Sorry for the lack of updates, development has slowed down a bit as I've been back and forth on . both PCB and the case's design language. Here's a quick summary of what's going on:
PCB has been completely redesigned. Will feature a seperate PCB for the switches and a daughter board that contains the electronics.
Minor changes to the case have been made to accomodate tented bottom and new layer-status LEDs.
Prototype printed in my workshop and the case design is now frozen. What you see here is 90% likely what gets produced!
Battery has been chosen! 3000 mAh battery which should last well beyond a year per charge.
Order prototype PCBs (for real this time)
Compile and test firmware (for real this time)
Continue looking for a manufacturer for tented bottoms
If you're interested in the details:
The overall case design language has not changed. A few minor changes were made in an attempt to make it feel sleeker, easier to mount bottoms, and accommodate some of the new electronics. To this end, here's a list of what was changed:
Increased the chamfer around the top outer edge to 2mm, up from 1mm
Added a 1mm chamfer around the bottom edge so it's less sharp
Reduced gap between bezel and keycaps
Increased thickness of the case by ~1mm to accomodate battery and possible thermal expansion
Added 3 LED holes to the inner bezel
Added USB port hole (flush to the case)
Bottom mount screws changed from M3 to M2
Added corner relief to the plate switch cutouts so Kailh switches insert more smoothly
And to show what all that looks like in motion, here's the latest printed proto:
The LEDs will be in the bezel next to the top right most switch. I've always been a huge fan of the LED indicators on the Uniqey C70 and figured that with the large bezels, we could do something similar. I think here they're out of the way and don't stand out, whereas if I had put the LEDs up at the top of the case, they'd be front and center since they broke up that otherwise large, smooth surface.
The USB port hole isn't shown in the images above, but here's a quick render of where I decided it will go:
The bottom mounting screws were moved from M3 to M2 due to a few issues. First, with the acrylic bottom, countersinking for M3 would leave the acrylic very weak and crack prone. The issue there is that you end up having ugly exposed heads. Going down to M2 allows me to countersink and hide the screw heads better. Secondly, for the tented bottom, M3 takes up so much space that the corners of the tented foot are very very thin (since the M3 screw heads are almost the radius of the corners). By going down to M2, it makes the tenting feet much easier to fabricate and attach.
As for the tenting itself, the foot present here is a 5 degree foot. I have files for a 10 degree foot too. My plan is to offer these 3d printed with an SLA printer to produce a high quality bottoms to match the tops. I got quotes for machined bottoms made of polycarbonate (plastic over metal because of signal propagation) but the bottoms would have cost more than the case and plate combined. With SLA printed parts, there is no MOQ so for people interested in a high quality tenting kit, that's an option, or I can print them on my printer and while they'll be lower cost and quality, you can get a feel for what tenting feels like and order an SLA printed tenting kit for yourself in the future. I'm looking for an affordable supplier of SLA printing still, but the best quote I have so far would be $50 for the two tented feet.
And finally, for the low cost version, the overwhelming majority of you voted for 3d printed over acrylic. That's really really easy for me because the Bireme files were designed from the ground up to be easily 3d printed (no supports, reasonably large surface area to prevent lifting on unheated beds). The STLs for the files will be released prior to the GB so you can print them for yourself at the local library or school or if you have your own printer. If you like the layout, you can either grab a PCB alone or get a metal plate to help add some heft to your case. I can say right now though that even the 3d printed plate is very rigid and feels great to type on.
The PCB was completely redesigned, again. When I started to lay the traces for the right hand side PCB (only the left hand side was done at the time of update #1), I realized that the positioning of the wireless module was so tight that I had to move it on the other half, and flipping all the chips meant I had to completely rethink how things were laid out for the other side. I realized if I was gonna re-layout the PCB again from scratch, I figured I might as well put everything on a daughter board that I can layout once and works for both sides. It also solves all the problems with the long USB traces and having to run power from the USB port to the battery which were on opposite sides of the massive board due to space constraints. It wasn't elegant and I hated it.
The daughter board has been reviewed by my colleagues and I've gotten the OK to produce. I'm still mulling over some minor things (such as if the port should be recessed or not) before I pull the trigger, but it's definitely happening this week.
Again, if you want to give me some feedback about where to proceed from here, I'd really appreciate if you could fill out this short survey: https://forms.gle/t8sXabMtYtr5KXbP6
Thank you for your time everyone!