Battery Life on Motorola CE Devices

Our WT4090 devices suffer from short battery life. The MC3190 devices are not so susceptible but users report needing 3 replacement batteries to get through their 8.5 hour shifts, which seems like a pretty poor battery life to me.

Initially, this has been put down to the device itself. Because the battery is so small and the devices are in constant use, it was initially dismissed as a limitation of the device itself and nothing we could do to tune it.

We have since adopted third party software to run another operation at another site using the exact same devices and now experience a much greater battery life. The devices are in constant use, as they are on this site but are lasting a lot longer. This has led to some investigation to take place to see what the underlying problem is.

I am commonly told that our application design is the cause because there is so much white on the screen. A quick bit of research confirms that the devices use an LCD screen with a constant backlight. This means the screen colour does not affect our battery life so the problem must lie elsewhere.

Having been our first venture into the compact framework and device led operations, the application was initially written to read web services. To keep things simple in the early stages, all web services simply exposed a Data.DataTable object. Looking back, these are pretty big objects and serialize a lot of data over the wireless. We have since removed the majority of these and replaced them with purpose built serializable classes (with short property names to optimize them even more!)

We have also invested in better wireless antennas for our wireless infrastructure, this has seen some battery life extension on this site.

A thorough walk through of the application has also highlighted areas for improvement, reducing network traffic and CPU time on the device.

I think the moral of the story here is not to always assume the hardware is the limiting factor to problems; it's always worth a second look at internal practices to see if improvements can be made there. I still believe the WiFi infrastructure on our other site is superior (resulting in better signal strength and less power needed to maintain a reliable connection) but I will continue looking for more ways to squeeze battery life out of these devices.

Update 26 Jan 2012: Since making changes to the amount of traffic the devices download over the wireless network, the battery life has significantly improved - despite this now meaning more processing on the device itself.