improvements on Route 2

This commit is contained in:
BubblesToTheLimit 2016-09-10 22:57:49 +02:00
parent 9370c43dfe
commit 9d87a0b134
3 changed files with 59 additions and 3641 deletions

View file

@ -161,70 +161,78 @@
That's all for now, this guide will improve over time! (TODO!)
** Route 2: Dactyl Hand-Wiring Guide
Step 1: Creating the rows
*** Step 1: Creating the rows
Using the copper wire you solder each row together, notice how the thumb-part gets its own
row. If you take a close look you can see how I did do a messy job soldering, I had to redo
some of the parts because they weren't soldered properly. Do this for both sides. Which ones of
the pins you chose for the row doesnt matter, but for it to look clean you should decide for
either one of them
Using the copper wire you solder each row together, notice how the thumb-part gets its own
row. If you take a close look you can see how I did do a messy job soldering, I had to redo
some of the parts because they wouldn't be stable enough. Do this for both sides. Which ones of
the pins you chose for the row doesnt matter, but for it to look clean you should decide for
either one of them
[[file:wiring_create_rows.jpg]]
[[file:wiring_create_rows.jpg]]
[[file:wiring-create-rows-both-sides.jpg]]
[[file:wiring-create-rows-both-sides.jpg]]
You might want to go ahead and test each one of the 70 switches with a multimeter.
You might want to go ahead and test each one of the 70 switches with a multimeter.
[[file:wiring-create-rows-testing.jpg]]
[[file:wiring-create-rows-testing.jpg]]
*** Step 2: Create the columns
The next step is to create the columns by soldering the diodes.
The next step is to create the columns by soldering the diodes. Notice which way these diodes
are directed here, the way these diodes are directed, we have to have a firmware that "drives
the rows". (TODO: insert diagram here) Luckily in our firmware "ergodox-firmware" we have the
option that adapts to that. However if you want to try and use the "tmk_keyboard" firmware
which seems to be also interesting you want to wire the diodes the other way. This way the
"columns are driving". As a side note, for some reason the tmk_keyboard firmware things of rows
as columns and the other way around, so if you look through the source code it will also see
"drive rows", even though that means "driving columns".
Notice which way these diodes are directed here. First of all, they are all wired with the
black part towards the switch, it is important that you follow that pattern for all of the
switches. The end where that color-mark is denotes the direction of the diode. The way these
diodes are directed, we have to have a firmware that "drives the rows", which means the
electricity can go from the column through the pressed switch through the row, but not the
other way around (TODO: insert diagram here to explain this further).
The tmk_keyboard firmware could also be used for this kind of wiring, but it would have to be
rewritten, maybe in a beautiful way, such that it can also "drive rows". However I didnt go
ahead and perfomed that rewriting, so I had to stick with the "ergodox-firmware". In retrospect
I'd wish I had wired the diodes the other way around.
Luckily in our firmware "ergodox-firmware" we have the option that adapts to that. However if
you want to try and use the "tmk_keyboard" firmware which seems to be also interesting you
want to wire the diodes the other way. This way the "columns are driving". As a side note, for
some reason the tmk_keyboard firmware things of rows as columns and the other way around, so
if you look through the source code it will also see "drive rows", even though that means
"driving columns".
The first picture doesnt show how the 6 thumb-keys are actually individually connected to the 6
main columns.
The tmk_keyboard firmware could also be used for this kind of wiring, but it would have to be
rewritten, maybe in a beautiful way, such that it can also "drive rows". However I didnt go
ahead and perfomed that rewriting, so I had to stick with the "ergodox-firmware". In retrospect
I'd wish I had wired the diodes the other way around.
[[file:wiring-create-columns-left.jpg]]
The first picture doesnt show how the 6 thumb-keys are actually individually connected to the 6
main columns.
This second picture shows how the 6 thumb-keys are actually connected to the 6 main columns.
[[file:wiring-create-columns-left.jpg]]
[[file:wiring-create-columns-right.jpg]]
This second picture shows how the 6 thumb-keys are actually connected to the 6 main columns.
TODO: Insert diagram of how the thumb-keys are connected to the main clumns.
[[file:wiring-create-columns-right.jpg]]
The next step is to wire the Teensy 2.0 and the MCP 23018 in place.
Everything necessary for that really is the following circuit diagram
TODO: Insert diagram of how the thumb-keys are connected to the main clumns.
[[file:circuit-diagram.svg]]
*** Put the Teensy and the MCP in place
This next step is to wire the Teensy 2.0 and the MCP 23018 in place.
Everything necessary for that really is the following circuit diagram
[[file:rows-and-columns-connected-to-chips.jpg]]
[[file:circuit-diagram.png]]
Things to note here:
- The columns for the MCP go from 0 to 6, the columns for the Teensy go from 7 to D (counting
in hexadecimal) which makes great sence, since the MCP is left to the Teensy and we read from
left to right.
- When debugging this and using the ergodox-firmware, both sides have to be wired completely
and actually connected, dont expect the keyboard to work before that. The reason being is how
the ergodox-firmware stops completely if either the Teensy or the MCP weren't initiated
properly.
- In the circuit-diagram the columns on the left side (MCP side) actually go from GPA5 to INTA
(TODO: Update the circuit diagram and get rid of this point)
- The LEDs are handy for debugging, definately go ahead and connect those. When the keyboard
starts successfully two of the LEDs light up shortly.
- From the Teensy over to the MCP go exactly 4 connections. The blue, the red and the two green
ones. This is done using the TTRS jacks and cable (TODO: Add picture here how the jacks are
wired)
- Yes, the connection from B4 to VCC doesnt seem to make any sense, but the B4 port actually
gets used directly in the ergodox-firmware, so just connect those connections that make no
sense on first sight and either dont question it or find out why this is necessary by going
through the firmware in great detail.
[[file:rows-and-columns-connected-to-chips.jpg]]
Things to note here:
- The columns for the MCP go from 0 to 6, the columns for the Teensy go from 7 to D (counting
in hexadecimal) which makes great sence, since the MCP is left to the Teensy and we read from
left to right.
- When debugging this and using the ergodox-firmware, both sides have to be wired completely
and actually connected, dont expect the keyboard to work before that. The reason being is how
the ergodox-firmware stops completely if either the Teensy or the MCP weren't initiated
properly.
- In the circuit-diagram the columns on the left side (MCP side) actually go from GPA5 to INTA
(TODO: Update the circuit diagram and get rid of this point)
- The LEDs are handy for debugging, definately go ahead and connect those. When the keyboard
starts successfully two of the LEDs light up shortly.
- From the Teensy over to the MCP go exactly 4 connections. The blue, the red and the two green
ones. This is done using the TTRS jacks and cable (TODO: Add picture here how the jacks are
wired)
- Yes, the connection from B4 to VCC doesnt seem to make any sense, but the B4 port actually
gets used directly in the ergodox-firmware, so just connect those connections that make no
sense on first sight and either dont question it or find out why this is necessary by going
through the firmware in great detail.

BIN
guide/circuit-diagram.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 140 KiB

File diff suppressed because it is too large Load diff

Before

Width:  |  Height:  |  Size: 158 KiB