[alsa-devel] [PATCH 1/1] ASoC: TWL4030: Wait RAMP_DELAY on headsetl_event: POST_PMD

[Jarkko Nikula]

On Tue, 12 May 2009 16:16:08 +0300
Peter Ujfalusi <peter.ujfalusi at nokia.com> wrote:

> On Tuesday 12 May 2009 15:21:55 ext Jarkko Nikula wrote:
> > On Tue, 12 May 2009 15:16:40 +0300
> >
> > Peter Ujfalusi <peter.ujfalusi at nokia.com> wrote:
> > > > Looks like the beep is caused by the ramp operation since beep
> > > > is gone in power-down when reverting both of your patches but
> > > > it's still present in power-up.
> > >
> > > But the 'tuck' still there after you have reverted the patches?
> >
> > Yep it was there even it sounds bit more silent when going to longer
> > ramp times. Beep was present only with '109/81/55 ms'.
> 
> Now I'm even more confused...
> Without the patches that you have reverted, the ramp delay does not
> matter after the playback. It just cuts the VMID and that's it.
> On startup the ramp is initiated. If you set the ramp delay longer,
> than it is possible that the beginning of the audio is played already
> when the VMID is ramping up.
> 
> So with the ramp (down) delay - the patches that you have reverted -
> the 'tuck' was worst after playback, than without the proper ramp
> (down) delay handling?
> 
Hopefully I didn't confuse you too much :-)

So with your patches reverted, the low-frequency pop was gone in
power-down since there were no ramp implemented but there still were
the sharp snap. I didn't notice were there any difference on that.

Then this low-frequency pop went more silent as the ramp time
increased. Allways in power-up phase and also in power-down when your
patches were in implementing the ramp down.

Third observation was that the beep was present with ramp time of
'109/81/55 ms'. Without your patches it was present only in power-up
and with your patches is was also in power-down phase. So it sounds to
me that beep was there when the ramp up or down was going.

> I can see on the scope, that the ramp up also has kind of 'tuck'
> problem, but it comes when the VMID is quite low, so it is harder to
> hear it. By making the ramp delay longer, this 'tuck' happens when
> the VMID is close to 0, so it makes it less audible.
> 
I think for me this low-frequency power-up and -down (when your
patches are in) pops are caused when the capasitors connected into
HSOL/HSOR are charging/discharging through the headphones.

What I'm wondering how it is possible to hear such a slow ramp like 100
ms?


-- 
Jarkko