I’ve had my oneplus 3 for like 5.5 years and the battery still lasts all day.
1 Like
Based on an e-mail that went out to many of us, you should be able to get a refund now.
I’ve had my iPad Air since November 2013 and the battery still lasts for like 10 hours (and of course I would like to replace it with a GNU/Linux tablet some day). The charger gets warm during charging, then it cuts off when the tablet is charged and it remains cool to the touch even if I forget to unplug it. Does anyone know if Purism battery-powered devices do this? @nicole.faerber @MrChromebox @mladen
I don’t want to talk about battery run time now since this is a complex issue.
What I can tell though is that you can expect a similar behavior from pretty much all Librem devices, i.e. during charging the DC adapter will get warm since it needs to supply significant power to the device for charging and operation. After the battery is fully charged the current goes down. If the device is not active (suspended, screen off etc.) current will stay low and thus the charger will cool off. How much depends on the further power draw of the device. If the device is disconnected altogether but the charger remains connected to mains it will cool off.
Cheers
nicole
2 Likes
Thanks, Nicole!
So is there any problem just leaving a Librem device plugged in to the charger?
I ask because I did leave my Librem 15v4 on the charger a lot, and I’m wondering if the battery died because of that. With the new battery I am setting GNOME Clocks to remind me to unplug.
This give me an idea, also because everyone seems to be out of the stock of the Librem 15 battery (U753-TS44-111). Well, the Purism shop now shows a used one in stock. The Librem 5 battery is 4500 mAh (not sure of the voltage), 100 mAh more energy than the Librem 15 battery: 4400 mAh at 11.1V. What if all Librem devices could use the same battery (or multiple batteries in series)?
My Librem 13 version 4 is plugged into the charger 24/7 (except for rare occasions that I take it out somewhere).
Leaving a device plugged into the charger should never be an issue. The charging system should monitor the state of the battery and limit the charging current as the battery needs. If it doesn’t do that it means it is flawed or broken.
4 Likes
Exactly, and I ask because elsewhere around these forums floats the idea that it is the user’s job to monitor the charging or the battery will be ruined. So I’m asking the hardware gurus.
1 Like
Woohoo, OK, now we are indeed going down one of the battery rabbit holes 
So first about the batteries themselves. I’m afraid it’s not really possible to share much of these between devices, especially between the L5 and the laptop(s) - totally different package, voltage etc. Also in these mobile devices (phone, laptops) you will always try to make best use of the available physical free volume in them for battery volume - volume in the end means capacity. So you design these battery packages to fill this volume as good as you can. We could probably design a kind of holder that can carry three or four L5 batteries in parallel to hold them in the L15 or L14 battery compartment (a pretty cool hack idea but this will end up in less capacity than the original packs. And it would need extra electronics. The laptop battery packs put together multiple cells, the L5 battery is just one cell so you need to add logic to combine multiple of them. (and I think the L5 battery is already thicker than what would fit in, say, a L15).
So while I see the point I am afraid, meh, not going to work well 
For the charging indeed recharging a Lithium based battery all the time to 100% reduces its life time. The way this is handled in a device depends. In the L5 this is handled by the Linux kernel charge driver. In the laptops this is handled (and thus controlled) by the so called Embedded Controller (EC).
In the L5 we have a small threshold for recharging but kept this low so that you can recharge it to full easily. With the L5 we assume that it will be used most of the time without the charger and when it gets connected it is supposed to get charged, that’s what most users will expect to happen. So that’s why the threshold is low.
In the former laptops (L13 and L15) we did not have much control over the EC - or pretty much none. It was not yet free, totally proprietary and we could only take it or leave it as-is. In the L13 and L15 the recharge threshold is also very low, i.e. it will pretty quickly start to charge the battery to 100% again, even if it is just barely discharged. From what we know today this is not really helpful for the battery - but also sadly very common in the industry.
That’s why we put a lot of effort (and money) into the Librem14 and freed the EC there, sourcecode is here:
This is the actual code we ship on the Librem14 from the very early beginning. This free EC firmware by default implements a 90% threshold, i.e. only if the battery capacity is below 90% it will start a new charge cycle. By default it will then charge to 100% to gain full capacity and run time. Ideally you would set these even more conservatively! On my L14 that I use for my daily work I set the start threshold to 40% and the end threshold to 90%, i.e. it will only start if the capacity drops below 40% and it will then only charge to 90% of the total capacity. This works totally fine for my daily use cases and it will extend the overall battery life time a lot. In the L14 these threshold can be controlled by the user from Linux commandline (and are then stored in NVRAM).
Cheers
nicole
5 Likes
Wow, the Librem 14 is more excellent than I thought. Bravo x 2! I told a friend about it, she bought one, she got it about a month ago, and she is very pleased.
Thanks for the detailed info!
Since I’m on a Librem 15v4, my Unembedded Controller is GNOME Clocks. 
I guess I should write a little utility to use the output of upower and pop up a GNOME shell notification saying “Unplug your charger!” One pops up when the battery is low at 10%, but not when it’s high. Maybe someone has already done that…
UPDATE: Someone HAD already done that - with a GNOME extension: Full Battery indicator. I modified extension.js so the notification pops up at 95% instead of 100%.
Question for @nicole.faerber: is 95% charge too high for maximum battery life? You mentioned 90%. I guess I need to go look at Li-ion battery life curves.
3 Likes
Battery University’s article on prolonging the life span of Li-ion batteries says:
Most Li-ions charge to 4.20V/cell, and every reduction in peak charge voltage of 0.10V/cell is said to double the cycle life. For example, a lithium-ion cell charged to 4.20V/cell typically delivers 300–500 cycles. If charged to only 4.10V/cell, the life can be prolonged to 600–1,000 cycles; 4.0V/cell should deliver 1,200–2,000 and 3.90V/cell should provide 2,400–4,000 cycles.
I assume this info is for car batteries, because most phone batteries are rated at 3.8V, but the same principal applies that lowering the voltage will dramatically extend the cycle life of the batteries.
Most laptops, tablets and phones used cobalt-lithium-oxide batteries, since they have the highest energy density (by weight and volume), but they also degrade the fastest of the different types of Li-ion.
2 Likes
Thanks @amosbatto, good info. Right now I’m at 77% charge on a Librem 15v4 and the pack voltage is:
$ upower -d | grep voltage
voltage: 11.76 V
$
Since the battery is a 3-cell pack, 11.76 V/3 = 3.92 V per cell.
It will be interesting to see what the voltages look like as the pack charges and what the peak voltage ends up being.
2k? I’d buy a macbook pro instead
Yes, he rather missed the point.
1 Like
Update to this thread:
I received my confirmation email on 06/10/2021.
I requested a full-refund.
3 Likes
Keep us posted, I’m curious how long the refunds are taking to process. I mentioned in another thread, I received my confirmation on 09/29/2021. I’m electing to receive my device still.
“Hello Arthur,
I have created a ticket for your order to be canceled and refunded.
Please keep in mind that it will take several weeks for the refund
process to be complete.”
Wow.
08/11/2021 - Still no refund received.
Not acceptable considering their purported influx of cash! SMH
I can’t think of any legitimate reason that a refund should take more than 3-5 days. Can anybody provide a non-shady reason for this?
1 Like