Phase difference based sensing is not at all new! Not even for audio signals. To sense direction the sensors need to be a decent portion of a wavelength apart. That is the most basic physics part. So we really do need more details.
AND the very low nose IC amplifiers have been available for a whole lot of years.

 

If you are going to measure the phase differences, you need to control the gain and phase response of the buffer amplifier. Since transistors have a variable beta because of manufacturing tolerances, your only option is an op amp based buffer.

 

If you are going to measure the phase differences, you need to control the gain and phase response of the buffer amplifier. Since transistors have a variable beta because of manufacturing tolerances, your only option is an op amp based buffer.

That's rather an over-simplification. It would not be difficult do design a discrete transistor amplifier with a well-define gain and phase.

 

Certainly a discrete component amplifier can be built without too much effort. making four of them identical would take more effort. An IC amplifier can provide more gain with fewer parts in the same space.

 

Certainly a discrete component amplifier can be built without too much effort. making four of them identical would take more effort. An IC amplifier can provide more gain with fewer parts in the same space.

In my experience there is more variance in the capsule response than the preamplifier. I have built OpAmp and discrete transistor variants for ECMS, large diameter condensers, phantom powered microphones, etc. Its an interesting chat, but the question in this post is do I need a preamp at all between a biased ECM and the ADC given that they are both on the same PCB. As I've said, the ADC datasheet only references a direct connection with a differential MEMs microphone (low impedance output). I have raised a ticket with TI who make the ADC, who may be able to add more clarity. Thanks.

 

I thought that I had posted that the TS needs to understand the available input signal voltage reading capability of the A/D device and also the output signal voltage from the electret microphone, based on the amplitude of the sounds to be detected. The difference between those signal levels is where an amplifier will undoubtedly be required. We do not have those numbers and so we can provide guesses based on our experience.

 

I thought that I had posted that the TS needs to understand the available input signal voltage reading capability of the A/D device and also the output signal voltage from the electret microphone, based on the amplitude of the sounds to be detected. The difference between those signal levels is where an amplifier will undoubtedly be required. We do not have those numbers and so we can provide guesses based on our experience.

I don't think I need external gain, as I can use the PGA inside the ADC to use the full dynamic range of the ADC. Its more to do with whether it is inadvisable to not buffer it from a signal integrity point of view. As I say, the only example in the datasheet is the MEMs microphone which has low impedance output.... Lets see what TI have to say...

 

Your electret mic is not "high impedance like a crystal mic, since the output is developed across that resistor. Besides that, consider that the main reason for impedance matching is to provide the best power transfer. How much input POWER does the A/D's PGA require??

 

Your electret mic is not "high impedance like a crystal mic, since the output is developed across that resistor. Besides that, consider that the main reason for impedance matching is to provide the best power transfer. How much input POWER does the A/D's PGA require??

I'm not talking about matched impedance at all! Again- I am asking if the load of the ADC will distort the signal from the ECM if they are directly connected. If so, then I can see that the better solution would be to add some sort of buffer. If not, then the better solution from a cost and complexity point of view would be a direct connection. I put the question on this forum because I thought that somebody may have had to make the same design decision the past. Lets see what TI say.

 

OK, the impedance discussion was in another thread. Sorry about that.
If the input signal amplitude does not force the input of the internal amplifier into a non-linear portion of operation that changes the impedance, then there would be no distortion caused to happen in either direction.