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 register 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.
Pine64 discontinued the PinePhone Pro unfortunately
the PinePhone Pro looks the most promising one of the bunch
I’ll have to advocate for the Librem 5 over the Pinephone Pro for the following reasons:
- The Pinephone Pro has officially been discontinued as of August 2025 [source].
- The Librem 5 and Liberty phones are still in production [source].
- Librem 5 PCB board design files are also available - not just schematics [source].
- Purism is already working on a Librem 5 version 2.
- Purism is pushing toward FSF RYF certification for the Librem 5 and future models.
I own a Librem5, and let me tell ya, it’s not a daily phone, hardware is just way too slow. Even with sxmo it lags a lot, opening a browser is a whole ordeal for it. Meanwhile when I tried my friend’s PinePhone Pro, it felt a lot better. Oh, and for context, I’m currently semi-dailying a OnePlus 6 with NixOS.
I’ve daily-driven my Librem 5 since March 2023. I will certainly not state that this device will meet everyone’s needs or expectations and would consider myself a patient prosumer, but comfortable daily use is possible and is proving easier in testing the next major PureOS release (crimson).
I guess we have vastly different expectations from our phones, then. At a minimum, I need to:
- Have reliable, snappy maps with precise GPS (for trekking)
- Be able to interact with my bank on the go, at least via a web app
- Be able to chat with people via Matrix
- Get transit routing via a web app
And in my experience, Librem5 just doesn’t have enough processing power and RAM to do any of those quickly and reliably. It was not comfortable at all, e.g. the browser kept filling up RAM and locking up the device with constant swapping, and finally OOMing. GPS took 5-10 minutes to get a lock, even with AGPS, and after that wouldn’t reliably keep it. Both Nheko and NeoChat were slow and laggy. It also died after 4-5 hours of suspend with a modem on, unacceptable for a reliable daily.
OnePlus6 is a rocketship in comparison, and performs all those tasks with ease. The battery also lasts for an entire day with conservative suspend settings (but with the modem on), and for a couple days in airplane mode (e.g. while hiking in the mountains).
Purism is already working on a Librem 5 version 2.
Do you have a source? i could not find anything.
Yes, I am a Purism staff member. Here are a couple Purism forum search results of things I have posed related to the Librem 5 version 2:
Very Interesting. any ETA? will it have faster hardware like a faster CPU? will there be a fundraising campaign like kickstarter?
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.
Considering their recent hardware reveals, I want a valve steamphone with a fully open system and modularity a la fair phone (or like their new VR headset) One model every 4/5 years would be perfectly ok for me.
A valve / framework / fairphone teamup would be a dream. It’ll never happen but I’d pay unreasonable amounts of money to see it
deleted by creator
omfg yes!
that’s feature complete enough
So what features do you really, really need?
Working hardware drivers. Stable phone and SMs applications.
Have yet to see either on any Linux Distro for Fairphone
I too need my sadomaso application, almost as often as online banking applications
Having already gone to e/OS and degoogled, avoiding apps on the play store - I’ve just been using the webapps via Fennec for banking, and its been fine. No notifications… But these days most of my important bank notifications can be emails.
Like all hardware supported out of the box such as Wi-Fi, Bluetooth, NFC.
Well, when a product manager tells me “everything is a priority”, to me it translates like “nothing is a priority”, because there is not something more important.
I think being maximalist here is a losing strategy.
No, the hardware needs to be completely functional. I’m not asking for a native Signal app or banking apps here, but if I purchase a phone with wifi Bluetoothand nfc, I expect it those things to function. That’s not maximalist, that’s minimum viable product.
Librem-5 is failing only at 3, right? Pine phone fails at 1 and 3 …
2 and 3 are the same thing. 1 isn’t something I care about too much.
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 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.
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.
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.
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.
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!
can’t see from the pic (more of them appreciated), but it looks like you can desolder the ports to save on space.
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
i love this kind of thing. yeah you can definitely desolder the ports and try to squeeze it closer to the screen, you can rewire them with smaller connectors if you need to. round the printed chassis down and you got something much more ergonomic.
others mentioned you can run postmarketos, but i’m sure you can also get android running too. with the low resolution it is at least running smooooth.
also aliexpress is your friend. a lot of stuff is already designed and made and can be a stepping stone into making it work better as an actual phone, and as “inspiration” if you really want to eventually design the circuits involved into a single board. things like the charging and control circuit for that battery, for starters. or better screens/batteries.
i wish you best of luck!
you could try one of these guys and a cm5 saves you some space https://www.pi-shop.ch/nano-base-board-a-for-raspberry-pi-compute-module-5
This is a very helpful suggestion, thank you! I have been having some issues figuring out spacing, the battery sticks out like a sore thumb right now, so if this can save me some space I may end up moving in this direction for further prototyping. Thanks!
OMG, what?!? Holy fuck, this is amazing! This is incredible! Way to go!
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
Happy to see this out there, love the idea. Watching the repo may mess around with this at some point when I have more time.
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.
This seems like a very ambitious project and a great learning experience for someone working on their own. For a similarly ambitious project, check out the Liberux NEXX. The project didn’t reach its crowdfunding goal, but they did make some progress before rolling up the rugs.
I don’t know if they’re looking for contributors or if you’re in a place to contribute, but most of the project is open source. You could probably get in touch with them and ask for any advice, successes and failures, and even if they have parts (such as their dev-board) that they can give you access to.
If you’re really going to do this you need a RISC-V processor SOC. If you look around online there’s a few places where you can obtain these.
For the my layperson friend reading over my shoulder… erm, why?
SOC has a small form factor which can fit neatly in a phone sized device (this us why all phones have them) and RISC-V is a completely open source processor instruction set that can be customized to whatever function you wish to implement.
Are the processors good? Like, does performance compare to the alternatives? (I’m assuming these are alternatives to like an ARM based SOC?)
If you are using a Pixel with a Tensor processor then you are using a RISC-V SOC.
Oh cool. Thank you for the info. I hear people talking about RISC-V a lot, but was nearing the parks and rec meme of “I don’t know and am too afraid to ask”, lol.
This is Lemmy not reddit. Unless it’s a political thing on .ml feel free to ask and most people will be happy to answer your questions, especially if it has to do with Linux or FOSS.
Not trying to dump on your efforts or anything here, but you’d be better off first defining a scope of what you’re trying to do, and then work off an existing hardware or software platform.
You can get phone dev kits for cheaper than $200 if you just want to build something that works without Android, but if you intend to take that further and design some of the software experience, you’d be better off just working or contributing to something that already exists.
A single person can’t even begin to touch on the fundamentals of what it takes to run a phone experience in that that we currently understand and use them. Touch UX, software<>hardware integration, peripherals like cameras…it’s A LOT. Doing it well as a single developer is just not going to happen.
If your goal is simply to not have to buy another shit Google-infested phone, you can get a cheap that runs other things right now.
From the sounds of it, it’s just a hobby project for fun for OP. I don’t think there’s anything wrong with doing something just for the sake of it.
Plus, if OP’s project catches steam, it provides more competition and cult followings which adds to the flair of hobby phones and its attraction thus helping to pave the way for more adoption and development.
This is definitely the angle I’m trying to work. The more people who know how to build these sorts of devices, the more software and hardware will be supported, and standard custom software/hardware pairings can be documented. But we won’t know what works till we try it
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.
