[Updated] Wireless GH60

[Strelok]

Progress:[████1░░░░2░░░░3░░░░4]
1:Conceptualize
2:Prototype
3:Revise
4:Produce (whatever that ends up meaning)

What we're working on here is a collaborative effort to make the GH60 wireless for users that want it. A wireless 60% board would be small, portable, be able to connect to phones and tablets, and most importantly, there wouldn't be a wire running across your desk ruining the Feng Shui.

Originally I was thinking it would be pretty easy to just stick an Arduino and a battery in the case and everything would automagically work, but agodinhost has actually figured out how to do it. komar007 has made some revisions to the GH60 to provide for expansion boards like this one, but we still need to know what people want.

If you're interested in the project, please give your input on what you would like to see

These are the options various options available to us according to agodinhost:

1) GH60 current version + usb charger + 5v booster + bluetooth:

* 5v, running @16Mhz
+ easy to build
- requires a minor GH60 PCB review
- some wiring required
- need someway to fix each module inside the case
- not modular at all

2) GH60 current version + usb charger + bluetooth:

* 3.3v, running @8Mhz (the firmware need to be tested at this speed)
+ uses fewer components
+ easier option to build
+ cheaper option to build
- requires a minor GH60 PCB review
- some wiring required
- need someway to fix each module inside the case
- not modular at all

3) Shield for 1 OR 2

+ no wiring required
+ shield fixed by connectors or sockets
+ so so modular
- not so easy to build, requires a new PCB for the shield itself
- requires a minor GH60 PCB review
- thicker (3 stacks)

4) GH60 v2 + usb charger + 5v booster + Teensy 2.0 + bluetooth AND
5) GH60 v2 + usb charger + Teensy 2.0 + bluetooth:


+ easy to build
+ cleaner matrix design
+ modular
- requires a major GH60 PCB review
- some wiring required
- need someway to fix the modules inside the case

6) GH60 v2 + Shield - usb charger + 5v booster + Teensy 2.0 + bluetooth AND
7) GH60 v2 + Shield - usb charger + Teensy 2.0 + bluetooth:


+ cleaner matrix design
+ no wiring required
+ shield fixed by connectors or sockets
+ modular
- requires a major GH60 PCB review
- not so easy to build, requires a new PCB for the shield itself
- thicker (3 stacks)

8 ) GH60 v2 + Complete custom controller PCB.
+ cleaner matrix design
+ cleaner controller design
+ no wiring required
+ controller fixed by connectors or sockets
+ smaller
+ modular
- requires a major GH60 PCB review
- hard to build, requires a new PCB for the controller itself
- slower option (more items to design)
- more expensive option
- more error prone

[WhiteFireDragon]

Controller, firmware, hardware that transmits info, integrating all this on the PCB, and a receiver on the other end. Definitely won't happen on this first round of GH60.

[Strelok]

Controller, firmware, hardware that transmits info, integrating all this on the PCB, and a receiver on the other end. Definitely won't happen on this first round of GH60.

I assumed that, I was thinking more along the lines of Arduino Micro + Buetooth Module + batteries.

[mkawa]

the major problem is powering the board via batteries. a bluetooth tx/rx chip is possible to add via the hackerpads, but it's not clear that you could hook a battery up to the VCC hackerpad and not fry something

[mkawa]

regardless, this would probably requiring designing a daughterboard and a custom casing to accomodate the extra board

[TDub]

Not entirely in context since the GH60 already has a controller, but I saw this on kickstarter and thought it might be interesting for building a wireless keyboard project:

http://www.kickstarter.com/projects/kenburns/tinyduino-the-tiny-arduino-compatible-platform-w-s?ref=category

It has an optional expansion to work with batteries, which might eliminate the battery complexity that everyone seems to dread.

[Strelok]

I've been doing some research and apparently the real issue is finding something both small and capable of being a USB host, rather than a USB device.

[hasu]

You need radio transparent case, Lipo battery, charger chip, and bluetooth module which supports UART or SPI. You'll be able to connect module to GH60 spare pins. Then rest of work is only software job.

Or I guess you can just take apart this and push its guts into GH60 case if it has enough room.
http://handheldsci.com/kb

[mashby]

When I got back into mechanical keyboards, I really thought I'd miss having a wireless keyboard. Surprisingly, it hasn't been that big of a deal -- which is probably due to the cable I got from Mimic. 

That being said, it would be nice to have the option to be wireless, so I'm curious where this going. Very interested.

[Aleksander]

I was planning on trying to do a wireless conversion with THIS
Is connects as bluetooth HID and accepts serial HID stream from the controller in the keyboard.. so some coding on the keyboard-end, the thing I linked to and a LiPo battery... at least that is my theory...

[Strelok]

I was planning on trying to do a wireless conversion with THIS
Is connects as bluetooth HID and accepts serial HID stream from the controller in the keyboard.. so some coding on the keyboard-end, the thing I linked to and a LiPo battery... at least that is my theory...

I'm looking at a similar Bluetooth HID SMD, but Arduino branded, connected to an Arduino Micro. I've used microcontrollers before but not since highschool. It's looking like the complicated part is going to be connecting the keyboard to the controller as a USB device, since most controllers aren't naively USB Hosts. Some people I'm talking to have recommended removing the controller from the keyboard and taking switch input directly so remove a layer abstraction, which would potentially be simpler and more power efficient. I may also be able to reverse engineer Komar's firmware for the switch input, with permission of course.

[mkawa]

listen to hasu. you'll have much easier access to the SPI pins than you will to the USB pins. the trickiest thing will be switching between charging and voltage supply modes on the lipo's supporting circuit without frying anything (assuming you can supply at all from the broken out vcc and gnd pins that komar supplies)

[Strelok]

You need radio transparent case, Lipo battery, charger chip, and bluetooth module which supports UART or SPI. You'll be able to connect module to GH60 spare pins. Then rest of work is only software job.

Or I guess you can just take apart this and push its guts into GH60 case if it has enough room.
http://handheldsci.com/kb

So what you're saying is that I can, take the raw SPI output from keypresses to the Arduino, and write firmware send that via Bluetooth to the computer, right?

[mkawa]

neg.

atmega <-> SPI <-> bluetooth PHY

when powered but not plugged into usb, the teensy firmware writes to the bluetooth phy. otherwise, it writes to usb. you can also pull other cool tricks like clocking the avr down on battery power, etc

[hasu]

Sparkfun module Aleksander linked to looks like nice and it can be controlled via UART from ATMega on GH60.

I also tried bluetooth mod  on my HHKB with bluegiga WT12 connected via UART.
But I'm not using this anymore, there are many problems to solve.
Power management(saving/charging) and Bluetooth pairing are difficult part for me.

Just FYI. Don't use this for your design because this is just experimental attempt without any document.
circuit:
https://raw.github.com/tmk/tmk_keyboard/master/keyboard/hhkb/doc/Bluetooth_img/BT_circuit.jpg
source code:
https://github.com/tmk/tmk_keyboard/tree/master/keyboard/hhkb
result:

[mkawa]

the difficulty with the sparkfun module is the one i mentioned earlier (and you mention again ). the pinout and circuit is inappropriate for powering/charging. sparkfun has other modules that are just mounted SMD bluetooth PHYs that are better suited for the purpose, but have to be combined with a "smart" battery board of some kind (which i don't see on their site, unfortunately )

[mashby]

This makes me wish I hadn't thrown away my Apple BT keyboard after I broke it. Would have been interesting to source out just the BT and battery parts.

Don't know if it'd have been of any use and I would have had to send it to one of you to actually do anything with it since I'm a newb at this stuff.

Our course now that I think about it, I don't know that this would be replicable. Step 1: buy a $70 Apple keyboard! 

[mkawa]

yah, the raw parts are cheaper than that

so is making a small run of the pcbs required. the trick is just picking the battery circuit components, as before. most bluetooth phys i've seen have no trouble with either uart or spi.

[agodinhost]

I saw here that when using bluetooth it would limit the keyboard to a 6KRO (assuming normal hid, no special drivers at the OS side).

It's pretty easy to connect this bluetooth module to teensy (assuming that we still have free pins). It uses normal tx/rx to communicate with the atmega and this hidapi library would help to avoid headaches.

And regarding the lipo charger sparkfun sells this open source module ...

It's a pretty cool project but I think that it would require more time updating the firmware than the hardware (I'm reeeeeeeeeeeealy slow writing C/C++ code).

PS: this bluetooth module that I linked above seems to be the cheapest one out there, BC whatever - it's programmable too but forget about it, the compiler price is absurd.

[mkawa]

that bluetooth board looks like it only has rs-232. i don't think the atmega komar used has native rs-232, only spi and i2s

[mkawa]

the sparkfun lipo charger module might be interesting, but i suspect we'll either need to buy them and pull the usb plugs off or clone the board design to properly subsid to our board and on top of that break the 5v signal from the usb on komar's board to route through the socket the lipo board plugs into. overall, non-trivial

[agodinhost]

that bluetooth board looks like it only has rs-232. i don't think the atmega komar used has native rs-232, only spi and i2s

Native rs-232? It uses normal tx-rx, we will need a logic level converter (to convert to and from 5 and 3.3v) but I did already saw a few guys using the module without it.
The Leonardo board uses the same atmega32u4 chip used on Teensy 2.0.

This project uses this very same module.

[agodinhost]

the sparkfun lipo charger module might be interesting, but i suspect we'll either need to buy them and pull the usb plugs off or clone the board design to properly subsid to our board and on top of that break the 5v signal from the usb on komar's board to route through the socket the lipo board plugs into. overall, non-trivial

Agree, we will have to have the module - otherwise we will have to make our own arduino+charger module (hmmm - I like this idea too). This module is simple - no big deal on it (take a look at the schematic) it uses only 6 components.

Sorry, I forgot that the teensy operates at 5v (not at 3.3v) - so it will require a 5v booster too.

Something like this:
1) Remove teensy usb connector
2) PC USB goes to the charger USB
3) The lipo goes to the charger input
4) The charger output is connected to the 5v booster input
5) The booster output goes to the teensy's 5v and gnd
6) The charger D+ and D- goes to teensy D+ and D-.

There is a second module with the charger AND the booster in only one board, here.

I don't have none of this modules at here - I do have only the charger ic and the 5v booster (I'm a poor bastard - always looking for cheap alternatives) but I wanna give a try, just don't push me, I'm slow!

I did used another simpler charger IC before (maxim) but it's out of stock and seems it will not be for sale anymore.

I just found the same 5v booster on ebay, less than 2 bucks!!!
>>UPDATED<< This 5v booster is ready to be used (the one that I've linked above will require to remove the usb connector)
The USB charger on ebay, US$ 1.55

[mkawa]

that bluetooth board looks like it only has rs-232. i don't think the atmega komar used has native rs-232, only spi and i2s

Native rs-232? It uses normal tx-rx, we will need a logic level converter (to convert to and from 5 and 3.3v) but I did already saw a few guys using the module without it.
The Leonardo board uses the same atmega32u4 chip used on Teensy 2.0.

This project uses this very same module.

right, the board uses an old-school serial interface. for interchip communication the teensy avr supports SPI and i2c _natively_. it can speak any arbitrary interface though with appropriate programming of the gpio pins. the code you linked to does just that. the only issue then is that komar has to break out the gpio pins on the board.

anyway, because of the power issue, komar has to build support into the pcb, so if people want this, NOW is the time to suggest a nice modular schematic that will allow the gh60 board to support power and wireless daughterboards. as it is, the wireless might work with appropriate firmware, but the power is probably not going to...

eta: also remember that if major changes are required to the pcb, we should be able to produce or obtain these optional daughterboards in bulk at reasonable prices. if all you want is to build a one-off, i can definitely see a way to hack up even the beta pcb to support this stuff. it will be ugly, but for a one-off, ugly should be a-ok..

[Strelok]

that bluetooth board looks like it only has rs-232. i don't think the atmega komar used has native rs-232, only spi and i2s

Native rs-232? It uses normal tx-rx, we will need a logic level converter (to convert to and from 5 and 3.3v) but I did already saw a few guys using the module without it.
The Leonardo board uses the same atmega32u4 chip used on Teensy 2.0.

This project uses this very same module.

right, the board uses an old-school serial interface. for interchip communication the teensy avr supports SPI and i2c _natively_. it can speak any arbitrary interface though with appropriate programming of the gpio pins. the code you linked to does just that. the only issue then is that komar has to break out the gpio pins on the board.

anyway, because of the power issue, komar has to build support into the pcb, so if people want this, NOW is the time to suggest a nice modular schematic that will allow the gh60 board to support power and wireless daughterboards. as it is, the wireless might work with appropriate firmware, but the power is probably not going to...

eta: also remember that if major changes are required to the pcb, we should be able to produce or obtain these optional daughterboards in bulk at reasonable prices. if all you want is to build a one-off, i can definitely see a way to hack up even the beta pcb to support this stuff. it will be ugly, but for a one-off, ugly should be a-ok..

I want to do this but all the electronics lingo in here is a little above my head. I would totally put in for a GB.

[agodinhost]

... the only issue then is that komar has to break out the gpio pins on the board.

Yup, it will need pins 0 and 1.
>> UPDATED <<
Yup, it will need pins 7 and 8 when using teensy 2.0.

anyway, because of the power issue, komar has to build support into the pcb, so if people want this, NOW is the time to suggest a nice modular schematic that will allow the gh60 board to support power and wireless daughterboards. as it is, the wireless might work with appropriate firmware, but the power is probably not going to...

+1 modular

eta: also remember that if major changes are required to the pcb, we should be able to produce or obtain these optional daughterboards in bulk at reasonable prices. if all you want is to build a one-off, i can definitely see a way to hack up even the beta pcb to support this stuff. it will be ugly, but for a one-off, ugly should be a-ok..

IMHO a modular approach would help to simplify future modifications and improvements.

[mkawa]

i think the challenge is power.. it's not insurmountable, but you have to route the 5v from the usb to the expansion socket before going to the controller circuit (or anything else...) and then provide an idiot-proof (because i'm an idiot and will inevitably screw this up) way to jumper the power when the expansion socket is not filled

note that you have to do this all while fitting into the case designs that exist and are being worked on (!!)

[agodinhost]

I want to do this but all the electronics lingo in here is a little above my head. I would totally put in for a GB.

It's not hard as it seems - but it would require some wiring and re-rerouting in the GH60 board, the atmega32u4 pins 20 and 21 will have to be wired to bluetooth tx rx (this is the hardest part, IMO).
The Charger and booster are quite easy (assuming the linked modules).
Give me some time then I'll try to help and post some pictures (I've already warned - I'm slow)

[mkawa]

can you design a single board that can handle these things based on those two boards and that routes the external pins to the same form factor as the gh60 hackerpads? i'm going to try to run another set of prototypes of the gh60 board, and it would be just about right, if the gh60 is made more modular at the same time, to prototype any modules we might want..

[agodinhost]

i think the challenge is power.. it's not insurmountable, but you have to route the 5v from the usb to the expansion socket before going to the controller circuit (or anything else...) and then provide an idiot-proof (because i'm an idiot and will inevitably screw this up) way to jumper the power when the expansion socket is not filled

Didn't understood this "expansion socket" part.

note that you have to do this all while fitting into the case designs that exist and are being worked on (!!)

Yup, I can help at the tests and proof of concept of these modules that I've proposed.

I'm not used to Kicad (I'm used to use eagle) but I don't have any compatible GH60 or the case.
+1 Komar
8-))

[agodinhost]

can you design a single board that can handle these things based on those two boards and that routes the external pins to the same form factor as the gh60 hackerpads? i'm going to try to run another set of prototypes of the gh60 board, and it would be just about right, if the gh60 is made more modular at the same time, to prototype any modules we might want..

I'm planning to test this whole idea on my breadboards at here, it will be easier to me:
First the charger and the booster, then the bluetooth (using the charger and booster).
After that, assuming that we still have two free pins, we will have to re-route a bit the design in order to free the tx/rx pins in order to assign then to the bluetooth module.

IF there is no free pins we will have to review the matrix ...

At this point I need to ask about which approach to take in this modularity idea:
1) just connect the original "modules" as they are (charger, booster, teensy, bluetooth)

2) "develop" our own modules (harder, not sure if it's a good idea, I don't like to reinvent the wheel but it would be smaller and would require less conectors I think)

Re-routing the gh60 board will take some time - I did already had that in mind when I first saw the GH60 design.
I'm quite new on keyboards layouts and I'm kinda afraid to mess up with the layout - it will take sometime as I still need to learn how to use Kicad.

We still need to decide which lipo to use.
Any 1s 1c low discharge would be good for me - but we still need to decide, there are too many out there.

[mkawa]

komar needs to redesign the pcb to get this, but there will be four free pins

by "socket" i mean that both boards should be designed to be mated by a commercially available male header and female socket. the gh60 beta board has something that's spaced about right for some molex product somewhere, but it only provides access to SPI. if you read the gh60 threads you'll see that he's planning on expanding this area to include the free gpio pins

[SmallFry]

The GH60 files are open source, if you guys wanted to tinker with them.

[agodinhost]

komar needs to redesign the pcb to get this, but there will be four free pins

Awesome, he already did an excellent work!

by "socket" i mean that both boards should be designed to be mated by a commercially available male header and female socket. the gh60 beta board has something that's spaced about right for some molex product somewhere, but it only provides access to SPI. if you read the gh60 threads you'll see that he's planning on expanding this area to include the free gpio pins

You mean this?
I like the idea - but a molex requires wires and I do hate wires! The crimping tool is way worst to use.

I was thinking on something simpler like those 40p Single Row 2.54m Double Male pin

You don't even need to use the female header - you can use the male header to solder both boards. The female header would be optional (thicker).

Something like what Dmw did into his HH:

[mkawa]

molex sells pcb-mounted headers and sockets too. no wires

molex, TE, etc. all have full lines..

[Strelok]

I was thinking horizontally mounted female headers and 90degree male pins when I pictured it in my head.

[agodinhost]

Please, could you guys post some links or pictures of what you do have in mind?
I'm trying to figure it out but I'm no parts expert (I'm a humble developer - underpaid and living on a jungle!)

[agodinhost]

It's a shame that this board is retired.
It works at 3.3v so it does not requires a 5v booster, uses the atmega32u4 and has a lipo charger.
We would have only to wire the keyboard matrix, plug the bluetooth module on it's xbee socket and connect the lipo.
Since it's using 3.3v it works at 8Mhz - to work at 16Mhz it will require the booster in order to step up the 3.7v lipo output to 5v.

They also have this another, not retired however out of stock, version here.
I don't know the difference ...

Well, it's open source, we can still take advantage of it's idea right?

By the way: Teensy is not 100% open source, they use one modified LUFA version (you have only the hex file - they didn't open the source of it).
I would go to arduino micro - it's quite new (2 months old, tops) and it's kinda hard to find at here yet.

[hasu]

That boards seems to be nice and we can learn from that schematic
3.3V and 8MHz configuration is clearly better in terms of power consume and It doesn't need 5-3.3V level shifter to interface with Bluetooth module.
Regarding power saving I don't know ATmega32u4 is better choice than other controller like PIC, ARM or etc. But this controller is common and almost only option for DIY keyboard around here.

Teensy's halfkay bootloader is not open but I think other tool/example sources is open. I'm not a fundamentalist, so to me no problem to use their product Shipping cost is a problem to me.

[agodinhost]

3.3V and 8MHz configuration is clearly better in terms of power consume and It doesn't need 5-3.3V level shifter to interface with Bluetooth module.

True, less 3 components for the tx line (the rx line does not requires a level converter - or it's the opposite?) and it will not require the booster too.
Way simpler.
I'm not sure if 8Mhz it's enough to run the tmk_keyboard firmware ...

Regarding power saving I don't know ATmega32u4 is better choice than other controller like PIC, ARM or etc. But this controller is common and almost only option for DIY keyboard around here.

PIC is really good and cheap - but it's kinda jurassic (sorry, it's) and all firmware code that we already have would have to be revised (too much work, IMHO).
8152? same as the PIC above.
ARM is not cheap and the simpler ARM is way too complex for me (Cortex 8? A9? too many options and as far as I know there is no cheap developer board for all these options yet - just a few). Too much for just a keyboard (IMO) and it would require firmware review too.
ATMEGA is cheap, well documented and the 32u4 is good for an usb ic without host (the at90 is way better - has complete host but it is not so cheap or easy to find).

Teensy's halfkay bootloader is not open but I think other tool/example sources is open. I'm not a fundamentalist, so to me no problem to use their product Shipping cost is a problem to me.

I'm on the same page man, the shipping cost to Brazil is crazy!
My last order on ebay was one Dell AT 102W, I got it by 20 bucks - the shipping cost were 50!!

I would say to continue with teensy - it uses the 43u4 anyway. It's not a big deal to change it later if we decide for that.

[hasu]

3.3V and 8MHz configuration is clearly better in terms of power consume and It doesn't need 5-3.3V level shifter to interface with Bluetooth module.

True, less 3 components for the tx line (the rx line does not requires a level converter - or it's the opposite?) and it will not require the booster too.
Way simpler.
I'm not sure if 8Mhz it's enough to run the tmk_keyboard firmware ...

Also I'm not sure and never tried it at 8MHz, but my HHKB runs without problem at 12MHz with ATMga328p(and V-USB) now.
I think It will and it should work at any proper cycle. If it doesn't work I'll work on that problem.

For a while I'll also stick with 32u4 because I've got spare 20 chips of that  from digikey

[mkawa]

cheaper forwarding to jp is something that can probably be set up no problem, since base usps postage will probably be fine. brazil otoh i'm not sure about; how is the govt mail service there?

arm chips are getting pretty cheap -- look at teensy 3.0 and the other lightweight cortex micro platforms that are popping up. pic's time is just generally over -- it is just a dying isa.

the issue with future keyboard designs will not be avr or arm but probably finding auxiliary MCUs for doing high current PWM of lots of analog devices: ie, RGB LED controllers. arm might get us slightly faster communication with these devices, but the cost of the led MCU will probably be much higher than the cost of the usb host micro..

[Strelok]

I would go to arduino micro - it's quite new (2 months old, tops) and it's kinda hard to find at here yet.

I have an Arduino Micro on hand, but all I've done with it thus far is make it blink.

[agodinhost]

Also I'm not sure and never tried it at 8MHz, but my HHKB runs without problem at 12MHz with ATMga328p(and V-USB) now.

ATMga328p DIP is simpler to solder and we won't need the crystal nor the 2 capacitors that it would require if not running at 8Mhz! Less components ...

I think It will and it should work at any proper cycle. If it doesn't work I'll work on that problem.

Deal, my C++ is worst than my english and it's way slow (my LUFA and V-USB knowledge is basic).

For a while I'll also stick with 32u4 because I've got spare 20 chips of that  from digikey

Rich guy! I do have 6, all atmel free samples. The Digikey or Farnell shipping cost for a resistor to Brazil is something like US$ 50,00!

---

... brazil otoh i'm not sure about; how is the govt mail service there?

Far from good: problems at the customs office, the postman sometimes fell in love by your package, sometimes the salt of the sea just mess up with your stuff and last but not least Murphy law is always there. The ETA for anything from US is 1 month, from China is 3 weeks (don't know why China is faster - just guessing that the guy at the customs office think it's something fake or cheap).

... look at teensy 3.0 and the other lightweight cortex micro platforms that are popping up.

I never used this version, 3.0 right? I don't know about it's acceptance, still didn't saw a living soul using this board ...

... the issue with future keyboard designs will not be avr or arm but probably finding auxiliary MCUs for doing high current PWM of lots of analog devices: ie, RGB LED controllers. arm might get us slightly faster communication with these devices, but the cost of the led MCU will probably be much higher than the cost of the usb host micro..

The number of free ports is important too, IMHO.

---

I have an Arduino Micro on hand, but all I've done with it thus far is make it blink.

Camon!!! you just turned it on - it cames blinking already!
8-)

[Strelok]

Please, could you guys post some links or pictures of what you do have in mind?
I'm trying to figure it out but I'm no parts expert (I'm a humble developer - underpaid and living on a jungle!)

Bottom is what I had in my head when I started this thread, but I don't know how difficult USB is to program for.
Top is what I imagine after reading your EE jargon.

[komar007]

Catching up with the thread...
As far as I understand, what I need to add is a header for the expansion board. The expansion board should have BT and power chip with a battery and a battery charger.
So I need to provide the 4 remaining I/O pins, VCC, GND and VCC output, so that the expansion board can provide power either from a battery or connect VCC input to VCC output in USB mode, and charge the battery when USB is connected.
VCC will be connected to VCC output on the motherboard with a solder bridge. When one wants to use the expansion board, they'll desolder the bridge. That's how I see it...


EDIT: or maybe that VCC and VCC output solution is unnecessary. Just VCC should be enough.

[TDub]

Another link that might be useful:
http://little-scale.blogspot.de/2013/01/teensy-xbee-simple-native-usb-wireless.html

A setup with teensy and xbee (2 of each) which has the battery stuff all figured out. Not sure if it can do charging of the batteries though.

[agodinhost]

... As far as I understand, what I need to add is a header for the expansion board ...

Err, nope, as far as I understood the idea is to build one modular keyboard matrix without any specific controller on it. This board will have some kind of headers to connect it's columns/rows pins to our external controller bundle. It will required a major change since we will have to remove all teensy stuff from the current version and re-route the columns/rows pins to these headers (we still didn't decide about which headers to use).

Komar, I think it would be better to wait a bit - we are still deciding which path to take.

I'll add some sketches to let this idea clear, later today.

[komar007]

Okay,
I'll break out the pins anyway though, since I need them for hacking purposes.

[komar007]

Double post:/