View Full Version : An idea for Kenbak-1 data entry assistance

07-02-2007, 08:26 AM
Here's an idea for Kenbak-1 data entry assistance that's inexpensive and doesn't interfere with the operational authenticity of the replica. It could even be adapted to original equipment Kenbak-1 units.

Start with getting an inexpensive PIC or similar microcontroller chip from Parallax or similar vendor. As long as it has plenty of general purpose I/O pins, it'll probably work. Take the Kenbak-1 front panel push button switch set signal output wires, sever them from the motherboard, and connect them to inputs on the PIC. Then connect the same number of PIC outputs to the appropriate Kenbak-1 motherboard pads. Allocate a few extra connections on the PIC for a serial port, and then load the PIC with a little monitor program.

In the PIC's default state, all signals from the Kenbak-1 front panel are passed through untouched and there's no change in operation. However, when given the right inputs from an external source via its serial connection, the PIC will temporarily takeover the Kenbak-1 front panel push button input and supply its own input instead.

A few more modifications and maybe some glue chips could have the PIC handle all the front panel controls along with relaying the front panel indicators along the serial link to the external controlling source.

Grant Stockly
07-02-2007, 05:21 PM
You read my mind. :)

I was thinking about using an AVR. They have enough EPROM to store around 30-50 memory images of the kenbak. So you could keep your kenbak programs inside the kenbak...

Too many ideas!

07-07-2007, 07:48 PM
I wonder if there ever were 30 to 50 different programs ever written for the Kenbak-1.

Do you know if the original designer ever had a prototype for a card reader for the otherwise superfluous slot in the front panel? I figure it would take about a meter of paper tape to hold an entire memory image, so it's hard to see how that much data could be recorded on a small card given the state of technology at the time.

The HP-65 calculator had a 100 byte magnetic card capability, but it didn't arrive on the scene until 1973 (two years after the Kenbak-1). A few years later, the HP-67 was announced (I bought one), and it used higher density recording to store a full 256 bytes per card.

Grant Stockly
07-07-2007, 08:43 PM
I asked about that, and this was his response:

The slot on the front panel (that is covered up) was for a possibility that
never materialized. I thought that I might be able to use a punched cord.
The holes in the card might perform the same functions as the input

I guess I imagine something like a tile sample ring of cards???

07-08-2007, 01:00 AM
The standard paper punch cards of that era had eighty columns and twelve rows for a total of 960 bits capacity. But a programmer needed access to an IBM 029 keypunch or the like and this wasn't something to be seen in anyone's home.

Paper tape would have been a better idea for the Kenbak-1 as tape punching Teletypes were fairly common at the time. And it was a fairly mature technology that had been in use for many decades.

See the "ppt" program on your favorite Unix:

| o o. oo|
| oo .o o|
| oo o.oo |
| oo . o |
| oo . o|
| oo o. oo|
| o o.o o|
| oo . o|
| o . |
| ooo . |
| oo . o|
| ooo . |
| oo .o o|
| ooo . o |
| ooo .o |
| oo . o|
| ooo . |
| oo .o o|
| o . |
| oo .o o|
| oooo. |
| oo . o|
| oo o.o o|
| ooo . |
| oo o.o |
| oo .o o|

07-08-2007, 01:17 AM
Although it doesn't output a full eighty columns, here's the result of the unix "bcd" program:

| ] ]] ]]]] ] ] ] ] |
|] ] ]] ] ] ] ]]] |
| ] ] ] |