Me too. Can’t wait to get mine. Got a while to wait still though. I’m an Oct 18 order.
Is it normal to manufacture a phone which respects your freedom nowadays and is recommended by the Free Software Foundation? Is it typical for a phone to have 6 innovations in comparison with all previous devices? How many such phones do you know? The path Purism have chosen is the hardest, so – at least to me – this is just fine to have all those problems, I expected them.
Also, it’s not necessary to read all the forums, you can just ask like OP did.
2 Likes
they were ironic questions 
I didn’t intend to flame the Original Poster nor anyone else here. My post was to try to turn-around some paradigms that may not be necessary nor useful in some cases. In some cases, maybe those paradigms might be important for real reasons. That’s fine too.
I can’t see why it’s so important to go a whole twenty-fours on a charge or a whole week on a charge, just for the sole reason that such is possible on an Android or Apple phone. It should have more to do with how you need to routinely use the device rather than needing to reach what might be more of an emotionally-driven benchmark that is a marketed feature of other phones. To me, plugging the phone in more often in exchange for a real Linux root access is a good trade off. It takes maybe five seconds to plug in the phone. Of course you do need to have access to an external source of power and if you can’t get that access, that could be a deal killer. I thought that the Tesla example was a good one. If you rarely travel long distances and have a not-too-bad work around for distance travel, then the inability to go 400 miles on a five-minute refill becomes more of an emotionally-driven desire as opposed to a practical one. You could say “well, I only drive fifty miles per day to and from work. But I am used to five-minute 400 mile refueling in my older vehicle. So no way, no Tesla for me, even if I only leave my local area once per year and never travel more than one hundred miles in any normal day” (and there are charhing stations everywhere if I ever need them). In this example, people purchase a Tesla because they like other benefits that it offers and the rare-case need to implement the work-around rarely affects them at all and when it does, it has no significant effect because they have a good work-around (Like a spare battery for the Librem 5 or plugging it in at certain available times).
One reason I leave the phone charging through the night is because I also need it to stay on all night while it charges. If a family member or close friend has a crisis at 3:00 AM, I want them to be able to reach me then since I have no landline.
Does anyone know if there is a way to charge the Librem 5 batteries outside of the Librem 5? If a second battery can be charged while you’re using the other battery in the phone, that increases your ability to use phone for long periods of time. I could probably carry four or five L5 batteries in my jacket pocket (a separate baggie each to keep them from touching eachother electrically). Five batteries plus one in the phone gives you more than twenty-four hours of non-charging time of use. But I will need a five battery separate charger to keep them in at night, while they charge. Should it hurt my pride or ego that my phone has one such limitation when I am one of the few people on the planet with a freed phone?
The first time I plug my Librem 5 phone directly in to an overhead projector in a business meeting and put up a real Linux desktop, everyone there is going to ask how I did that. They’re probably not going to ask about the battery life or notice the slightly larger size of the phone. I’ll bring in a mini-mouse, a roll-up keyboard and charger (in my pocket).
Has been discussed. Probably. Don’t know for sure. To be determined.
Four or five spare batteries, with their four or five dedicated chargers, might be overkill but OK.
Rather than using separate baggies, another approach could be to keep the batteries bound flat together. A baggie has the advantage of stopping the battery touching something else.
Yes, with the four or five battery strategy, I would only be willing to charge them all at home over night and would expect to have one single charger that I could drop them all in to at once. If there is no external charger available in the market, there are a lot of people here who could build their own. Anyone can carry a few extra batteries around without any significant burden. Most of us have at least a few hours for a part of the day to plug the phone in to a power source. I have a Note 9 and still have a USB-C charger in my car and one in my office at work and one next to my bed. I rarely use the one in my car or at work. But if I need them for rare occasions, they are there. I never carry a charger around with me on my person. I just plug in if/when/where needed. A part of the baggie idea is flexibility. I wouldn’t have a large solid object in my pocket. I could take only as many or as few as needed. Most days, I would leave them at home and plug in more often. I might keep a spare battery at work or in the glove box of my car.
1 Like
AFAIK you can also replace Librem 5 battery without rebooting, so why not use it as a charger?
1 Like
One scenario might be where you’re away from any ability to charge your Librem 5 for several hours at a time. You need your phone with you, but you can’t charge it because you’re walking around a lot while using it. So after several hours of heavy use, you go back to your car or your office and swap out batteries. You take the battery off of its external charger and put it in to your L5. The battery that you just used up in the L5 then goes in to the external charger. Four hours later, you do the same again.
You can do that with a simple cronjob, since you have control over the charging chip.
1 Like
the cronjob I know how to use but I have no idea how to manage the charging chip: could you give me some useful information for me who are fasting electronics? Thanks
regards
Unfortunately I have no skills to do so. So I put my idea into this community in hopes that someone find it useful and important enough to take this job. But I also hope to find other ways to give something back to this community, once i got the device.
1 Like
Let me stop you right there: is it normal to preorder or crowdfund a project that promises to deliver a fully working phone, then get scolded in the comments because you’re unaware that that promise hasn’t been delivered upon when you finally get the phone in your hands?
Is it normal to send out the “work in progress” without including some note warning the user of the current issues, reassurance that they’re being worked on, and some links to more information? To temper their expectations a bit and soften the blow a bit?
On the other hand, I think it’s perfectly reasonable to preorder a phone and waiting 3 years without checking up on the status in the forums. When I give you money in exchange for a promise of a future product, I expect you to keep up your end of the bargain without me having to constantly check in on your work towards fulfilling that promise. And if something serious crops up, I expect you to keep me in the loop. Not me having to find out through a third party, or having to chase that information.
When this guy paid for his L5, he ordered what at that time was presented as a Linux phone. Nobody expected it to have the rich app ecosystem of Android or IOS, but everyone expected something that would at least function flawlessly as a dumbphone (phone, SMS, able to hold a charge for a day), with some basic functionality like calendar, calculator, camera etc…, but little else. At that time, there was no talk about crappy runtime. All that talk came later. So no, as an early backer, he has absolutely zero obligation to appease the fanboys by not complaining about the phone not living up to the expectations Purism set during their crowdfunding campaign and preorder phase. He didn’t pay you, he doesn’t owe you squat.
What is wrong with you that you think it’s not only OK for Purism to overpromise and underdeliver, but for you to get hostile towards people who voice their disappointment?
4 Likes
Was thinking along the same lines. Another nice option down the road would be a toggle switch between “charge to full potential” and “prolong battery service life”. The reason EV batteries don’t wear down as fast as cellphone batteries is that they don’t charge to the absolute maximum limit, but stay well below that, where there’s less wear and tear on the battery chemistry. For those of us who would get enough runtime out of the battery-optimised setting, it would be nice if we could prolong the service life of the cells.
But best of all would be if we could get full control of the battery charge characteristics (within safe limits), at least while the phone is on. That way we can implement our own stuff. For example, it would be cool to implement a “discharge to storage capacity, then shutdown” for when you don’t expect to use the phone for a prolongued period of time. Like we do with our LiPo packs in the RC world.
[EDIT] Ah, I see that’s possible. Thanks for the info, @dos. That’s gonna be fun to play with. 
1 Like
I agree that Purism did set unrealistic expectations, by promising to provide a privacy/security phone for normal users. To be fair to Purism, however, it had little way of knowing in August 2017 what problems it would be facing. The i.MX 8M Quad didn’t start mass production until Jan. 2018 and the first SBC’s based on the chip didn’t ship until March 2018. In Aug. 2017, it would be hard to predict that the mainline Linux drivers for the i.MX 8M Quad still wouldn’t support suspend-to-RAM and the MIPI CSI2 camera interface in Jan. 2021. If you are mad about the poor battery life and non-functioning cameras, then you have to honestly address the question of whether you would have been willing to live with the performance of the i.MX 6 or A64, because Purism would have had to use an old chip which already had well-supported mainline drivers.
We can criticism Purism for marketed the Librem 5 as a privacy/security phone for normal people, but we have to honestly address the question of whether Purism could have paid for 3.3 years of development if it had marketed the Librem 5 as a early-adopter dev device for Linux geeks. How many pre-orders would the phone have generated if it were marketed to reflect its current development status? To be frank, I probably would not have pre-ordered the Librem 5 if it had been marketed as a dev device for Linux geeks, because I didn’t understand all the issues at play. Today I understand the importance of developing Phosh and libhandy for the future of mobile Linux, and I understand that supporting i.MX 8M is vital for the future of RYF devices, but I didn’t understand any of that when I pre-ordered the Librem 5. All I knew is that I wanted a Linux phone and hated surveillance Capitalism.
The Librem 5 actually has pretty good runtime. What it lacks is suspend and wake on call support. If we are going to criticize Purism on this point, we should also be criticizing NXP for failing to provide good mainline Linux drivers for the i.MX 8M and for failing to properly document its CSI2 camera interface.
We should also look at how long it has taken Crust to provide good power management for the A64. The A64 started shipping in June 2015, but development of Crust didn’t start until Oct. 2017, and only recently did Crust get good enough that the PinePhone can last 2-3 days on a single charge. If we are going to demand a phone that runs on 100% free software, it is unrealistic to expect it to function flawlessly when released.
Purism’s marketing of the Librem 5 did set unrealistic expectations, but it seems to me that we have to acknowledge the economic realities of funding development and the fact that mobile Linux has little hope for success if some company doesn’t step up and fund its development. None of the major Linux companies care about mobile Linux, and anyone who thinks 100% volunteer labor will make Plasma Mobile or Ubuntu Touch into a viable alternative to Android and iOS needs to look at the 75 commits that Lomiri has received in the last year and try to use Plasma Mobile on the PinePhone.
3 Likes
I see what you’re getting at, but that’s implying it’s OK for a company to accept people’s money under false pretenses because there’s a higher goal at stake. I find that a very dangerous line of reasoning.
[EDIT] That said though, I don’t think this is what they had envisioned either. Maybe this is just me projecting my expectations when I backed, but I think the idea was to ship a phone that would have all the basic functionality working pretty decently, but with a rather limited selection of phone-friendly applications.
Ooh, that’s great news! I’d be interested in more info on that. Which distro do I need to run to get those runtimes out of mine? Because last I checked, it was still around 8h idle, screen off, Wifi on (but turns itself off after a while, after which you have to manually disable and re-enable it for it to come back on), and no SIM inserted. Haven’t seen any system (e.g. kernel) updates on that UBPorts it came preinstalled with for quite a while now.
[EDIT] Apparently, I want to be on Manjaro + Posh then, correct?
Idea for someone: Break out a Arduino, add a power transistor as a switch, male and female USB connectors, and write some simple timer control software to get the phone ready for you when you wake up. You’d have an aftermarket product and I would buy one.
USB power --> switch controlled by Arduino timer software --> PHONE
Sure, here’s the datasheet of the charger: https://www.ti.com/lit/ds/symlink/bq25895.pdf
The most useful section is 8.4 Register Maps. You can talk to the chip with i2cset/i2cget. Be careful though, this is a low level tool so it will happily let you shoot yourself in a foot!
You can check the current value of a register with:
sudo i2cget -f -y 3 0x6a <REG>
and set a new value with:
sudo i2cset -f -y 3 0x6a <REG> <VALUE>
If you want to disable charging and let the phone operate out of USB power, you’re interested in register 0x03 (page 35) and its CHG_CONFIG bit.
purism@librem5:~$ sudo i2cget -f -y 3 0x6a 0x03
0x1e
purism@librem5:~$ sudo i2cset -f -y 3 0x6a 0x03 0x0e
purism@librem5:~$
This disables charging, but will discharge the battery in case USB doesn’t provide enough power. You can check USB input current limit with:
purism@librem5:~$ cat /sys/class/power_supply/bq25890-charger/input_current_limit
900000
This shows that the current limit is set to 900mA. You can set it in register 0x00 (IINLIM, page 32). For instance, this sets the input current limit to 2.5A:
purism@librem5:~$ sudo i2cset -f -y 3 0x6a 0x00 0x30
purism@librem5:~$ cat /sys/class/power_supply/bq25890-charger/input_current_limit
2500000
purism@librem5:~$
Make sure that your charger is able to provide the current you set there!
You can also completely cut the battery from the power path, so it doesn’t charge or discharge at all and the phone operates solely on USB power (in which case it turns off if it tries to use more power than provided via USB) - see the BATFET_DIS bit in register 0x09 (page 41).
No worries, not that long ago I also didn’t know how to do it
11 Likes
thanks ! I will try to exploit what you tell me.
Thank you again.
regards
People commenting on the PINE64 forum say that they are able to get 2 days of battery life with recent Arch/Phosh, but I couldn’t find commits in the Crust code that are doing this, so maybe it isn’t Crust, but something else which is allowing this. At any rate sleep and wakeup are now more reliable.
FWIW, I can ssh into my phone (LineageOS) while it’s “sleeping”. It usually takes a while before it answers the request, but it shows conclusively that it’s not just “suspend-to-ram”. I don’t actually know if it’s “suspend-to-ram with regular wake-ups” or if it’s running at a super-low clock rate with the wifi put into a form of “sleep and check for packets every few minutes” or what.