dn300 keyboard emulator progress (and q)

[tos...@gmail.com]

Hey all,

I've been able figure things out to the point where I can connect an 8P8C (non-keyed) cable + rpi pico + usb keyboard and type things and they on the screen and they work. For instance `DI N` + `LD` prints out the node id + fails (the dn300 is the only token ring machine I have.)

But here's the rub: I can't seem to locate a Vcc pin on the keyboard connector.

State of my understanding:

8P8C pin Function
(kb side)
=====================
1 ??? (1 and 2 seem to be tied together)
2 ???
3 ???
4 KB_DATA. 5v (from a pull up resistor?) from the apollo side
5 KB_CLOCK. putting out 1200 Hz which matches the baud rate
6 KB_GND
7 ???
8 ???

KB_GND connected to the rpi pico's GND.

KB_DATA connects to a level shifter (with 5V coming from the external supply powering the pico and 3.3V coming from the pico's 3.3V output), and is also tied to GND through a resistor.

The 3.3V input to the level shifter connects to one of the pico's UART tx pins. the connection is (afaict) not bidirectional. 8N1 @1200 baud.

Current theories about how the keyboard gets power, and why I'm not seeing +5V anywhere but KB_DATA, range from:
1. Vcc is on one of the 5 unknown pins and there's something shot on the apollo side that I need to repair/replace.
2. The keyboard somehow powers itself from the current it can steal from KB_DATA (diode + caps + .. whatever)
3. There's an additional KB_SENSE that when asserted properly causes +5V on another of the pins.

#1 is definitely plausible. I've had to replace other things (oscillator for the ACIA the keyboard connects to, in particular.)
#2 Seems pretty crazy
#3 I think I've ruled out. unless there's something like #1 affecting that pin too.

Any thoughts?

If there's anyone out there with a dn300 (there must be someone someplace :) I'd also _love_ to see photos of the circuitry inside the keyboard.

Will get my schematic in order and put everything up on github shortly.

-c