[maemo-developers] [maemo-developers] Measuring power consumption of 770

From: Igor Stoppa igor.stoppa at nokia.com
Date: Wed Mar 15 13:44:43 EET 2006
On Wed, 2006-03-15 at 12:59 +0200, ext Teemu Nikkilä wrote:
> On Tuesday, 14. March 2006 22:42, Claudio Scordino wrote:
> > > Why don't you just wire a battery to the battery pins and connect your
> > > meters to these wires?
> 
> I'd investigate this approach, maybe even building a battery "emulator" - a 
> good quality 3.7 V regulated voltage source, if this is enough to fool the 
> charging circuit.
> 
If I'm proposing the battery, there's a reason ... and no, i won't go
any further, but anyway faking a battery is not the scope of this study,
according to my understanding, so using a real one will be just ok.

> >
> > The voltage and the current provided by the battery is much different from
> > the values provided by an electric cable connected to a socket (220V and
> > 50Hz here in Italy).
> > That's why our measurement tools wouldn't work with such a small current...
> 
> Without knowing your tools let's screen the problem - the battery is 1500 mAh 
> and it can deliver power to the device for hours, say 5 h. This means the 
> average current draw during normal operation would be around 300 mA, most of 
> which goes to the screen backlight and radios, I guess.
> 
> If and when your algorithm can save some power, the difference in current 
> would be very small and rapidly changing anyway. To make an exact 
> measurement, you would probably need very good measurement equipment and also 
> you would want a data logger with reasonably fast A/D converter to see and 
> record what really goes on during the execution.
> 
Anyway, IMHO, any power saving study should be done only after gathering
info on the processor itself and the power management software actually
in use.

The code is freely available in the kernel tree from maemo.org, while a
TRM for some processor similar to 1710 (which requires NDA, afaik)
should be available from TI website, for their Innovator boards.
> -Teemu
> 
> >
> > Many thanks,
> >
> >                Claudio
> >
> > _______________________________________________
> > maemo-developers mailing list
> > maemo-developers at maemo.org
> > https://maemo.org/mailman/listinfo/maemo-developers
> _______________________________________________
> maemo-developers mailing list
> maemo-developers at maemo.org
> https://maemo.org/mailman/listinfo/maemo-developers
-- 

Igor Stoppa, Nokia M / Tampere Finland

More information about the maemo-developers mailing list