[gta02-core] Backup battery underdimensioned

[Rask Ingemann Lambertsen]

On Wed, Aug 19, 2009 at 02:35:58PM -0300, Werner Almesberger wrote:

> Also, the PMU's data sheet specifies only a maximum total supply
> current, leaving it unclear on which path this goes, and if it
> also applies if there's no main battery. Also, this doesn't say
> much about the average current.

   This bit on page 16 makes me worry (last six words):

	Save

	In the Save state, the power management section is supplied from
	either the system voltage (if VSYS > Vth(syspres)) or the backup
	battery (if VSYS < Vth(syspres) and VBUBAT > Vth(bubpres); see
	Figure 6). Only the internal supplies, the 32 kHz oscillator and the
	real-time clock will be active. The GPIOs will maintain their state.

   In particular, GPO has a push-pull output which we use for EN_USBHOST.
It's not clear from the PCF50633 manual how much that affects current drawn
from the backup battery or which state consumes the least power.

> plus we have a number of
> potential time sources (GPS, GSM, NTP if we have IP connectivity),
> so resetting the clock after a battery change may be an option.

   At least with the default ext2[1] settings, this causes two runs of
e2fsck: The first one because the last mount time is in the future and the
second one because the fs has gone 3000+ days without being checked.

[1] Yes, I know, people shouldn't be using it, but ...

-- 
Rask Ingemann Lambertsen
Danish law requires addresses in e-mail to be logged and stored for a year