[gta02-core] ECN0019 (was Re: ECN0010: LDO2/LDO4 swap implemented)

[Werner Almesberger]

Rene Harder wrote:
> Apart from that, it's exactly what I had in mind,

Good, I guess it must have been C4307 that threw me off.

> except for relocating
> the choke. I would prefer to handle this in a different ECN,

Okay, makes sense.

> combined
> with moving C4301/2 and C4305 to the codec side of the choke.

I was wondering about them as well. Joerg said that having them
close to the microphone, before the choke, would be better, "to
give RF a path to GND, so actual HF voltage isn't that high".

I'm not sure this makes sense to me. So I simulated an LC and a
CL configuration [1]. I found that they perform pretty much
identically when input and output impedance are high. They do
show a marked difference if the output impedance is very low,
and also a low input impedance causes some slight differences,
with L-at-input showing better performance.

[1] http://svn.openmoko.org/trunk/gta02-core/docs/ecn/ecn0019/lc-or-cl.sch

- Werner