What Can I Do With a Terminal Model M?

[lkong]

Try this:
https://geekhack.org/index.php?topic=61878.msg1721863#msg1721863

And yes...  you can also do it to the 1392595, etc etc.... 
Just go to my gallery of pictures in my sig to see more transformations.

Show Image

If i had a faulty controller I'd rather fabricate a alternative controller for that instead of trying my luck on ebay.

I'm trying figure out here is: is looping on print "wFF" in debug console a software issue or a hardware one.

Thanks for the advise

[Snowdog993]

You can order the controller from Unicomp as well.

The replacement logic assembly for 1391401 is $20. To order use the link below:
http://pckeyboard.com/page/KBDCFG/CUSTOM
Enter qty 2
In the customization box enter "logic assembly 1378170"

Edit: I had to dig through all the Unicomp service tickets to finally find it.  Same controller.  Same price too.

Oh yeah!  I nearly forgot! The 4pin 4P 4 pin Arduino jumper cable.  I bought them to make my job easier.  Worth every penny.  You can get the M/M version or F/F version and use 4-pin headers.  Your choice.

http://www.ebay.com/itm/171445075322

I believe that covers everything.  All in all it is rewarding to do this project.

[Snowdog993]

If i had a faulty controller I'd rather fabricate a alternative controller for that instead of trying my luck on ebay.

I'm trying figure out here is: is looping on print "wFF" in debug console a software issue or a hardware one.

Thanks for the advise

In your particular case, you may want to get the RJ45 adapter from Orihalcon.  What this IS would be a soarer's adapter that directly plugs into the RJ-45.  You plug one end into that, and the other into the USB port.  Do all your remapping etc.  At least you would know if your terminal controller is good.

http://www.ebay.com/itm/281476431381?afsrc=1&rmvSB=true

The language that controller speaks is different than a PC would understand.  I am sure your terminal controller is good.  That's not the problem.  It is the translation language of the controller to your PC.  You might as well speak "wFF" with it all day long.  Your PC does not understand terminal keyboard controllers with a simple wiring.  You have to have a translator.  (Soarer's, Teensy, Orihalcon etc.)
Note: Be sure you don't have the connections mirrored as Fohat mentioned if you have a Teensy.  It might be that you flashed the wrong firmware too.  Be that as it may you still might want to try the other way.

--- OR ---

Just get a keyboard controller that works properly to begin with.  Pull the terminal controller out, put in the 1391401 controller.  That's all I am talking about.

Now may I please stop hitting my head against the wall trying to explain this? 
I think I spent more time editing and re-editing this message till my eyes have crossed!
I hope I helped.

[dfj]

wFF is printed out by soarer's software, (and also by TMK, perhaps?) in a debug console.

It means the converter wrote FF to the keyboard, repeatedly, with no response.

Presumably either the wires were not hooked up correctly, (easiest error is to accidentally swap clock and data, as mentioned) - or there is some more exciting problem with the controller.
  Feel free to drop by irc some time, we can likely figure out which with a few quick tests, otherwise - the forum. :)

I should be awake in about an hour... making coffee.
 
 

[Snowdog993]

wFF is printed out by soarer's software, (and also by TMK, perhaps?) in a debug console.

It means the converter wrote FF to the keyboard, repeatedly, with no response.

Presumably either the wires were not hooked up correctly, (easiest error is to accidentally swap clock and data, as mentioned) - or there is some more exciting problem with the controller.
  Feel free to drop by irc some time, we can likely figure out which with a few quick tests, otherwise - the forum.

I should be awake in about an hour... making coffee.
 
 

I went to bed after that post.  Just woke up now.

I realized what you were talking about when I saw that.  Yeah you might have a problem with the firmware or the terminal controller.  I really don't know, and I am not an expert on that.  There are several that know a LOT more than I do about getting the converters to work.  Fohat or Orihalcon would be much more helpful in this respect. 

Sometimes I prefer simple than fussing around with that.  That's part of the reason why I made the post.  It is an alternative.  Plus it makes for an easy option.  (IBM pun)

[Skechup]

You can also map the keys on the computer if I am correct. But I might be wrong on that one.

[lkong]

Please do a sanity check on my connection before I declare death on this controller.
firmware is the latest master from hasu's github repo.

[dfj]

1) The solder joins at the 5 pin connector (only four of which are used) look shot. should touch those up, likely whatever bent it also broke those joins - but the traces are likely fine, so it's an easy thing to fix and rule out.
  if hasus stuff runs on d5 and d2, then you are good, there... clock is on the lower left pin, and data on the corner pin beside vcc and gnd.
gnd is in the middle between clock and data, and vcc is alone on its row... (how I remember them)

So - I'd reflow those solder joins - and don't leave a controller resting with B3 and B4 possibly touching a grounded bolt. :)

I can't see what is going on with that brown component next to the connector either - might want to double check it isn't shorting to anything, and has good joins as well.

[lkong]

1) The solder joins at the 5 pin connector (only four of which are used) look shot. should touch those up, likely whatever bent it also broke those joins - but the traces are likely fine, so it's an easy thing to fix and rule out.
  if hasus stuff runs on d5 and d2, then you are good, there... clock is on the lower left pin, and data on the corner pin beside vcc and gnd.
gnd is in the middle between clock and data, and vcc is alone on its row... (how I remember them)

So - I'd reflow those solder joins - and don't leave a controller resting with B3 and B4 possibly touching a grounded bolt.

I can't see what is going on with that brown component next to the connector either - might want to double check it isn't shorting to anything, and has good joins as well.

that brown component is a 1.5k resistor, which i read somewhere acts as a pull up between Vcc and Data. The board doesnt work with or without it.

I reflowed all the joints and checked traces. Same problem.

It's time to buy a new controller. thanks a lot for the help.

[hasu]

I better have pull up resistors on both data and clock lines, I think that short wires works without pull-ups but not sure.
The wiring looks ok to me I don't know why it doesn't work and I don't have this converter hardware to test right now.

Try connecting clock to D0 and data to D1 and build firmware after changing lines in Makefile like this.
This pin usage is compatible to Soarer's so you can use it in case

Code: [Select]

#
# PS/2 protocol implementations
# USART is recommended if it is available, others are for reference purpose.
# INT implementation will drop simultaneous key strokes.
#
#PS2_USE_USART = yes # uses hardware USART engine for PS/2 signal receive(recomened)
PS2_USE_INT = yes # uses external interrupt for falling edge of PS/2 clock pin
#PS2_USE_BUSYWAIT = yes # uses primitive reference code


[fohat.digs]

acts as a pull up between Vcc and Data

I follow Soarer's advice and have pull-ups between both data and clock to power.

[lkong]

I better have pull up resistors on both data and clock lines, I think that short wires works without pull-ups but not sure.
The wiring looks ok to me I don't know why it doesn't work and I don't have this converter hardware to test right now.

Try connecting clock to D0 and data to D1 and build firmware after changing lines in Makefile like this.
This pin usage is compatible to Soarer's so you can use it in case

Code: [Select]

#
# PS/2 protocol implementations
# USART is recommended if it is available, others are for reference purpose.
# INT implementation will drop simultaneous key strokes.
#
#PS2_USE_USART = yes # uses hardware USART engine for PS/2 signal receive(recomened)
PS2_USE_INT = yes # uses external interrupt for falling edge of PS/2 clock pin
#PS2_USE_BUSYWAIT = yes # uses primitive reference code


This thing took a unexpected turn here:
The controller board might be ok.
So Ive unplugged teensy from the controller, still getting wFF from hid_listen.
Then I decided to check if the teensy board is faulty.
It's not! I tried to load blinky.hex from teensy's official website, works fine, the hid console output is correct too.

Then try soarer's converter (1.12).
I got this output first time plugged in:

Waiting for new device:...
Listening:
wEE

remaining: FFFC


Keyboard ID: 0000
Code Set: 1
Mode: PC/XT

R05

Still not registering key strokes though.
 

[Snowdog993]

I better have pull up resistors on both data and clock lines, I think that short wires works without pull-ups but not sure.
The wiring looks ok to me I don't know why it doesn't work and I don't have this converter hardware to test right now.

Try connecting clock to D0 and data to D1 and build firmware after changing lines in Makefile like this.
This pin usage is compatible to Soarer's so you can use it in case

Code: [Select]

#
# PS/2 protocol implementations
# USART is recommended if it is available, others are for reference purpose.
# INT implementation will drop simultaneous key strokes.
#
#PS2_USE_USART = yes # uses hardware USART engine for PS/2 signal receive(recomened)
PS2_USE_INT = yes # uses external interrupt for falling edge of PS/2 clock pin
#PS2_USE_BUSYWAIT = yes # uses primitive reference code


This thing took a unexpected turn here:
The controller board might be ok.
So Ive unplugged teensy from the controller, still getting wFF from hid_listen.
Then I decided to check if the teensy board is faulty.
It's not! I tried to load blinky.hex from teensy's official website, works fine, the hid console output is correct too.

Then try soarer's converter (1.12).
I got this output first time plugged in:

Waiting for new device:...
Listening:
wEE

remaining: FFFC


Keyboard ID: 0000
Code Set: 1
Mode: PC/XT

R05

Still not registering key strokes though.
 

It's going mode 1 instead of mode 2?  Maybe that's the problem?  It should be going mode 3 to mode 2.
That could be the issue right there.

acts as a pull up between Vcc and Data

I follow Soarer's advice and have pull-ups between both data and clock to power.

Agreed Fohat.

I'm assuming it's TMK?
https://geekhack.org/index.php?topic=27272.0

[Snowdog993]

It's going mode 1 instead of mode 2?  Maybe that's the problem?  It should be going mode 3 to mode 2.

I was wondering if there was a resolution to that controller issue.  Did you wind up swapping the controller out?

[lkong]

It's going mode 1 instead of mode 2?  Maybe that's the problem?  It should be going mode 3 to mode 2.

I was wondering if there was a resolution to that controller issue.  Did you wind up swapping the controller out?

I will try to throw in another teensy first, which i used for a point to point wiring keyboard.
Then I'll see if i should replace the controller.

According to the source code wFF is the code for reset and teensy keeps pooping those in console even without connecting to the model m controller sounds super suspicious.

[Snowdog993]

It's going mode 1 instead of mode 2?  Maybe that's the problem?  It should be going mode 3 to mode 2.

I was wondering if there was a resolution to that controller issue.  Did you wind up swapping the controller out?

I will try to throw in another teensy first, which i used for a point to point wiring keyboard.
Then I'll see if i should replace the controller.

According to the source code wFF is the code for reset and teensy keeps pooping those in console even without connecting to the model m controller sounds super suspicious.

For some odd reason, I am still wondering if you ever got that keyboard working!

[lkong]

It's going mode 1 instead of mode 2?  Maybe that's the problem?  It should be going mode 3 to mode 2.

I was wondering if there was a resolution to that controller issue.  Did you wind up swapping the controller out?

I will try to throw in another teensy first, which i used for a point to point wiring keyboard.
Then I'll see if i should replace the controller.

According to the source code wFF is the code for reset and teensy keeps pooping those in console even without connecting to the model m controller sounds super suspicious.

For some odd reason, I am still wondering if you ever got that keyboard working!

here's the updates.
I purchased universal model M keyboard converter from here: http://phosphorglow.net/store/products/universal-model-m-usb-controller/
Lots of keys are still not registering, although some are functional.
I pulled out membrane and traced it.
Turns out the "row" membrane is faulty: http://i.imgur.com/F82k63r.jpg
There's a fold close to the connector, thus the connection is really bad (connected in certain angle).
I'm in this too deep now to bail out.
I've contacted unicomp asking for replacement part number.
I'll report back on that.

[lkong]

Instead of waiting for Unicomp I managed to patch the trace by myself.
I used copper foil tape (although amazon item description said adhesives are conductive it's not) and conductive paint.
http://imgur.com/a/tbUD4#NbGvVbu
http://imgur.com/a/tbUD4#nQFq1pW

[Snowdog993]

So is everything all good now?  Sounds to me you got it all sorted out.

[lkong]

So is everything all good now?  Sounds to me you got it all sorted out.

Everything is pretty much working now.
'X' is a little hard to register, but im too lazy to unscrew 20 bolts to fix it tonight.
It's such a long journey, and i'm glad i got to the end of it.

[paicrai]

adapter