Author Topic: Boston, an Austin-derived compact 120%  (Read 6853 times)

0 Members and 1 Guest are viewing this topic.

Offline Pylon

  • Thread Starter
  • Posts: 852
Boston, an Austin-derived compact 120%
« on: Fri, 08 May 2020, 00:59:21 »


The layout for this was inspired by the Austin and also by the 7-row Thinkpad keyboard you find on Thinkpads of the T420 generation and older. The general idea was to make a narrower full-size that retained both the 2u numpad "0" and traditional 2x3 Ins/Del/Home/End/PgUp/PgDn nav block, by expanding vertically. This would allow for an easy transition for full-sized users like me (I've been using full-sized keyboards my whole life)

Much like the Austin, I also generally wanted to use keys that were included in base kits for group buy keyboards, so I wanted to stick to things like a 6.25u spacebar and avoid things like putting a Print Screen key in R2, for example.

This first led to something like this:


I then realized that for the same vertical room I could add a large number of programmable keys above the F-row, and there weren't a lot of 120% keyboards out there, especially not narrow ones, so I figured I might as well add a bunch of keys, resulting in the layout you see on top.

Since this keyboard is an Austin derivative, and because I grew up in Boston, Massachusetts, I decided to name the project "Boston".

Everything except for the case and plate is posted on this Github repo:
https://github.com/bluepylons/Boston

I intend to post the case and plate after a group buy, possibly after a couple of rounds.

General specifications:
  • 121-keys
  • Traditional 2x3 Ins/Del/Home/End/PgUp/PgDn nav block
  • Full-sized numpad with 2u "0" key
  • Uses keys found in GMK base kits (except for programmable keys)
  • USB-C
  • Single-color backlight support
  • Electrical schematic and design largely copied from the Austin
  • Probably will use QMK
  • Controller on a daughterboard that fits under the F5-F8 keys, to reduce prototyping costs (small <100x100mm PCBs are dirt cheap, while large keyboard-sized PCBs are easily $100 per revision) . I currently intend to use an STM32F072 just like the Austin, though I may design a backup controller around an AVR IC (likely AT90USB1287) as QMK appears to be much better supported on AVR.
  • All SMD parts are on the daughterboard, except for backlight LED current-limiting resistors, to keep the main keyboard PCB buildable with through-hole only (unfortunately due to space constraints SMD resistors are necessary if backlight LEDs are to be installed)
  • Case design is TBD - I'm designing the PCB to be tray mountable, but I intend on  mounting via the plate (most likely via top-mount). I most likely will design a couple cases

The keyboard PCB is pretty much done, and I'm working on the controller right now, with work on the case to start after that.  Most recent files are available on the Github repo above.


I might eventually run a GB if this works out well, though that's a ways away.

Many thanks to:
  • The Acheron project - the KiCAD library and the source files for the Austin have been invaluable for designing this board.
  • The designers of the Austin (Driftingbunnies,  PheonixStarr, and Gondolindrim)
  • Gondolindrim, for assistance and feedback with the PCB design, as for running the Acheron project
  • KiCAD, for being an awesome free open-source PCB design tool
  • Keyboard Layout Editor
« Last Edit: Tue, 07 July 2020, 03:32:04 by Pylon »

Offline Maledicted

  • Posts: 2164
  • Location: Wisconsin, United States
Re: Boston, an Austin-derived compact 120%
« Reply #1 on: Fri, 08 May 2020, 09:04:33 »
Cool. Great job. I know the battleship and battle cruiser guys will like this.  :thumb:

Offline funkmon

  • Posts: 451
Re: Boston, an Austin-derived compact 120%
« Reply #2 on: Fri, 08 May 2020, 10:46:38 »
Show Image


The layout for this was inspired by the Acheron Austin and also by the 7-row Thinkpad keyboard you find on Thinkpads of the T420 generation and older. The general idea was to make a narrower full-size that retained both the 2u numpad "0" and traditional 2x3 Ins/Del/Home/End/PgUp/PgDn nav block, by expanding vertically. This would allow for an easy transition for full-sized users like me (I've been using full-sized keyboards my whole life)

Much like the Austin, I also generally wanted to use keys that were included in base kits for group buy keyboards, so I wanted to stick to things like a 6.25u spacebar and avoid things like putting a Print Screen key in R2, for example.

This first led to something like this:
Show Image


I then realized that for the same vertical room I could add a large number of programmable keys above the F-row, and there weren't a lot of 120% keyboards out there, especially not narrow ones, so I figured I might as well add a bunch of keys, resulting in the layout you see on top.

Since this keyboard is an Austin derivative, and because I grew up in Boston, Massachusetts, I decided to name the project "Boston".

Everything is posted on this Github repo:
https://github.com/bluepylons/Boston

General specifications:
  • 121-keys
  • Traditional 2x3 Ins/Del/Home/End/PgUp/PgDn nav block
  • Full-sized numpad with 2u "0" key
  • Uses keys found in GMK base kits (except for programmable keys)
  • USB-C
  • Single-color backlight support
  • Electrical schematic and design largely copied from the Austin
  • Probably will use QMK
  • Controller on a daughterboard that fits under the F5-F8 keys, to reduce prototyping costs (small <100x100mm PCBs are dirt cheap, while large keyboard-sized PCBs are easily $100 per revision) . I currently intend to use an STM32F072 just like the Austin, though I may design a backup controller around an AVR IC (likely AT90USB1287) as QMK appears to be much better supported on AVR.
  • All SMD parts are on the daughterboard, except for backlight LED current-limiting resistors, to keep the main keyboard PCB buildable with through-hole only (unfortunately due to space constraints SMD resistors are necessary if backlight LEDs are to be installed)
  • Case design is TBD - I'm designing the PCB to be tray mountable, but I intend on  mounting via the plate (most likely via top-mount). I most likely will design a couple cases

The keyboard PCB is pretty much done, and I'm working on the controller right now, with work on the case to start after that.  Most recent files are available on the Github repo above.
Show Image


I might eventually run a GB if this works out well, though that's a ways away.

Many thanks to:
  • The Acheron project - the KiCAD library and the source files for the Austin have been invaluable for designing this board.
  • The designers of the Austin (Driftingbunnies and Gondolindrim)
  • Gondolindrim, for assistance and feedback with the PCB design, as for running the Acheron project
  • KiCAD, for being an awesome free open-source PCB design tool
  • Keyboard Layout Editor

In for the group buy cause I'm too dumb to make this. It looks awesome! A compact battleship!

Offline Tactile

  • Posts: 1432
  • Location: Portland, OR
Re: Boston, an Austin-derived compact 120%
« Reply #3 on: Fri, 08 May 2020, 11:16:00 »
Kinda reminds me of something...

241867-0
REΛLFORCE

Offline Rayndalf

  • Posts: 472
Re: Boston, an Austin-derived compact 120%
« Reply #4 on: Sun, 10 May 2020, 21:44:18 »
Kinda reminds me of something...

(Attachment Link)

Those are really cool, I'm guessing it was modeled after the G80-1800 (but not manufactured by Cherry), but they needed some more keys for their (medical IIRC) equipment. A shame the original keycaps are pad printed. The layout adds a lot keys with a pretty minimal increase in size.

Offline Pylon

  • Thread Starter
  • Posts: 852
Re: Boston, an Austin-derived compact 120%
« Reply #5 on: Fri, 15 May 2020, 03:29:29 »
Some various updates:

Finished a first pass at laying out the STM32F072 controller:

 242589-0


I also started working on a top-mount CNC aluminum case in Fusion 360. So far I only have the top half of it somewhat CADed, though I have a general idea of how to make it:

I also will probably design a sandwich and/or 3D-printed case, since CNC aluminum might end up costing more than I'm willing to pay.
242591-1
242593-2

Also I realized there's an awkward space above the Esc key when starting designing a case. After weighing different options (hole, cutout, triangular cutout, angled notch), I  decided to just throw a rotary encoder in there, as I had three unused I/O pins on the controller.

Offline Pylon

  • Thread Starter
  • Posts: 852
Re: Boston, an Austin-derived compact 120%
« Reply #6 on: Fri, 15 May 2020, 03:33:34 »
(and I just realized two halves of the ground plane on the controller aren't connected - remember to always run  DRC folks)

Offline Maledicted

  • Posts: 2164
  • Location: Wisconsin, United States
Re: Boston, an Austin-derived compact 120%
« Reply #7 on: Fri, 15 May 2020, 08:21:04 »
Nice. Looking good.

Offline Pylon

  • Thread Starter
  • Posts: 852
Re: Boston, an Austin-derived compact 120%
« Reply #8 on: Tue, 19 May 2020, 02:57:11 »
Top-mount CNC case is getting there. A couple more details to work out, especially on the plate. Spending a lot of time fighting Fusion 360 unfortunately.

242937-0
242939-1
242941-2
242943-3

I tried to make the case easily machinable on a 3-axis with minimal setups (most pieces should be doable with just 2 setups), and avoided tapers (other than 45-degree chambers) to avoid having to do 3D profiling on a 3-axis CNC (which adds a lot of machining time and expense). That led to a 3-piece case, with the bottom piece serving to set the angle and also as the brass weight.


« Last Edit: Tue, 19 May 2020, 02:59:36 by Pylon »

Offline Pylon

  • Thread Starter
  • Posts: 852
Re: Boston, an Austin-derived compact 120%
« Reply #9 on: Tue, 19 May 2020, 03:29:03 »
One more render from the back, because why not?


242945-0

Offline Maledicted

  • Posts: 2164
  • Location: Wisconsin, United States
Re: Boston, an Austin-derived compact 120%
« Reply #10 on: Tue, 19 May 2020, 10:04:27 »
It is shaping up well.  :thumb:

Offline Rico

  • Posts: 104
Re: Boston, an Austin-derived compact 120%
« Reply #11 on: Tue, 19 May 2020, 10:23:42 »
What a nice project !
Is it you first try ?

Offline Pylon

  • Thread Starter
  • Posts: 852
Re: Boston, an Austin-derived compact 120%
« Reply #12 on: Tue, 19 May 2020, 13:19:39 »
Yeah, this is my first attempt at a custom keeb. And thanks!

Offline Rico

  • Posts: 104
Re: Boston, an Austin-derived compact 120%
« Reply #13 on: Tue, 19 May 2020, 16:22:04 »
Congratulations, it is a very nice keeb :)

Offline Pylon

  • Thread Starter
  • Posts: 852
Re: Boston, an Austin-derived compact 120%
« Reply #14 on: Fri, 19 June 2020, 04:39:21 »
The prototype is physically together! Finally got my parts in. Case from SuNPE, plate from Sendcutsend, and PCBs from JLCPCB.

First parts that came in were the case parts from SuNPe (apologies for the dirty background). I did some manual deburring, as I explicitly asked them not to deburr (as I wanted to preserve the raw as-machined surface of the aluminum, and SuNPe frequently deburrs using wirebrushes and scour pads that results in a more brushed finish)

245223-0

Next parts that came in was the plate, from Sendcutsend  (in a package with some panels for a different project)

245225-1

I did the prototype plate in low-carbon steel to safe a bit of money, as I wasn't completely sure it was even going to fit (it fit, but there ended up being a couple of mistakes: two holes were too small for the screw heads I intended to use to fit through, and the cutout for the rotary encoder was too small to allow the rotary encoder to be installed on the PCB  after the plate/switches were installed). Steel was also one of the very few materials you can get in 1.5mm thickness here in the US.

I wanted an interesting finish on the steel, and steel can develop some interesting oxide colors when you heat it up, so I put it over a stove to try to get it to oxidize in interesting patterns. DO NOT DO THIS WITH A STOVE AS I DID - this ended up warping my plate due to uneven application of heat, initially by quite a bit. I managed to correct it mostly by applying heat to different sides, but it ended up still being slightly warped. The side you heat up will contract slightly as it cools, which causes warpage, as any welder will tell you.

245254-2
245227-3

The warpage is a bit more apparent in this shot (I clearcoated the plate also with rattlecan clear spray paint):

245229-4

The plate flattened out pretty well when I installed it in the upper case with screws. There was a spot in the middle (underneath around where the F5-F8 keys would be) that didn't have any nearby screw-down  points, so I tried to flatten the plate when I w as soldering the switches in, using a hose clamp against the case:

245233-5

In retrospect, if I was doing this again I would use an oven, though that would result in more uniform color, and getting interesting combinations of colors requires uneven application of heat, which will likely result in warpage. There might be some possible combination -  the differences in temperature to get different colors are quite small, and you could preheat using an oven to get most of the way there, and then get different colors using a flame, which would reduce warpage compared to heating it over just a flame.

Anyways, the rest of the build:

Turns out soldering 120 diodes in is a chore:

245235-6
245237-7
245239-8

Also the diode for the numpad Enter key hits the stabilizer, so I had to solder it to the underside.

On the controller front, I started with the buck controller circuitry that takes 5V down to 3.3V for the STM32 and LED backlight (I went with a buck instead of an LDO to provide more available current for the backlight, due to the 5V 500mA limit).
245241-9


« Last Edit: Fri, 19 June 2020, 05:04:49 by Pylon »

Offline Pylon

  • Thread Starter
  • Posts: 852
Re: Boston, an Austin-derived compact 120%
« Reply #15 on: Fri, 19 June 2020, 04:49:50 »
(continued) It was fortunate that I tested the buck converter on the controller daughterboard before soldering the other components on, as I noticed the buck was outputting 5V out instead of 3.3V (which likely would have damaged the STM32 microcontroller had I soldered that in). Turns out I mistakenly routed the Feedback pin on the buck controller to ground (WTF was I thinking?), which results in the controller increasing the output duty cycle (and therefore  output voltage) because it thinks the output voltage is too low.

245245-0

Fortunately this was fixable with some scraping of traces and soldering a wire in to connect the pin to the correct place.

245247-1

Unfortunately I didn't take great pictures of the rest of the build progress or of the finished controller daughterboard.

Other mistakes I discovered:

  • Cherry-style PCB screw-in stabilizers aren't installable after the plate is installed. I unfortunately only realized this after soldering in around 15 switches, and spend over an hour with a solder sucker getting all the switches out and removing the plate.
  • I accidentally forgot to order the rotary encoder in  my Mouser order
  • I made the cutout in the plate for the rotary encoder too small to install the rotary encoder on the PCB after the plate is in place. I'll probably have to dremel or otherwise enlarge the cutout.
  • I forgot to install a pair of M3 screws in the PCB before putting the plate on (these screws were necessary for holding the controller on securely), and the holes I made in the plate large enough for a screwdriver head to fit through was not large enough for the screws to fit on after the plate was installed. General lesson is to design your keyboard so that parts can be installed after the plate is on.
  • For female headers, you can't substitute 2 20-pin female headers for a single 40-pin female header. The edges of the headers are too wide for the two headers to fit right next to each other in the same footprint as a single 40-pin header. Ended up having to trim the headers with a flush cutter (fortunately there were some unused  pins in the middle of the header, so losing a couple of pins from trimming was ok.


I built it up with MX Clears, and threw on an OG Cherry WoB set I had with a GMK Neon RGBY mod set. Sorry for the terrible pics - I'll post some better pics of the completed board tomorrow - these shots were done in very little light .

245249-2
245251-3

Next up is getting the software for the keyboard working (QMK and the like). Since it's electrically pretty similar to the Austin I'll be starting with the Austin QMK firmware and making the appropriate changes.
« Last Edit: Fri, 19 June 2020, 13:40:16 by Pylon »

Offline Jojjelito

  • Posts: 12
  • Location: Sundbyberg
Re: Boston, an Austin-derived compact 120%
« Reply #16 on: Fri, 19 June 2020, 06:22:51 »
Pure porn! Should be made into a group buy. I’d be on this like white on rice!

Offline Capsy

  • Posts: 346
Re: Boston, an Austin-derived compact 120%
« Reply #17 on: Fri, 19 June 2020, 09:50:07 »
Amazing work pylon.

Offline Zustiur

  • Posts: 235
Re: Boston, an Austin-derived compact 120%
« Reply #18 on: Fri, 19 June 2020, 10:20:08 »
Fortunately this was fixable with some scraping of traces and soldering a wire in to connect the pin to the correct place.

(Attachment Link)

Unfortunately I didn't take great pictures of the rest of the build progress or of the finished controller daughterboard.
Ah the good-ol' yellow wire fix. I had to use that on my first PCB because I forgot to connect VCC!

Offline Pylon

  • Thread Starter
  • Posts: 852
Re: Boston, an Austin-derived compact 120%
« Reply #19 on: Fri, 19 June 2020, 18:51:04 »
Thanks everyone! I'm working out the logistics of doing a GB for this (I put up an IC thread a while back). Also posted some better photos there.