An update on the Farphone's battery
far.computerThe real question is "what the hell is a farphone"?
Apparently not a misspelled Fairphone, as I originally thought. I wouldn't mind an article about that battery.
No actually, it is a Fairphone after all.
Well, it is a misspelled Fairphone :) You can see in the picture of the phone that the (paint? sticker?) of the letter i has fallen off
Your answer lies on https://far.computer
In short, far phone is the phone which powers far computer which is in turn served from https://far.computer
Funny. I really thought it was a mistake :)
> this webpage is hosted on a drawer-bound fairphone 2, running postmarketos
“Farphone” is a given name of a specific fairphone (android smartphone) that the author uses to run their web server.
The name likely comes from “fairphone” with the “i” scraped off - see the photo: https://far.computer/
These days it is rare for a phone to be able to be used without a battery. The reason is that the max energy consumption when the CPU and GPU are running 100% exceeds the wattage that the device can accept over USB PD.
I've had success using a large capacitor instead of the battery. To keep it charged I connected its positive leg to the 5v USB pin through a diode.
I was pleased to discover my old PinePhone allows this. It makes development much easier when having to swap the SD card every few minutes, and allows for simple power cycle via USB switch.
What I cant figure out is how to detect power usage from the PMIC when in that configuration. ie seems to still assume power draw happens via a battery.
That can't be true right? PD (and some Chinese standards) have insane wattage allocation/allowances, there's no way that a mobile CPU can pull over that amount, it's more that they don't support it.
> device can accept over USB PD
Accept, not what the power supply can supply. Cell phones aren't being made with massive laptop-sized 100W accepting circuitry.
> Accept, not what the power supply can supply. Cell phones aren't being made with massive 100W accepting circuitry.
Surprisingly they are, supervooc goes up to 120w now (which is insane), there is also a Motorola phone which also does 68w via PD standard
https://m.gsmarena.com/oneplus_15-review-2898p3.php
https://www.gsmarena.com/motorola_moto_g_stylus_5g_(2025)-13...
As for past through charging, it's actually apparently more common then I knew.
Google does it as part of their "adaptive" charging (hits 80% then does pass through) and the OnePlus etc also do it with a settings for gaming.
(Older article confirming this https://www.androidauthority.com/phones-with-bypass-charging...)
Unlikely a web server would see such usage patterns.
While I’ve seen plenty of swollen and deformed phone batteries, I’ve never personally seen one that has burned. Obviously it’s happened in the past with certain phone/battery models, but I’d imagine that it’s actually very rare now days?
On the other hand, I have seen cheap 18650s spontaneously start smoking even when they weren’t plugged in to anything…
You can’t be too cautious with spontaneously combustible stuff.
So the Fairphone 2 runs on just a USB cable with no battery inside?
I'd like more details. Many devices won't run on USB cable alone: they won't start without a battery as well.
"bypass charging" is what you need to look for. theres a small few devices that have support for it at the moment
Fairphones are designed to be modular and repairable, I'd imagine it's much more robust than the average smartphone.
Yup that's right!
I would have hooked the smartphone to a small solar panel. The natural daylight cycle would have made sure that the smartphone kept having charging and discharging cycles.
I doubt the traffic hitting it would be sufficient to drain the battery overnight.
That's what the project cited as inspiration does: compost.party
an even smaller https://solar.lowtechmagazine.com/about/the-solar-website/ ?
Why not just have the charge controller "unplug" it if the battery is full?
All of the controllers do that! But then battery starts to self-discharge, eventually a controller detects "huh, the battery is no longer charged" and start charging again.
Over years, this can accumulate enough charging cycles so battery gets worn down.. And old batteries have even higher self-discharge, so the cycle accelerates. If you are lucky, the battery lasts long enough. If you are unlucky, you end up with "spicy pillow". If you are super unlucky, and charger's temperature sensor fails (or was never installed), or battery gets punctured - you got a fire.
Not all of them apparently. I'll have to dig for a schematic(they exist, but places want money), but it seems my Dell laptop from 2019 uses the embedded controller as the BMS.
As, somehow it managed to turn all four cells in the pack into pillows. Which indicates a shockingly flawed balancing system
Because the charge controller likely does not run software that can easily be modified by the end user?
I would’ve expected a hardware-based lithium-ion charge controller which would continue to work regardless of what software runs on the main CPU(s).
Does Farphone run Far Manager?
[flagged]
"When disagreeing, please reply to the argument instead of calling names. 'That is idiotic; 1 + 1 is 2, not 3' can be shortened to '1 + 1 is 2, not 3."
I have multiple devices with lithium batteries plugged in 24/7. A kindle that I use as a display for example. So far nothing exploded. If exploding kindles were a thing I guess I would have heard.
I have not had anything explode, but I have had Macbook batteries expand on me on two separate occasions to the point where the case was visibly warped. Both times I was away from home, so it was extremely inconvenient.
you use a kindle as a display?
For home automation, how much energy my solar system produces, birthdays, weather forecast. Stuff like that.
did you flash it or do you just use the browser?
I flashed it. Don’t know if always on works with the browser. But jailbreaking is easy and I think even reversible. Then you can ssh into it and even setup a cron to download a picture and refresh the screen.
that's cool
i may have a case to buy a kindle then
any model recommendations?
The fact that they can doesn't mean they will.
On older devices the controller might make some assumptions that holds true with a new battery, but very much doesn't with an old and worn one.
My Macs have all been sensible about it, but I've seen Windows machines with batteries that just died from being plugged in all the time not even 10 years ago. Even if that specific instance was just a bad battery and not due to a charge controller, I have no faith in Random Windows or Android OEM Number 582 doing this correctly.
For devices that are fixed, I'd prefer to eliminate the potential of there even being a problem in the first place.
There’s basically zero risk for these cell phone batteries outside of freak accidents, speaking as someone who who’s been building packs since pouch cell Lipos first started coming out for model airplanes back in 2008/2009.. That’s because in a single cell configuration, there’s no way for the charge controller to run up an imbalance and overcharge one of the cells.
I somewhat agree with you. As my last comment suggests, I have a lot of experience running phones as AP’s including phones with dual cell configurations.
Where things go off the rails is situations where extreme heat can be present (shoving phone in direct sunlight in window with hot climate is a bad move) another thing they don’t tolerate well and people don’t talk enough about this is deep discharging the batteries frequently. This causes a breakdown of the SEI membrane and makes it so future recharging generates more heat and gas. This will cause expansion and might cause a short/failure if poorly designed (galaxy note 7).
Three times I've been lazy and set up an old phone or tablet as an always-plugged-in stationary device without excising the battery, and that has produced two spicy pillows and one completely dead battery (phone wouldn't even boot when plugged in, until I replaced the battery). Granted, these were all 5-10 years ago, but I do not trust the batteries and their controllers in these devices.
Nowadays if I want to leave a device plugged in I crack it open, remove the battery cell, solder on a power supply and capacitor, and then do the nonsense with rooted Android to keep it from shutting itself down.
fwiw I’ve used 24/7/365 plugged in phones as AP’s in multiple locations for a decade or so now, never had an issue. Past few years I use the battery threshold to set them to 70% charge and they don’t move from this for months at a time.
What roasts the lifetime of my laptop batteries is compiling with gentoo, but again never an issue with catastrophic failure and I have 20+ years of experience with that as well.
I've gone through a dozen or so LiPo-utilizing portable devices at my property in the Mojave desert. All it takes is a single season for many of these batteries to swell up to such an extent the enclosures split open.
Ostensibly they contain charge controllers and temperature sensors, yet they're unable to prevent this outcome when the ambient temperature exceeds 110F day after day while the device stays on in a hot attic w/usb-c pd connected.
Fortunately I haven't had any burst into flames yet, but after a few years of seeing this pattern repeatedly I stopped deploying anything containing LiPo batteries at the property.
YMMV - but IMHO it's prudent to exclude these batteries from such unattended, powered 24x7 devices.
This has nothing to do with being near full charge and everything to do with the temperature.
Leaving such devices plugged in 24x7 unattended creates opportunities for such ambient conditions to develop unnoticed.
These mobile LiPo-using rechargeable devices simply aren't ideal for being left powered while unattended, day after day, year after year. All it takes are some sunny days with normally closed blinds left open and you may slowly be cooking that device you forgot about that Just Works but is a ticking time bomb after the 10th day it spent an hour in direct sunlight while charging.
AIUI the high temperatures are particularly damaging to the SEI (Solid Electrolyte Interface) layer, which is the layer preventing dendrite growth. If that layer breaks down enough for dendrites to grow, you're in internal short / fire-risk land.
Excellent advice. Did you swap any of the cells with a different chemistry?
Not really, there was a brief excursion in kludging a ZTE MiFi device to use a DIY NiMh pack of AA cells when it refused to stop self-destructing its OEM LiPo batteries every summer. (I use a MiFi hotspot for a cheap security camera network)
It worked as a stop-gap but I've since replaced it with a GL.Inet X300b ruggedized hotspot without any batteries.
There's no UPS for now... if I went the route of wanting uninterrupted power at the property I'd probably put a battery bank underground outside to power the entire building. It's not worth risking anything rechargeable inside the place given how hot it can get, and how long I sometimes go without visiting.
Cheers, Thanks for the info.
I'm not sure how generally applicable it is. When you have several acres of undeveloped land full of sand at your disposal it's relatively trivial to dig a pit, mix some concrete from the sand you excavated, and pour a subterranean cellar to house a battery bank and other hazardous infrastructure. Nobody would even notice it happened/exists.
The situation is far more complicated for folks in apartments or high density housing.
Most can, but you do get reports that sometimes they don't, and better safe than sorry.
I'd guess it would have more to do with heat, though.
That's silly. Batteries don't like to be kept at 100% all the time, not unlike your lungs which doesn't want to stay filled all the time (which is uncomfortable for your muscles even if you ignore the carbon dioxide).
e.g.: MacBooks discharge the battery down to 80% by using the battery even if it's plugged in by citing "Rarely used battery", and keep the battery at 80% for at least half a day, then charge it again.
Li-ion is an adversarial chemistry. You need to take care of it or the battery bites back by puffing up or losing capacity very fast, or becoming an indoor firework.
For how long?
I have a fire tablet I hucked into my wall to use as a home assistant console.
It's been in there for 5 years now, always plugged in.
I've been using an old Pixelbook for 3 years that I had laying around for an always on Home Assistant control panel (but with adaptive dimming that defaults so extremely low brightness to reduce heat until approaching it).
I noticed a few months ago that the battery has started to significantly swell pushing the back panel apart, but I haven't worked up the motivation to try removing it yet. From what I can ascertain from user experiences, it actually does boot sans battery but haven't confirmed, it would be awesome if that worked and would probably give me 5+ more years of usage.
So in my case, data point is that 3 years of 24/7 use/charging of an old laptop/tablet was enough to push it over the edge and finally swell. It's really a shame how so many otherwise usable devices that could be wall mounted turn into e-waste because they won't run without a battery. With USB-C PD a well-designed device should be able to get whatever power is needed on demand but manufacturers don't really have any incentive to future proof for the .01% of users like me who would benefit.
> I have a fire tablet I hucked into my wall to use as a home assistant console.
OK, but "fire" is right there, in the name.
Right, but, this is just a single data point anecdote, right? Let’s say for same of argument that there is a 5% chance of overheating in five years - and another sub-percentage where this causes a fire, depending on where it is plugged in.
95%+ of people would report “zero problems here, all concern is overblown”.
Safety doesn’t work that way?
I have a Samsung tablet in a non-flammable container waiting to be disposed of after being mostly plugged in for 3 years then suddenly swelling quite alarmingly. YMMV.
Until the heat death of the universe, and a couple beyond.