AC-coupled input impendance is about 2.2kOhm.

 

[Bo Deadly]

So what circuit are you going to use? Just brainstorming off the top of my head I am actually drawing a blank about what a good circuit would be in this case.

For a DI you don't need amplification so I don't think you want a FET in there. If anything you want attenuation (10dB maybe). So you need a >30K resistor across the input but  you don't want the source impedance seen by the iPhone to be that high (preferrably as low as possible I would think). Nor do you want the load impedance seen by the remote source to be that high (10K maybe).

Another thing to consider is if it should be balanced. If the iphone isn't plugged into anything (meaning it's floating on battery) then it probably doesn't matter. But if the cable is plugged in then you're connecting the ground of the source to the ground of the iPhone cable and whatever it's connected to (a computer).

And because you have the bias you need at least one capacitor - two for a balanced circuit.

And if it's passive, the circuit should probably be as close to the iPhone input as possible and yet it probably won't fit in the jack housing.

It will be interesting to see what you come up with.

 

[bezen4uk]

Hi,
Thank you for your input!
I doodled few circuits and found that simple PNP follower would be best, though I should check how it works with guitar cable capacitance.

 

[Bo Deadly]

Hi,
Thank you for your input!
I doodled few circuits and found that simple PNP follower would be best, though I should check how it works with guitar cable capacitance.

So this is for guitar? Then you don't need that much attenuation - 30k / 2.2k is too much. And you don't need the 50k. If you drop the 50k and change the 30k to 22k that would be better but it's still -20dB attenuation which is too much for guitar. You probably want -10dB attenuation but you can't go lower than 22k or you're going to go over the current limit (2.75 / 30k = 92uA?). Maybe add a capacitor accross the 22k but that will cause the input to clip. You could break down the 22k into to 10k and partially bypass to get a high shelf. It's a feature!

But otherwise I like it. The input and PNP is good. Probably don't need the first 1M and a TVS diode like SMBJ15CA would be a pro touch.

 

[PRR]

> don't need that much attenuation

No; but he needs some. This type mike won't put out over 0.1V maybe 0.3V. Guitar often beats 0.3V, traditionally 0.5V, and some modern rigs go well over 1V. So I would expect 4:1 to 10:1 padding to be needed.

 

[bezen4uk]

Microphone's input full scale sensitivity  is 40mV (info from web), so deep pad is required.
Though, didn't check myself.

 

[duffharrisnyc]

Microphone's input full scale sensitivity is 40mV (info from web), so deep pad is required.
Though, didn't check myself.

Hi - for some reason I can't message you. Just wanted to ask if you still have any Gefell MV692 conversion kits.

Thanks!