Another Battery Test

This time, I tested the longevity by charging to full, turned off all three killswitches, set the phone aside, then only activated the screen occasionally to check the charge (except when I was sleeping).

The battery drained from 100% to 3% in just over 14 hours.

My final check of the power level at about the 14-hour mark actually showed a remaining charge of about 35%, but that was apparently inaccurate, because a few minutes later I noticed that the phone had shut down. When I turned it back on, the level was at 3%. So the charge indicator is still way off, as has been explained before.

I have this phone on WiFi most of the time, with cellular sometimes on and sometimes off, and I’m getting from 6 to 8 hours comfortably, depending on what I’m doing with it.


FYI, it took about 4 hours to recharge after that 14-hour drain.
(And yes, I’ll probably have to buy a new battery from Purism if I keep this up! Lol!)


Thanks @amarok. I added your info to the FAQ:

When you buy an extra battery, please share if you find a generic charger that works to charge it. One good thing is that you can use the Librem 5 without a battery if plugged into the USB cable, which means that you can hot swap the batteries without turning off the phone.


I recommend adding “from a nearly depleted battery” to your sentence here:

“4 hours to recharge the battery”

Because it takes much less time to recharge when the battery isn’t so empty.

1 Like

My first test (using different parameters) yielded 10 hours:

The difference being that I did not have the killswitches off.

1 Like

I changed it to this:
“He also reports that it took 4 hours to recharge the battery after draining it down to 3% charge.”


With the disclaimer that until the percentages shown are close to accurate, it is difficult to be certain about the details.

The most definitive test is: Put the phone in the state (HKSs) that you want to use it, fully charge it, leave it completely idle until it shuts down. That’s the number of hours, regardless of whether the percentages shown are accurate.

For me that would be two tests, roughly speaking - one for home (nothing killed) and one for out-and-about (WiFi killed).

for me the most important part about accurate battery reporting is to not deviate too much from reality between 25-33 % since i don’t plan to leave the El-Cinque battery drop bellow that point. i’ll use any means available at my disposal to prevent that from happening since i want to conserve the battery cycles as much as possible.

good stuff ! carry on !

Very informative topic! reference

I am glad that someone out there is testing out the battery life. Usually, I would get amarok’s estimate in battery charge. However, I recently have the hub device. That hub really took a drain, reducing the maximum charge within 4 hours estimate. I guess that hub devices are power hungry devices. I would make note of the hub’s temporary use (exclude hub connection when charging, hub consumes too much recharge rate). I also hope that the XFCE desktop environment (interface) is considered for PureOS of Librem 5. I used the GNOME DE a few times with PureOS of desktop. Despite its pleasing looks, GNOME demands plenty of power and probably state of the art tech for flawless operation(s). Yes, it is true that PureOS uses the 64 bit data architecture. It doesn’t mean it has to be glossy in its development goals.

Here is my idle phone plan.
I intend to keep the phone idle with only a battery (no hub connection), a disabled cellular connection, a enabled wifi connection, a disabled bluetooth connection, zero screen brightness, zero audio volume, and a active web server. My goal is a idle phone running only on battery (no charging, no hubs) as a web server with 8 hours estimate uptime. While cellular connection is very convenient at times, it does not have the web hosting requirements. Unless, I am okay with ethernet connection, which is on the power hungry hub, which is better off with charge, I am stuck with wifi connection. I also believe that screen brightness demands a lot of power consumption, not to mention stacking overheat, which near bricks OS operations. Same can be said for audio volume. As for the web server, I am sure that its interaction may affect both data usage and power consumption to a degree. I am not sure if there is a heavy host by hosting a web server on the Librem 5. I have less confidence in application servers such as game servers. Imagine the constant data transfers! Imagine the burnouts! Anyways, I hope that I can achieve 8 hours estimate uptime with this plan.

In the case of failure, I consider to connect a power bank to the idle phone with the set parameters. The power bank is my contingency plan should the phone loses all battery charge due to variable(s). However, I do not have first-hand experience with the power banks. I recall that there were two types of offered power banks. I would like to know more information, specifications, and performance of the power banks.

If someone can provide feedback to the post context/content, it would be most helpful to my situation. I know this post is enduring to read. Thank you for your consideration.

It’s true that I have yet to conduct performance tests. I have too much thoughts to work in a strict mental state. It would be better to write them down somewhere. Once I implement my plan(s), all I can do is to hope for the best.