Wifi / let's have RF stuff powered with high current source all the time because...

[William Kenworthy]

On Tue, 2008-08-26 at 22:07 +0900, Andy Green wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Somebody in the thread at some point said:
> 
> | I have no memory of any calls coinciding.  And yesterdays was a clean
> | boot in the morning, with no calls up until the problem ~an hour later.
> | Thats not to say a call came in and didnt register - but I have had none
> | asking why I didnt call back :)
> |
> | Every case though was during gps use: openmoko-agps was run first until
> | lock was obtained, then tango-gps.  In only the last case though, did I
> | notice the FR being warm/hot to touch.  Otherwise it was the battery
> | being drained at an unusually fast rate - stood out when I checked the
> | battery charge level after restarting (I have a fair idea where the
> | battery is at at most times.)
> |
> | Not really scientific or smoking gun stuff I'm afraid.
> 
> No for a problem report we're lucky to have it that good, thanks for
> posting about it.
> 
> What was getting hot?  Battery?  Some particular area of the Freerunner?
> 
> You don't use GPRS?
> 
> GPS stuff doen't seem itself to have a way to pull the amount of current
> we're talking about without something bad having happened.  But it does
> remind me of an ancient bug report about the UART traffic from GPS
> getting corrupted if you had debug board connected.  I wonder if there
> is some general problem with UART traffic corruption hiding somewhere
> that is also affecting GSM call lifetime or somesuch.
> 
> - -Andy
...

No GPRS.  I should have pulled the battery and checked what was getting
hot but didnt - was in a rush to get to a meeting so other than wanting
to stop whatever was causing the heating I didnt think to dig into it -
next time!

The whole device was hot, especially the screen. When using GPS in the
car, I place it against the speedo horizontally (fits nicely, stable and
doesnt hide the important bits - the GPS works fine, but there is only
the plastic sun shade part of the instrument cluster over it.) with the
display switched to landscape, and no power saving.  I do this often,
and the device is usually barely above ambient temperature if at all.
Nothing was different (i.e., dash lights or other possible causes of
heating happening)

I wouldnt call it too hot too hold, but uncomfortably warm from the
point of view it was IMMEDIATLY noticable as soon as touched, and images
of flaming batteries and melted plastic ran through my mind :)
Probably no worse than if it was left on the dash in the Australian
summer sun, but not expected on a cold winters morning about 8 degrees
outside and the device not exposed to sun etc.

I posted once before a figure for current draw and I think you responded
that it was much higher than expected, but I have not noticed it being
that high again - I now have a small script so I can extract the figure
quickly without having to navigate the /sys hierarchy. - next time I
will be ready :)

BillK