[Rene Harder Mo  30. November 2009]:
> I think, the idea was to build a RC filter (R4305 & C4306) with a really
> low cutoff frequency to get rid of any noise introduced by the MICBIAS
> line.
> However, in this case R4310=0 Ohm fcut->inf (MICBIAS internal resistance
> not take into account). So to get the same effect as in the current
> design we would need to increase R4310 which is not possible because
> this would also increase the voltage drop across R4310 and R4305. 
> Currently, we are already operating the microphone near the minimal
> required voltage so an additional decrease will probably result in
> malfunctioning of the mic.

That's why we'd better off with 2 * 1k, and in case we need we could create
a R-C-R lowpass by rplacing the other 0R with a further 1k and equipping the 
C. A 100uF creates a AC short to GND, so we could keep the symmetrical design 
of 1k from big C to mic, and one 1k from micn to GND. And we still have a R-C 
lowpass built from the new R. If we dn't need the big-C fix, we can leave out 
the C and make the one R 0R.

btw: 1k + 1k (+ 1k) < 2k2 + 2k2

/j

Attachment: signature.asc
Description: This is a digitally signed message part.

_______________________________________________
gta02-core mailing list
[email protected]
https://lists.openmoko.org/mailman/listinfo/gta02-core

Reply via email to