This is an excellent article that will reveal some very interesting things about power draw while navigating. WARNING: You may not like what you read but it's very good to know.
After personally experiencing having my phone (A Verizon GN) end up LESS charged than I started out with on a 1 hour trip with navigation on while plugged in, I decided to run some tests.The first was to see how much power the Galaxy Nexus could draw from the USB connection. This turns out to be about 4.5W. (0.9A at 5V)
(Edit: Yes I tied the data lines together so the GN knew it was hooked to a charger and not a PC)
I ran all kinds of applications, benchmarks, turned on all the networking, GPS. Just sitting there idle or in sleep mode it only drew 3.5W but I could push it up to 4.5W which is fairly close to the max most USB adapters can supply. My PS can deliver 8W so it had plenty of power to offer. So no matter how powerful your charger is, the GN simply won't draw more than 4.5W. Forget about those 3A chargers.
The next test was a bit trickier. I wanted to measure how much power the GN was drawing from the battery. Tiny contacts and an expensive phone made me a little cautious but I figured out a fairly safe way of doing this, and here are the results.
In sleep mode it only drew 0.04W with occasional spikes of 1W.
Now the real test... Google Maps.
- Powered on and idle it averaged 3W.
- Running a CPU benchmark it stayed at 4W.
- Downloading a large file over WiFi got a steady 4.2W.
This varied a LOT. Minimum was 4W with some steady peaks of 5.1W!
That seems to be the problem. The GN can draw more power from the battery than the charging system can provide. In the real world the power draw is going to be worse because of all the switching of cell towers and map data downloading. I wasn't about to take my very delicate wiring job on the road, so will just have to guess at the final numbers, but just sitting still was enough to overpower the charger.
I doubt this can be a software fix where a hardware register can be changed to let the charging system draw 1.5A or something.
This also means that a 2A or 3A power supply is not going to help. As long as you have a 1A supply, that's all the GN will take. Do make sure it's wired right or it will only draw 0.5A, and be warned some USB jacks on stereos and computers might only give you 0.1A.
Now on to something else... the AMOLED screen! Does keeping it black really save power? The answer is... sort of!
I ran several tests with the device in airplane mode and just showing a solid color on the screen on the dimmest, and brightest screen settings.
For extra fun...
- Dim black: 1.8W
- Dim white: 2.0W
- Bright black: 1.9W
- Bright white: 2.9W
So if you run the screen at the dimmest setting, it really doesn't matter what the screen is showing. If you run it at the brightest... then white really does draw more power. A lot more. If you can set Google Maps to display the screen mostly black, that could save a significant amount of power during navigation.
- Bright Red: 2.5W
- Bright Green: 2.6W
- Bright Blue: 2.8W
I was unable to find a way of turning off the display without putting the phone to sleep, so no idea how much of that power shown above is due to the screen and how much to running the CPU and all the other functions.
Anyway, I still have my spare battery rigged up so if you have any other good ideas for tests I can try them out.
(Edited for formatting and added in color power consumption)