Bugzilla – Bug 16640
Date/Time gets lost with the new processor
Last modified: 2012-03-09 07:28:49 UTC
Power off the Baby and power back on after 10 minutes and the date/time is lost.
The Baby and MSP430 can communicate through the i2c bus because the following actions work fine: 1. The Ambient Light Sensor is working fine. 2. The battery/charger is functioning fine. 3. The knob works fine. Also even though the Date/Time is lost the MSP430 firmware is fine and there is no need for reprogramming it. During run-time and by using the hwclock command I monitor the date/time on MSP430 and it is fine. The MSP430 keeps the date/time correct while there is no AC removal. This proves the followings: 1. The MSP430 is functioning fine 2. The processor is working fine. 3. The communication is fine. I changed the MSP430 to make sure that it detects the AC removal and it does.
From my debugging code it seems that after the AC is removed we execute the code that should put MSP430 into low power mode(LPM). Based on the fact that we lose the date/time it seems that there are still some components running after the execution of the LPM. I would like to see the current measurement after the AC is removed to confirm this, so I assign this to John.
*** Bug 16553 has been marked as a duplicate of this bug. ***
Powering on the Baby with the new processor: 1. Everything works fine, which includes Battery/Charger, Ambient Light Sensor, Remote, etc 2. Power off the Baby and power on after 1 minute. The Baby boots up and the Date/Time on MSP430 is fine. 3. Set the default Date/Time in MSP430 to something other than 0 to make sure that the MSP430 date/time is used during power up and that is the case. 3. Power off the Baby and power on after 5 minutes. The Date/Time is lost. The only thing that could cause this is if one of the hardware component does not go to sleep and is taking a lot of current OR the supercap is not functioning properly.