OK, when to REALLY expect updates

Well, it was quiet for the first half hour, and after that it went over the lake.

If it's at 18000 feet, the batteries are cold and the voltage is low, so it won't try to update very often. We saw fully charged batteries go to ~3.6V at -10F or so.

Here's the complete, overly complicated update algorithm in order to try to survive cold temps / low batt / transatlantic journeys.

When it goes to sleep between 6 AM (local!) and 7:59 PM (OR: There's zero GSM signal at all, not even roaming)
-- Sleep for 15 minutes
-- If battery is lower than 3.9V, sleep for 30 minutes
-- If battery is lower than 3.8V, sleep for 1:30
-- If battery is lower than 3.7V, sleep for 2:30
-- If battery is lower than 3.5V, sleep for 5:30
-- If battery is lower than 3.4V, sleep for 8:30 (Hopefully, solar cells help here
-- If there's messages stuck in the outbox, sleep another 30 minutes

Otherwise (night time or zero signal)
-- Sleep for 1 hour
-- If battery lower than 3.8V, sleep for 3 hours
-- If battery lower than 3.7V, sleep for 5 hours
-- If battery lower than 3.6V, sleep for 7 hours
-- If battery lower than 3.5V, sleep for 9 hours
-- If battery lower than 3.4V, sleep for 11 hours (Sadly, seems to be typical for mid-way charged batteries at -18F)

Also, it'll send VGA images if there's anything in the outbox at powerup.

Batteries seem to recover nicely from sleeping at -18F. I had some batteries in the freezer that recovered fine.

If the phone wakes up and pukes due to low battery, there's a backup alarm set for +8 hours. So if the phone went to 30,000 feet (-40F) and can't wake up at night, it "should recover" when the sun comes up. Probably.

So, yeah, wait 24 hours before calling it dead. :) I'm hoping there was no design failure, and we just got unlucky in the first 30 minutes of flight, and rural Michigan's TMobile network isn't friendly.

-CF