This is my 6th prototype for a mouse shell to replace the original Lofree mouse shell. I am a sucker for good design, and while this mouse had solid reviews the most glaring was that it was too small for most people. I agreed, but I also chose this mouse specifically because I could swap components. Unfortunately not many people have messed with making a replacement shell, and after 8+ hours of CAD and printing I can see why. However after some testing I managed to settle on this prototype, it's not perfect but it works for me better than the original shell. For full print details see below:

Print specifications:
-Machine: Print FDM or resin (FDM will require a little sanding and filler, see photos) with a standard 0.4mm nozzle
-Slicer: Printed in 0.15mm layer height, heavy infill for strength / 3 wall layers, I used Elegoo Rapid PLA+ and it's been surviving
-Orientation: You can print either upside down or right side up, either way I recommend dialing in your supports beforehand because it will require them. In my prototype photos I printed it on it's back which gave it a horrible texture in the palm, which I sanded and hit with putty (it's yellow because I haven't gotten around to painting it). With proper supports and a little sanding I think either orientation can work.

Notes:

• It fits fairly well into the existing connection structure, unfortunately one of my earlier tests cracked the connector on the mouse itself so my version wiggles, if you have any issues you can leave a comment because I will probably come back to this once my current prototype breaks down
• It would be worth trying to sand and smooth this as much as possible prior to use, the texture of prints tends to hold dust and dirt really well so minimizing that will help the appearance.
• This version I am calling a Large, in my hands it fits fairly well but I know the fitment could be improved. With a more accurate base model of the mouse I could also create different variations in size, but that will have to wait until I purchase a 3D scanner because my current Fusion files are a timeline nightmare held together with lost projections and loose approximations. I don't exactly know when I am getting my scanner so I am uploading this version until then, at the same time I will also post whatever I come up with for locating the support legs and the connector as a f3d and STEP so others can make remixes that match your style / hand size.

Feel free to leave any comments, on visual design ideas, size of the shell, hand feel. I am far from an ergonomics engineer so I'm basing this entirely off my use. As far as general mouse review if you're interested, the DPI controls are solid, the battery meter on the mouse is cool, it's got plenty of connection methods (and can be used while charging), and is pretty accurate for gaming / CAD. Battery life is probably closer to 1 week rather than 4 but that's more than fine for me, and with the larger shell I don't have to have a death grip on it to keep it from slipping out of my hand.

I encountered two main challenges with designing this mouse shell, the connectors had no square points to measure off from and the mouse shell was a bizarre organic shape that changed gradually.

My first step was to create a connector that fit the points on the back, which I took as many measurements with my calipers as I could and then made assumptions on what the actual values could be. I also used a drawing square to create datum planes I could measure off of and then copied that into Fusion 360. After I got some measurements (and used a chopstick to check the hole diameters) I created a resin test piece (Fig 1) to check my fitment. I also printed an FDM one and was happily surprised that it could print the connector crossbeam.

After solving this problem, the much more glaring one was the organic nature of the mouse shape. I noticed that the original shell used stand offs to touch the mouse at specific points, so I decided to copy that idea because it would be much easier to adjust the the lengths of the ribs than it would be to change the entire shape of the mouse. I tried guessing which, surprise surprise, was quite inaccurate, so I switched gears. Without a 3D scanner ( a little out of my current budget) or a Coordinate Measuring Machine (wayyyy out of my budget) I knew I would have to MacGyver this with nothing but cheap Chinese calipers and my 3D printer. I theorized a multi-caliper fully printed CMM (stay tuned for it I ever finish it) but for this case I elected the quick and dirty route as a proof of concept (Fig 2).

With the improvised CMM, I was able to record the depth at each point, and copy all of these points to Fusion. Then by drawing a line to connect them all, I had a (still rough but better) approximation of the shell. This got me close enough that I was able to print some more targeted tests, mainly just curved lines that I could print in minutes and then hold up to the center line of the mouse to check for fitment (Fig 3) . After that I started printing more prototypes, starting with just the top of the mouse and gradually expanding it to include the sides. After I was able to get a shape that fit, then I started tuning for the fitment in my hand. In Fusion I used the 3D spline tools to generate the surfaces, however this can get messy fast. This is where in hindsight I should have used the component linking tools in Fusion. If you make a dedicated file just for approximating the shape of the object you're trying to match, you can drag and drop that file into a new one to keep their timelines completely separate. This also means that if you make a prototype that doesn't work, you can always create a new file and drop in the base shape again to apply what you learned from the previous prototype without messing up your timeline (remember: don't fillet your grandpa).

Ideally A 3D scanner would be the right tool for the job, but the one I am looking at is the Moose at about $700 last time I checked, so if you don't have a spare grand to throw around creating your own jig / CMM isn't the worst option to get close. Then once you have an approximation that's close you can fine tune it by printing a few test pieces and then measuring off them. In Fig 3 I used a sharpie to approximate where I needed more thumb room, and then measured off the printed model to get dimensions. This probably isn't the best way of doing it but it did work surprisingly well so another tool in the toolbox.