Hello all!
Due to the recent statements by Google (as well as their track record the last few years) I’ve decided I do not want to use Android as a phone operating system anymore. But Apple is just as bad, if not worse. So I’ve decided to build my own custom device.
I am working on building a phone using a single board computer, right now I’m using the raspberry pi 5. This is still a proof of concept, but I want to share my ideas with others, so like minded individuals can start messing around with this idea in their own homes to further this goal.
You can view more images of the device here, as well as the step by step instructions here (these are still very rough and incomplete) https://github.com/muhammadmanwar/cheaphone OR https://codeberg.org/muhammadmanwar/cheaphone
Right now it just runs raspberry pi OS, with a different desktop look and feel. Everything that normally works in a pi 5 works on this device, additionally I am experimenting with a Mobile Broadband modem, to allow the device to text and call, as well as access internet, like a normal phone off wifi
The total cost is around 200 dollars, not including the 3d printer to make the custom case.
This project is barely off the ground, and I’ve got a lot to learn before I can stop relying strictly on the raspberry pi 5, my end goal is to custom design SBCs, and release those designs for free alongside the plans for the device, so that interested parties can select their own System on a Chip to use for the device. I need to get into designing boards, I’m interested in trying Stephen Hawes’ Lumen PnP (https://www.youtube.com/watch?v=JlkTcxh-9gA) for that phase.
But that’s for the future, for now, I’m hoping to get more people interested in the prototype so that I’m not the only one noodling around on this idea. I’d love some feedback, and if anyone was willing to put one together for testing, I would appreciate it greatly!
You must log in or # to comment.
I’m sure you are already aware, but just in case, there’s a lot of prior work in getting a truly Linux mobile phone.
There are ready-made devices like PinePhone (the PinePhone Pro looks the most promising one of the bunch), Librem 5, and Liberux Nexx. I think at least some of those companies publish schematics for their boards, you should probably check those out if you want to design your own.
There is also another direction, taken by postmarketOS and the like, to install Linux on a phone that shipped with Android out of the box.
It should be easy enough to install postmarketOS on your device, since it seems to have support for raspberry pi. The benefit of postmarketOS here is that it makes it really easy to install mobile Linux UI shells, like phosh, gnome-mobile, plasma-mobile, or sxmo. This will let you try all of them out and maybe pick one as a starting point for your software stack.
And the software side of it is the really annoying part. We’re missing so many components: Connected standby, an app lifecycle management, maybe sandboxing and a detailed, user-facing permission system. And then we need to go ahead and rewrite all the important Linux software to use these (not yet existing) interfaces.
I own a Pinephone, and I feel the Linux phone is within reach since the Nokia 900 (and its predecessors) and that was in 2009, so 16 years now. I believe any effort is very welcome, though. We badly need a good and free operating systems for this important device we all carry around and use hundreds of times each day.
Let’s not forget all the software that was done and kinda ready for the openmoko device. There were a few distros for that alone. Some were pretty stable but the biggest issue came down always to phone call quality and connectivity. I’m completely out of the loop but back then all the chips were so closed in themselves that they were basically black boxes. A huge effort was made to try and get them to work as well as possible, with varying degrees of success. I wonder if it continues being the same nowadays.
The N900 was such an amazing device, I still have it around and the UX was the best I’ve had in a phone, Maemo was truly great and way advanced to its time. That UX with the screen sizes we have today… I keep dreaming.
Phone quality on my Pinephone is nice except when it isn’t. We have a largely Free Software baseband running on the modem. You can tell it to run a high samplerate and it does VoIP and 4G. On the flipside it doesn’t always work and there were issues with additional stuff like echo cancellation, which then takes away from the experience. Openmoko was great as well. I think it was so very open, people could do silly things with the phone network. But then it’s old tech by now and the 2G (GSM) phone network is obsolete and they turned off the cell towers in most areas by now. And I think the software on it didn’t really translate to the phone generations after that. Afaik the entire software stack didn’t carry over and we’re using different dialers, messengers, calendars etc now.
- Connected standby is already somewhat possible (it’s actually done on the hardware/firmware side on most phones), it can work with something like ntfy with a fairly simple script IIRC
- We have sandboxing and permissions figured out pretty well with Flatpak (there are improvements to be had for sure, but all the basics are there)
The one main remaining barrier (apart from thousands of paper cuts everywhere and lack of apps), is indeed process lifecycle management. It’s the most complicated one to do, because in order to work well it requires apps to cooperate in some way, either by completely and honestly shutting down when not doing any work, or by providing ways to check if there’s any work to be done without running the rest of the app, or both ideally. None of the apps currently do that, so the only options are (1) just let apps do whatever they want, draining the battery, or (2) send SIGSTP to apps that are not in the foreground, losing background notifications and such.
Yes, good job at explaining the app lifecycle. I brushed over it because I had no idea how to phrase it in a concise way. And it’s really the million papercuts that stopped me from using the Pinephone. I tried.
Connected standby is already somewhat possible […]
Yes, it’s a very crude way, though. In practice I never got the messages from my wife to bring milk on my way home. I’d read them a few hours later in the evening after unlocking and tinkering with the device. What worked well is tell people to send SMS or call, because that reliably wakes up the device. But then people forgot they were supposed to communicate with me like in the 1990s.
[…] sandboxing and permissions figured out pretty well with Flatpak […]
We do. That’s how some modern distros like Fedora Silverblue work. But then it’s somewhat problematic with phones. These packages are supposed to come directly from the various upstream projects. And then the phone distros can’t patch them any more to deal with the peculiarities with the inconsistent ecosystem. So we’d either need to have a good platform first and all agree on it, or repackage everything in a way unalike how Flatpak is done today. And then it fills up storage fast with all the runtimes and dependencies and a phone has limited storage available.
I think hardware wise, that leads us to re-invent the laptop. Not a phone by any means. We need a large SSD for all the Flatpaks, lots of RAM to keep the software running in the background and a large battery stapled to it because none of it is energy-efficient enough.
It ain’t easy… I think we’re making (slow) progress, though.
A user facing permission system like what MacOS has is one of the things Linux is missing.
On linux an “app” is not as well defined due to the open structure of linux, which is the main problem in implementing this.
I want to be able to deny apps information and abilities, and also to give them false information.
Yes. I mean for most stuff we probably don’t need a complicated debate on definitions. I think we could do it for most desktop apps without further questions, like a browser, instant messenger, camera app, video editor, games, 3d printer slicer… They could all just do it and also not get permissions on the entire home directory right away. And we also have the groundworks for it. Linux has cgroups, thread isolation… permissions to use the webcam. There’s just the stuff in the middle missing and then we can’t build on any programming interfaces with the desktop apps.
Pine64 discontinued the PinePhone Pro unfortunately
When i look at postmarketos device wiki . librem 5 seems better, the chart lists the pinephone support for things like calls and SMS as partial. the librem 5 is the more expensive option but it seems like in practice at least some of the money for it went to good use. freedom isn’t free i guess.
What I wouldn’t give (or pay) for a 1. sleek, modern smartphone 2. running a pure Linux distro 3. that’s feature complete enough to daily drive
All of the current options available fall down in one of the three areas. Usually 2. and 3… mostly 3.
Doesn’t it run an Android kernel?
It does; but that wasn’t a blocker that OP mentioned (I read “pure” as Linux kernel with GNU coreutils).
Well Android doesn’t use a mainline Linux kernel and instead adds a bunch of hacks that break compatibility with some core Linux programs, so it’s not pure in that sense either
Just reading the reviews and it sounds like it’s got problems. GPS doesn’t work, mobile data is sketchy. That’s what I’m talking about. I’m fine to tinker and massage most of my devices into a working state, but not my phone. I can’t be messing around with terminal commands trying to get my gps working when I’m out on a trail for example. Can you imagine if there were an emergency and I first had to try and figure out why telephony was suddenly down before I could call emergency services? My phone is the 1 device that HAS to work flawlessly every time.
Which reviews had you read? Earlier reviews of the previous model aren’t going to be as accurate as a lot of updates have led to improvements. You’re not wrong that things aren’t fully out of the woods (as is inevitably the nature of things, at this point in time) but most people who drive it daily have said that things just generally work, usually, these days. Their Matrix/Telegram channel may offer a more accurate depiction of the current state of things.
Now, Android/iOS level performance may be what’s necessary to make the cut for you (that sounds sarcastic, for some reason, but I genuinely mean that as neutrally as possible) but I figured I’d mention since people will often mention the Liberux NEXX (a device which doesn’t even exist, yet) without even mentioning this one and it’s by far closer to an actual possibility of being daily driven out of everything else out there.
EDIT: for example, here’s a review from a few months ago: https://clehaxze.tw/gemlog/2025/07-20-flx1-actually-usable-linux-phone.gmi. Again, I don’t want to make it sound like bugs don’t crop up occasionally (I don’t we can expect otherwise this early in the game) but I do think this one’s potentially actually feasible at being a daily driver far closer to what you’re looking for comparatively to anything else out there, at the moment.
I just quickly browsed the reviews directly off the page you linked
The GPS one was from August, and skimming them again now, the impression I get is that it’s a great experience for a Linux phone but that it still requires a fair bit of technical know how.
Android/iOS level performance
That doesn’t sound sarcastic, and it pains me to say it but yes, for the core OS and features, I am unlikely to fully commit until it’s at that level of stability.
I don’t need wide app support, and I’m more than happy to tinker and run compatibility layers and whatever else for 3rd party stuff, but the core features of the phone have to be rock solid. That means GPS, WiFi, mobile data, SMS, MMS, Bluetooth, camera etc etc. need to be absolutely flawless. It’s a safety thing in my mind.
Not to mention when the rest of my hacky, cobbled-together, bad diy home infrastructure inevitably falls over, it’s usually phone I reach for to fix it. If that starts acting up too, I’m in deep paddle without a creek to stand in
ETAsk: do you own one? How do you find it? I’ve just read the review you posted and that is really promising
The GPS one was from August,
Ah; welp, I stand corrected.
That’s fair; most reviews I saw made it seem like it was one of the few options that actually could be daily-drivable (like, still probably infrequent crashes, slight small bugs, but all the crucial bits worked 95% of the time).
the core features of the phone have to be rock solid. That means GPS, WiFi, mobile data, SMS, MMS, Bluetooth, camera etc etc. need to be absolutely flawless. It’s a safety thing in my mind.
I mean, yeah; totally fair. I think I have just a smidge more flexibility. Like, I need battery (for obvious reasons) but I’m fine if I can get things working again with just a reboot. But I do have to go around far less than I used to (WfH, generally stay in most days). That said, my plan’s to keep my old phone around, just in case.
do you own one? How do you find it?
Not yet but I pre-ordered back in Sept. (maybe Aug.? A few months ago, certainly). Looks like current estimates are seemingly looking like some time in January.
Most of what I’ve read and seen have been promising but I’ve also just been ready to get rid of Android for, like, 5 years, now. I figure, even if things don’t work well enough for daily driving, I’m still supporting efforts to make it, finally, a reality but (naturally) I’m, also, really hoping.
I’ll let you know my impressions when it comes, if that’s something you’d want? Like I said, I often mention the device since it seems to go unnoticed in most Mobile Linux conversations but it’ll be good to have one so I can do more than just quote others’ experiences.
2 and 3 are the same thing. 1 isn’t something I care about too much.
Librem-5 is failing only at 3, right? Pine phone fails at 1 and 3 …
This is a sexy device. But can I ask you to make a repo on codeberg ? Github has been taking down repos that might be a threat to Big-Tech monopoly
Thanks for the suggestion, I mirrored the repo here: https://codeberg.org/muhammadmanwar/cheaphone
I may end up switching entirely to code berg, but I’ll see how bothersome it is to push changes to both before I fully migrate. Thanks again!
I’m not sure if you can do it the other way around, but you can set up a push mirror on codeberg so you only have to push to one forge.
i just want a linux phone with a slide out physical keyboard.
Yes. So say we all.
So say we all
This is the way.
I mean, so say we all.
Bring back the Nokia N900!
https://youtube.com/shorts/lDPZQpn4DeI
Something, possibly, to keep an eye on.
Fuck yes. Like a tiny little laptop or Gameboy Advance. I would replace my smartphone immediately.
deleted by creator
there’s a youtuber working on something similar, maybe yous can join forces: https://youtu.be/OgMdO0ckICg
Wow! That does seem really similar to what I’m doing. And they seem further along than I am. I’ll have to look into this project some more. Thank you!
I suggest adding a license . i recommend a copyleft license (there are copyleft licenses for hardware. for example the cern licenses).
I also suggest setting up a open collective. i suspect people might be more inclined to donate to a non profit then to for profit companies like purism and Pine64.
Why would you donate to pine64? They are selling things, just buy their stuff. Its like a donation but you get stuff out of it too!
When dealing with stuff like kickstarter campaigns. some people might not to risk the full amount (something like 600$), they might be interested in donating something like 10$ to help the project put out a product. then read up on reviews and decide if to go for it.
They already have products that don’t cost that much though. Do wish their website was better listed to compare specs of their SBCs though.
deleted by creator
Still, i try to act like an environmentalist and that means not buying stuff i don’t need. also a big part of that money will go to funding manufacturing costs and the development of new products for the same product line. unlike a campaign where a larger share of the money will go to developing a phone (and some of the money will go to give a return on investment to the owner, which is something i am fine with as long as there is no non-profit that can do the same work better).
Also for the CEO or board of directors it will be harder or even impossible to deduce that this signals a interest in a FOSS friendly smartphone.
Thanks for the suggestions! I’m not actually looking for any donations though. It probably sounds weird, but I don’t want to derive value from this, or even assign value to it, in the interest of keeping the information as freely accessible as possible. Not too get too ideological, seeking money often causes people to make a good idea bad, or to make a simple process inefficient, to make more money from it. I’m thankfully in a position where I can keep (slowly) working on this project in my free time, while still keeping my head above water.
That isn’t to say that no one else should make money from this idea. I just don’t want to personally.
I do like the idea of a copyleft license. I’ll have to look into it a bit more. Thanks again for your suggestions!
Thanks for the suggestions! I’m not actually looking for any donations though. It probably sounds weird, but I don’t want to derive value from this, or even assign value to it, in the interest of keeping the information as freely accessible as possible. Not too get too ideological, seeking money often causes people to make a good idea bad, or to make a simple process inefficient, to make more money from it. I’m thankfully in a position where I can keep (slowly) working on this project in my free time, while still keeping my head above water.
If you want to not get paid that is fine. but donating is the only way some people will be able to help make this happen. you could hire people using something like fiverr to do some of the boring stuff. money is just an efficient way to store and transfer economic resources. There is a significant difference often between a how a non profit allocates a economic resources vs a company that is owned by pension funds and mutual funds and is just trying to maximize a return on investment. Some of the best open source projects (e.g. blender signal thunderbird etc) hire full time workers.
I don’t necessarily disagree with you. And people do seem excited about the concept. I’m not even sure I’m far enough along to justify taking in donations though. So far I feel like all I’ve done is compile information that’s already available online into one document. I appreciate your perspective
Sounds reasonable. maybe take 3 months. spend about 30-50 hours working on this and see you can sustain the motivation to do this. then decide what to do next.
I think everyone has once thought of the idea of taking an sbc and a touchscreen and making a linux phone, cool to actually see one!
My thought (from before smartphones) was a computer you could mount on your belt or in a backpack, connected to a pocket sized screen that could unfold or unroll to become reasonable sized
At home you could plug it into your monitor, mouse, and keyboard
I think I imagined a touch UI
It would have run Linux, but it wasn’t intended to be a phone
deleted by creator
I’m not sure how to add more images to the post, sorry. But I uploaded images to a repo, in the pictures directory https://github.com/muhammadmanwar/cheaphone OR https://codeberg.org/muhammadmanwar/cheaphone
deleted by creator
Ah, the thicc phone.
What are the dimensions of the case?
50mmx130mmx82mm I measured it in freeCAD, so it might be a few millimeters off the real thing
deleted by creator
deleted by creator
Seems like the Pro is discontinued? https://www.howtogeek.com/pinephone-pro-is-discontinued/
deleted by creator
since the rpi5 cannot make phone calls and sms, i wonder if you or someone would design a attachment to add calls and sms (like to gpio pins for example)
Right now I’m using this “Gravity: CAT1 A7670G Global 4G IoT Communication Module”, connected through USB-C https://www.dfrobot.com/product-2802.html I am able to text reliably using a chat software, and connect to mobile data unreliably using ModemManager and mmcli. I haven’t been able to make a call yet, but I think this is due to a software issue, I’m still trying to get everything working reliably. Once I figure out how to do all of these things, I will add it to the guide in the repo https://codeberg.org/muhammadmanwar/cheaphone
i never knew that was a thing,may come handy.
When planning to go into mass production?
Why does it need to go on mass production? OP explained they want to get to a point where they share their design.
I keep repeating the same about Linux and other free software projects. The main goal is freedom, not market share.
OPs project seems to follow the same goals. And I find it awesome.
I meant the ability to order such a device. I just structured my opinion wrong. Because of sharing the device blueprints and software doesn’t mean that anyone will be able to create it by himself.
This is a really cool idea. Any work in the direction of more linux phone technology is always good. There are some linux phones out there already, and these devices have had some problems which is why there hasn’t been more adoption. If there is a way to do this with RISC and have decent battery life, that would be really exciting. Have you tried installing Phosh on it?
The newest, and most exciting, option right now is flx1s (https://furilabs.com/flx1s-update-2/).
One of the biggest problems is that, to my knowledge, there is no standard linux mobile App standard or, if there is, it’s not often used. There is a group working on this issue right now (https://www.fierce-network.com/wireless/mobile-linux-standards-group-formed).
For example, if I am using mobian or something similar and download thunar using sudo apt install thunar, then if I run thunar, it will run, but certain menus won’t display easily. In phosh, any sub-menu boxes will also pop up as a smaller pop-up box and to close it, you have to scroll through Apps to find the pop-up box and then close it. Generally I may be able to see the file structure on the left in Thunar but have a hard time seeing what’s on the right.
There are also things that can happen in which default panels (like the side panel) take up so much room that you can’t see what’s going on. For instance, if you try running gimp in phosh, you can barely see the image panel by default.
There are some Apps designated as mobile-friendly but even these sometimes don’t display correctly. Perhaps there should be a way to make it harder for Apps to be installed that don’t meet mobile standards and have weird menu glitches, such as making it harder to download Apps from repositories that are not mobile.
I’d really like to be able to run something like “flatpak-mobile install librewolf” and just get something that at least had a file with it to tell phosh how to display menus in the best way, if not a slightly altered librewolf.
Many people who used phosh would say “well just use waydroid” but the problem is that with play integrity api, many of those Apps won’t work.
In order for banking Apps and other Apps to run on linux and people to develop software, there really needs to be more adoption of mobile linux.
And yes, battery life was atrociously bad and completely unusable on the linux mobile devices I tested, which were a few years ago. It also got way too hot when just not doing anything, which was terrible. (In other words, if I took the device with me to Starbucks and got a coffee, it might get way too hot in my pocket; if I took it out and used the Internet for 20 minutes, the battery could die, and even if it didn’t, if I were waiting for a call during that time there was a good chance I wouldn’t get it. After getting back home, it would be totally dead and need a full recharge.)
Right now also, the main competitor to linux phones is Graphene OS with FOSS Apps and Graphene OS has better security features if someone is worried about their phone being stolen or seized. Data security is important to me and Graphene OS has a rate-limiting throttle to the password entry that even cellebrite can’t easily bypass and a function to auto-reboot. If the political situation in my country deteriorated even more, and arbitrary arrests started to happen more often, I would much rather be arrested with a Graphene phone than a typical linux mobile phone. Mobile linux for certain distros such as Mobian still offers robust encryption in before first unlock (bfu) if the password can withstand brute force attempts, but since there’s no hardware rate-limiter, the password has to be much more complex. Also, most people who use their phone frequently are not in BFU mode.
Graphene OS also requires a Pixel which does not have hardware switches and so a person must trust that there’s no exploit allowing certain components to be turned on or off which can be concerning when there is no way to definitively measure what the cellular modem is transmitting. Call me paranoid, but given what I know about how easy it is for someone smart to exploit computers, I actually don’t want a cell phone microphone to have power when I am talking about sensitive things or inputting passwords into computer systems and I do not want a camera that is built into the lcd part of the glass screen and can’t be easily covered because of the need to swipe up nor turned off without a switch, even if the cell phone has an incredible operating system that is very secure. Graphene, unlike most mobile linux distributions, is mostly very usable with no battery life issues, no weird display problems where Apps don’t display correctly and menus don’t work correctly, and no random reliability problems, mostly. I understand not wanting to rely on anything involving Android, however, given Google’s recent aggressive anti-privacy stances.
I’m excited about FuriPhone (https://furilabs.com/) and Purism’s Librem 5v2.
The thing that I believe would help Mobile Linux most right now is people having conferences and getting to know each other and discussing standards, specifically on user experience, linux mobile app standards, battery life, and reliability.
There are so many smart people in mobile linux and eventually it will get great but right now there are major problems with the user experience because of how Apps are displayed and battery life as well as things like reliability.
So any way to gather people to discuss the mobile linux user experience and to come up with standards to reduce these issues would help, or even to help list all the different problems so that they can just be enumerated and acknowledged and worked on would help.
Another way that would help is to have a mobile linux security group or conference to discuss things like standards for making mobile linux more secure from brute-force attacks if stolen or seized after being unlocked.
I’d really like to be able to run something like “flatpak-mobile install librewolf” and just get something that at least had a file with it to tell phosh how to display menus in the best way, if not a slightly altered librewolf.
This is a great idea
Really? How do I tell the right people?
All that is needed to achieve this is a package repository. A “mobile apps” repo for Flatpak. This isn’t that difficult.
