geekhack Projects > Making Stuff Together!
Muri ONE: Low Profile Wireless TKL
aabbcc:
--- Quote from: ice9js on Thu, 17 June 2021, 12:12:52 ---Another round of keycap protos :) As you can see the shape is a little bit different now as well with the scoop centered on the stem where previously it was offset to the top.
I decreased the radius on the surface by about 30-35% and it does feel quite satisfying now, interestingly especially so on modifiers. The stabilizer connectors also fit perfectly now.
Unfortunately this round is quite a bit more prone to breaking - I'm not sure if it's down to it not being completely cured or the resin itself, I'll experiment some more. But they did a great job as non-functional prototypes and I'm very satisfied with this new direction. That said, I'll probably do a few more runs using some 'extreme' settings just to make sure I actually have all bases covered before committing to one design.
(Attachment Link)
Other than that, I've soldered on the missing chips on one of the PCBs and will be going back to doing more software work for ZMK.
One question I have... is there any tricks to make reworking PCBs easier? I don't know what kind of solder they used there but it was close to impossible to reflow using hot air and even trying to remove it with a wick and an iron proved to be quite difficult as well (I use plenty of flux). Once I applied fresh solder to the pads, it was smooth sailing from there.
--- End quote ---
Caps look awesome!
As for reworking pcb. If you ordered a pcb with presoldered componentes they most likely use lead free solder which can be a bit trickier to desolder. Easiest way to desolder in this case is to add some flux and a small amount leaded solder i.e. Sn60/Pb40 or Sn63Pb37 on top of the soldered joint, this will make it much easier to then desolder or reflow.
Minahctrl:
following. really liking this so far can’t wait to see the outcome
ice9js:
It's been a while but I've been busy.
First things first, new brand name: Muri (pronounced [muri]).
Sounds good enough to me, and haven't seen it used yet for any keyboards or peripherals so that should solve that issue.
I implemented a ZMK driver for IS31FL3741 and I'm currently busy working on adding more advanced illumination support as well: splitting the rgb matrix into independent areas, per-key rgb etc.
The idea is that it should be possible to drive every single LED with one or two driver ICs saving GPIOs on the main MCU as well as maximizing customizability as nothing is ever hardwired for a specific purpose, rather configured in the software.
While that is still some time away, here's the mandatory christmas show:
video0.mov (1582.7 kB - downloaded 169 times.)
It's just a single underglow animation, although it's worth noting IS31FL3741 is now a viable ZMK alternative to ws2812b where you quickly run into power issues.
As for the rows not lighting up, I need to touch up a few pins with some more solder, eventually... not a priority as it's not a blocker for other work.
Other than that, I've finalized the design for the case. Not much has changed on the outside, but there was a whole 'inside' I had to redo as well. I thought I'd share a few more details.
The decision has been made and the wrist pad is here to stay. The rationale being:
- I know for a fact it improves comfort, both on and off the desk.
- It can serve a decorative purpose as well. Being a fairly easy part to manufacture, it's possible to experiment with different materials and finishes based on personal taste.
- Although the keyboard has a fairly large footprint, the space on the inside is very limited due to the super low profile (12mm for the entire case) and the undercut around the sides. Having the little bit of extra room in the XY plane does make it easier to make the whole thing more structurally sound and fit a respectably sized battery.
Since the new design doesn't have wings, I had to make a cavity and a holder for the slider for it to be able to detect touch 'through' metal. The sizing came out pretty convenient making it possible for it to double as a battery holder as well - the slider goes 'underneath' and the battery would be taped on top.
I've got some extra vertical space in there as well compared to the location on the old case, so I could potentially increase the battery capacity as well - currently at 1600mAh.
As the third and hopefully final revision of the PCB will feature fast charging, a larger battery is certainly an option.
I will be attempting another 3D print of the case some time this week - this time by myself and in resin. Due to it's size, I had to split the thing into smaller parts which I'll be then glueing together. It looks something like that:
I'll let you all know how that goes.
ice9js:
--- Quote from: aabbcc on Mon, 28 June 2021, 03:46:40 ---As for reworking pcb. If you ordered a pcb with presoldered componentes they most likely use lead free solder which can be a bit trickier to desolder. Easiest way to desolder in this case is to add some flux and a small amount leaded solder i.e. Sn60/Pb40 or Sn63Pb37 on top of the soldered joint, this will make it much easier to then desolder or reflow.
--- End quote ---
Thanks for the tip! While I didn't have much luck with abundant flux on its own, adding a little extra fresh solder to the pads before doing anything else worked like a charm.
ice9js:
The print came out so-so, with 3 parts not printing at all, other two coming out a bit crooked but nothing a bit of sandpaper can't fix.
Luckily, the bulk of the chassis came out near perfect so I could validate the fit and comfort of everything.
Not sure if I will be 3d-printing another case prototype - I feel like if anything, the next step will probably be CNC.
A bonus of printing the case in parts is you can get cool cross-section shots.
It seems I also measured the USB cutout correctly - they way it's designed the actual port sits behind 1.5mm of case rather than stick through it for a cleaner look. Kind of like what you get on a MacBook.
Navigation
[0] Message Index
[*] Previous page
Go to full version