Somewhat bricked neo problem

Gary Oliver go at l5.org
Sat Sep 15 19:33:25 CEST 2007


About 2 weeks ago my NEO stopped working.  This was about 48 hours after
doing an flash (from the fairly recent 2007.2 build.)  The unit was
plugged into my USB port on the desktop and was left 'on' but not
connected to GSM (no sim card in place, in fact.)  I had *not* flashed
u-boot to this point - it was as delivered.  Only rootfs and kernel had
been changed.

The problem seemed to be a dead battery, so I plugged in my 'spare'
(fully charged) battery and it came up, but within about 5 hours died
the same way.  Suspecting the charging process was broke, I pulled the
battery, unplugged the USB cable and waited about 5 minutes, reinstalled
the battery and plugged in to the USB cable again.  Voltmeter probing of
the battery contacts (while charging) showed it was getting juice, so I
figured it would come alive in a few hours.

Not so.  Now I have two fully charged batteries, but the NEO won't
start.   I should note that until this time, I had NOT opened the case
(other than the back cover to insert the battery.)

But the strange thing:  If I install the debug board connection, and
power the debug board, the unit will boot and run normally (although the
battery does seem to get a bit warm...)

While messing with it this morning, I decided to take a chance and flash
u-boot, kernel and rootfs from a full build last night.  While plugged
into the debug board, it came up and ran as expected.  However, when I
remove the debug board cable, it is still bricked.

One possibly significant thing: If I let the unit fully discharge by
removing the battery and letting it sit for several minutes, a power up
will (after about *15* seconds, not 3, briefly flash the screen with
some random pixels and then go blank.  The vibrator does not buzz.
After this, there is no activity from the NEO until I either 1) plug in
the debug board or 2) repeat the discharge process and try again.  I
have followed all recommendations with respect to power application,
when the debug board was installed (i.e. powered off, no battery, no USB.)

Has anyone seen this and if so, what was the remedy?

Given it still sort of works on the debug board, it's not completely
useless, but I can't cart it around and show my friends anymore :-(

For ref, the unit is a v4 from the first round phase 1 deliveries.  Up
to the point of failure it was working well.

-Gary





More information about the neo1973-hardware mailing list