Content originally posted in LPCWare by CodeRedSupport on Wed Oct 19 01:56:07 MST 2011
I have done a bit of experimentation at this end, and I think that part of the problem is simply that the signal delivered by many microphones is simply too low for the audio system/codec to handle. Certainly the microphones that I have access to are close to useless here.
Anyway, if you modify your code to have a 0 gain on the microphone input, then plug a line-in source into the microphone socket of the board, then the circuitry definitely seems to work. For example I can change our standard RDB1768cmsis2_AudioMeter to take its souce from the microphone rather than line-in socket. If I then connect up my ipod, the output through the headphone socket between the original and modified applications is basically identical (and quite acceptable to my non-audiophile ear). It might be worth trying to put some form of amplifier in place to boost your microphone input to line level, something like a Griffin iMic.
Anyway, I don't believe that there is an issue here with the board circuitry - but is likely to be a setup or other issue with the uda1380 codex itself. I suspect to make any further progress, we would need some input from a UDA1380 expert from NXP.
Regards,
CodeRedSupport