My Gta02s wont charge

Joerg Reisenweber joerg at openmoko.org
Mon Jun 23 03:34:35 CEST 2008


Am Mo  23. Juni 2008 schrieb Werner Almesberger:
> Joerg Reisenweber wrote:
> > Yep, you got it. That's what I say with "when we don't obey the specs...".
> 
> But which spec would we be violating ? If there's nothing connected,
> nothing happens anyway. If there is a USB upstream device or the
> adapter, we're allowed to draw 100mA without the need for further
> clarification.
Yep

> 
> So all we have to do is make sure the power we draw until we're in a
> better position to decide on the system's fate doesn't exceed 80mA
> (this is what the 100mA limit effectlively translates to, see the
> PCF60633 manual, table 91, page 75).
ACK

> 
> Since our system currently exceeds this 80mA envelope, we do wait for
> an indication that we have a power source that can (probably) give us
> what we need. It's in u-boot, board/neo1973/gta02/gta02.c, function
> wait_for_power.
Well I didn't look at this code yet, have no idea what it's actually doing 
(probe for 47k-ID?).
Anyway, until this function is telling us it's save to draw more than 100mA, 
we have to limit our current from USB, and we need to run the CPU with this 
current to execute the function mentioned by you. Then we are steady to boot.

> 
> Note that this isn't bullet-proof. We could do better by running the
> CPU at a lower clock speed. Also, given the excessive current drawn
> by the GTA02v5 LEDs, the LED that's being blinked as an indication
> in wait_for_power will probably bring down the system by pushing it
> over the 80-100mA current limit.

Agree, think we have the same idea of how stuff should work.
/j

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 194 bytes
Desc: This is a digitally signed message part.
Url : http://lists.openmoko.org/pipermail/openmoko-kernel/attachments/20080623/5f35968c/attachment.pgp 


More information about the openmoko-kernel mailing list