• Please review our updated Terms and Rules here

Determining keyboard protocol for dumb terminal

mykrowyre

Experienced Member
Joined
Jan 28, 2019
Messages
171
Location
Saint Augustine, FL
This may not be the correct forum but I figured people here would have the most experience with dumb terminals.

I recently purchased a 1992 Liberty Freedom (ASPECT?) amber dumb terminal.

It works, but the keyboard is not the correct keyboard, and it did some damage when the seller tried to use it.

I repaired the damage to the keyboard but sadly the keyboard can't be made to work with this terminal as it's a data/clock style keyboard.

Tracing the terminal mainboard circuit it seems to be serial, I found the keyboard RJ6 connector has the following pinout (see photo).

I've tried to find pinouts for other terminals of the time but haven't had any luck.

One signal pin is and input and the other an output of the 7407.

Not sure if it's serial TX/RX, or if it's data a data pint from keyboard and a clock pin from the terminal?

There is one pin which is not connected to anything.

Manufactured sticker date says 1992.

Finding an ASPECT keyboard is nearly impossible, so I am hoping they used a common protocol/keyboard design of the time but I don't know what that would be.
 

Attachments

  • PINOUT.jpg
    PINOUT.jpg
    133.5 KB · Views: 9
Last edited:
This may not be the correct forum but I figured people here would have the most experience with dumb terminals.

I recently purchased a 1992 Liberty Freedom (ASPECT?) amber dumb terminal.

It works, but the keyboard is not the correct keyboard, and it did some damage when the seller tried to use it.

I repaired the damage to the keyboard but sadly the keyboard can't be made to work with this terminal as it's a data/clock style keyboard.

Tracing the terminal mainboard circuit it seems to be serial, I found the keyboard RJ6 connector has the following pinout (see photo).

I've tried to find pinouts for other terminals of the time but haven't had any luck.

One signal pin is and input and the other an output of the 7407.

Not sure if it's serial TX/RX, or if it's data a data pint from keyboard and a clock pin from the terminal?

There is one pin which is not connected to anything.

Manufactured sticker date says 1992.

Finding an ASPECT keyboard is nearly impossible, so I am hoping they used a common protocol/keyboard design of the time but I don't know what that would be.
since you have the board out, some pictures and a dump of the firmware would be helpful, since I have no info on the Aspect
at http://bitsavers.org/pdf/libertyElectronics
 
more digging, this is a Freedom 120, evolved from the freedom one.
this terminal will kill most incompatible kbs because it regulates 12v to 5 inside the kb
you can find a pic of the insides, I REALLY wish these keydingos would dump the firmware
 

Attachments

  • Screen Shot 2022-03-01 at 11.24.26 PM.png
    Screen Shot 2022-03-01 at 11.24.26 PM.png
    1.7 MB · Views: 12
this is the correct keyboard on an Aspect 100 / Freedom 120
 

Attachments

  • s-l1600-5.jpg
    s-l1600-5.jpg
    229.6 KB · Views: 9
this is the correct keyboard on an Aspect 100 / Freedom 120
Hi Al, yes everything you stated is correct. Thats the auction, yes its regulated from 12 to 5 and the pinout was wrong so it destroyed a 74LS07 inside the wyse style keyboard. Repaired that and hoped it would work if I got the pinouts right, but the protocol appears different because the aspect uses only a single input and a single output instead of data+clk.

I will post some pictures of the terminal and main board and will dump the firmware. I did see that other auction and the original keyboard looks nearly identical to the keyboard I have so I can see how they thought it might work.

Excited to find out what if anything you conclude from seeing the mainboard.
 
Just a guess, but I remember putzing around with some hi-tek terminal keyboards years ago that were 12v and similar... Had a fun time trying to connect them to an old Unisys terminal. If I recall correctly it was generic TTL serial shift+ascii @ 9600 baud.

pinout for those was

  1. switch 20 (Reset)
  2. -INT, so an input to keyboard
  3. P27, so an output from keyboard
  4. ground
  5. DC in
  6. chassis
 
Last edited:
Thanks! If I can get enough information from the pictures about parts, osc freqs, etc. and the firmware there is someone
on the MAME team that likes to emulate terminals and we can get a simulation running to help puzzle out the keyboard
format. I'll check if there is already a simulation for any earlier models.

The protocol of Wyse keyboards is on the keyboard babel site, so if the ASIC inside wasn't toasted you can probably
build a protocol converter for it.

Actually, looking at https://geekhack.org/index.php?topic=51079.0 It may not have been a Wyse keyboard. A picture
of the board inside what you have would be helpful to ID it. It might be for ADDS
I have one of the 4000 series ADDS keyboards, and it looks a lot like it (5v, of course..)
There is one brand of AS/400 terminals that did the same thing using 12v and I blew out
the microcontroller in a normal DIN keyboard when I plugged it in. After that I always
check what voltage is going out on the DIN5 connector before plugging a keyboard
into that on a terminal I've never seen before.

Interesting.. no Liberty terminal emuated (the only firmware I have is for the 200)
but, looking at the pcb scan the keyboard pinout is the similar your unit.

G - T R G +12
 
Last edited:
Any clues on the terminal board itself? i.e., UARTs, etc. that might be connected to that 7407?
Ok, it's a SCN2861 DUART. Looks like only pins 10 & 11 (TXDB/RXDB) are used for keyboard. Crystal is 3.6864 as listed in the datasheet.

So thanks for asking me to dig further, it's clearly a serial keyboard and the serial protocol is described in the datasheet.

What are the odds of finding a keyboard which will work? I suppose the key mappings may also be different than other terminals.

Here's the keyboard it arrive with (which apparently is not the correct keyboard)


Photos and rom dumps follow
 
Last edited:
 
Here's the service manual for the Freedom 100 which might give some clues.

It does mention an RJ11 style jack for keyboard connection, and discusses keyboard operation, but I am not seeing (or possibly understanding) info about the serial connection.

 
a much higher res version of that is on bitsavers.
the freedom 200 has three 8251 serial chips, and I bet one of them was used for the keyboard

unfortunately, there is only a two-channel chip on the freedom 120 and no other obvious serial chip
it has to be there somewhere, maybe integrated in the 48 pin video attributes asic

i've sent the info off to MAME, hopefully someone there will be interested in getting it and the
Freedom 200 emulated.
 
if you have a way to send ttl 9600 baud data into the receive pin from some other device, you should be able to puzzle some of it out
the trick is identifying the non-printable keys, or if they are doing key-down, key-up protocol. sometimes they use setting the msb to
identify up or down.

the MAME guys have gotten really good at deciphering terminal keyboard protocols
 
Well, where I'd start (given the presence of a UART connection to the keyboard) is with one of those USB-to-TTL serial dongles. Just connect the serial-in/serial-out/ground leads up. Use something like minicom (Linux) or Realterm (Windows) and see what happens at various speeds, parity, etc. Likely, the keyboard is running at 9600 or 19200 n-8-1.

If you do MCU programming or use an RPi, you likely already have one of these in your kit.
 
Right thanks I was actually considering that as my next plan of action and my serial ttl board is already out on the bench. sounds it wasn't such a bad idea after all.
 
the freedom 200 has three 8251 serial chips, and I bet one of them was used for the keyboard

unfortunately, there is only a two-channel chip on the freedom 120 and no other obvious serial chip
it has to be there somewhere, maybe integrated in the 48 pin video attributes asic

i've sent the info off to MAME, hopefully someone there will be interested in getting it and the
Freedom 200 emulated.
I did trace the keyboard pins back to the dual uart so that is definiitely it.
 
Well this has opened up a world of keyboard hacking and collecting that I never knew existed. Pretty interesting actually.

I just tried using a terminal program and TTL serial device and just can't get it to respond to any baud or stop bit/parity settings.

I wired the RS232 connector into loopback as well, but nothing so far for the keyboard.

I connected to the RS232 port and was able to send to the terminal at 9600,8,n,1 with Wyse 50 escape codes and see it on the terminal screen so the UART seems to be working at least for the RS232 port.
 

Attachments

  • IMG_5606.jpg
    IMG_5606.jpg
    96.3 KB · Views: 12
Last edited:
Back
Top