Repairable Mouse

Hey everyone,
I’ve been toying with the idea of building a repairable mouse. Before I build an MVP (minimum viable product) I wanted to ask the community here for some ideas to add to what I already have. Currently, my ideas are swappable close to everything that makes sense so mouse switches, MCU, sensor, choosable cable/comms method, and being able to select the number of buttons. Some of these ideas may be omitted from the real product due to being too hard to manufacture, the product will have a similar philosophy to Framework. I likely won’t implement these features into the MVP, but a much more basic version of them, using this more for collecting suggestions


I’ve always replaced bad switches on my mice myself. That’s really the only thing that goes bad on them. And I think there are already some mice out there with easy replaceable switches.

To be honest, the amount of buttons, shape, and being wireless are all things I am particular about. Being more repairable will not influence me to choose a mouse that doesn’t fulfill all those specific requirements for me first.

To get some ideas, check out to see what they’ve done with their mouse and trackballs.


If you decide to do wireless, I have an opinion for you: AA batteries are a must! No mouse is truly wireless if it does not have swappable batteries that can be charged externally. Rechargeable AA batteries are the most consumer-friendly way to do this.


This sounds truly excellent. Something that is really lacking in the market. There is the mouse, of course. But it’s sounds like you’re aiming for something more modular with more options.

What I’d suggest & hope for is just a good base for building off of. With the models for the initial shell available so others can modify & print their preferred shape, size, button placements. You mention “being able to select the number of buttons” but what about enabling easy custom placement for extra buttons? FFC connectors with a few gpio + gnd on either the base MCU board or a daughter board. So if a user made a custom shell with a button board holder, they could place them in the exact location that works for them.

I’d suggest starting with something basic. At least, I think most would consider it basic?

  • Wired. For QMK compatibility, and for the ease of not having to deal with battery optimization & more limited component selections.
  • MCU with plenty of rom & ram, well-supported by QMK. The Ploopy mouse is great, but if you want more advanced functions, it now suffers greatly from having a limited MCU.
  • Tilt mouse wheel. Optical, mechanical encoder doesn’t matter as long as it has L/R tilt. I think the lack of tilt can be a dealbreaker to many.
  • Lots of buttons. Less benefit in reprogrammable buttons if you only have a couple extra. I think Ploopy suffers from this.

Opensource firmware, at least as an option please. QMK/Vial for wired, and for wireless BlueMicro, ZMK, nrf52 QMK fork, etc.


Thanks for so many points,

Right now, I’m planning on using an RP2040 as the main MCU, and yep, I do plan to use open-source firmware as much as I can, My plan is to offer two major variants, somewhat similar to FW, one being a prebuilt mouse with the option of swapping the shells but nothing major, then a DIY version where people can custom build everything except the main circuitry, and yes I do plan to offer some GPIO internally for adding buttons. Pretty much everything you said I agree with. Ploopy is good but doesn’t offer swappable parts.


IMO simply not hiding the screws behind the mouse’s gliding feet would be a big step towards easier repairability.

1 Like

You can take replacement teflon feet from aliexpress or ebay and use a hole punch right where the screws are. Never again do you have to remove the feet & risk messing them up in the process. Punch in from the teflon side, down towards the adhesive backing. That way the cut edges are pushed in.

I’ve considered not using screws at all for the exterior and using magnets and a small locking mechanism to take off the outer shell and maybe even using magnets to secure the gliding feet.


Decided on my sensor, PAW3395 from PixArt is currently the best model within their standard product line. As far as I know this sensor is used in a lot of high end mice.


Is that still under NDA? If so how will that play with the open source firmware?

I’ll try my best to keep it as open source as possible so that may require omitting parts of firmware/schematics that are for public release.


And I guess if one wants completely open, they could pop in a PMW3360 if wired. Or PAW3204 if wireless. Iirc the 3204 is open. The benefits of modularity.

Yep, they all use very similar interfaces so it should be easily swappable.

i am not sure if this is still in the works, but my corsair scimitar is starting to experience scroll wheel problems (again), prompting me to search for a more open source solution. if a thumb cluster is a possibility i am definitely interested. i am not sure how easily i’ll be able to source parts if my wheel continues to give me problems, so repairable options covered in buttons is exactly what ill be in the market for

I read that the corsair scimitar uses a 9mm high encoder. Looks like a TTC green core. Fixing a Broken Scroll Wheel – Corsair Scimitar Pro RGB — Will's Lab Aliexpress, ebay, amazon should all have them.

~edit~ Yeah, this says it’s a TTC green core Corsair Scimitar Pro RGB Review: Ultimate MOBA/MMO Gaming Experience - Gamer Necessary

1 Like