[H-GEN] Vintage computing mystery

[Arjen Lentz]

Adding to 

> On Fri, 2018-02-02 at 12:25 +1000, Jack Newport wrote:
>> I found this old board/card in a box of miscellaneous scrap computer
>> parts. It has been assembled by hand as you can see in the photos.
>> One of the chips is labelled with the date 15/6/92.
>> 
>> Pictures:  https://imgur.com/a/Aewpc

Further to Russell's contribution and the others.

Yes it's a hand-crafted Eurocard. Excellent.
https://en.wikipedia.org/wiki/Eurocard_(printed_circuit_board)

To the right of the CPU is the ceramic (brown) EPROM chip (the sticker on the glass is to prevent stray UV light from accidentally part-erasing), so it'll contain the program code to run on the CPU.
https://en.wikipedia.org/wiki/EPROM
If you can tell me the number under the sticker, I could tell you more about its capacity.
It'll be something like 27xx or 27xxx indicating the # of kilobits, which you divide by 8 to work out kilobytes.
So a 27x128 would be an 8K EPROM, and a 27x256 would be a 16K EPROM (those two were the most commonly used ones).
The x is an optional letter, for instance C to indicate CMOS architecture.

To the right of the EPROM is a static RAM chip, I think it's a 8Kx8, I used to use those myself (often stacked in a 16Kx8 ROM slot).
https://en.wikipedia.org/wiki/Static_random-access_memory
Static RAM keeps its charge as long as it's powered, it does not need refresh logic - more expensive but much easier to work with in circuits.

Below the EPROM is a chip with a code starting with PAL, I think that'll be a programmable logic array.
See https://en.wikipedia.org/wiki/Programmable_Array_Logic for details

On the far right are standard buffer chips, the 74x245 are bidirectional buffers.
Anything 74xx is a TTL IC, that means they work on 5V signal levels.
the LS designation after the 74 has meaning in the realm of lower power requirements and higher speed (LS = Low-power Schottky)
(ref: https://en.wikipedia.org/wiki/Schottky_diode)
Typically any external interface would use 244 or 245 chips.
244 is uni-directional, 245 is bidirectional with one pin controlling the direction.
The buffers provide circuit protection, they might blow but are cheap (a mere few dollars even in 80s and early 90s) whereas the chips they're protecting are more costly.


As Russell noted, the 1488 and 1489 are RS-323 / TTL level converters, proper RS-232 serial ports operate on 12V level, whereas TTL is 5V which is what the rest of the circuit board used.
Actually, the proper term is Line Drivers, see http://www.datasheetarchive.com/1488-datasheet.html

The two silver things are crystals, if you check the code on the one below the CPU, we'll be able to figure at what clock rate the CPU was running!

It's possible to read the EPROM and run it through a disassembler to see what the code was doing.
The electric logic is really simple, someone would be able to rig up something for that using an Arduino on a board in not much time.
http://www.eevblog.com/forum/microcontrollers/reading-out-old-ep-rom-on-arduino/

I am not familiar with 68xx assembly, I was more on the 6502 track at that time.
Finding 6809 disassembler is easy though, quick Google search finds https://github.com/Arakula/f9dasm and there'll be others.
And it should be fairly readable regardless of the CPU flavour, they were all fairly simple.

It's all straightforward stuff, reverse engineering of old gear is easy-as (as long as you have an old[er] sock nearby who can decode the magic numbers from the time).
Depending on what you want to do with it, it's cool computer-archaeology.


As for the purpose, it's overkill to create this design for just a simple serial interface.
Even a serial interface with a bit of buffering doesn't require CPU/RAM etc in the circuit, there were other options for that.
So the circuit specifically provided some intelligence in the interface.
It doesn't have a lot of RAM, so the RAM is mainly for the program to operate, rather than as a large storage buffer.
Without docu, I don't think you're going to find out what it was doing unless you disassemble the code.
Oh look even that's online now, https://archive.org/details/6809AssemblyLanguageProgramming
That book looks awesome, it appears to have all the necessary info on memory layout (e.g. where things need to be in the ROM, where the CPU starts executing on startup, etc) and so on.

So all the bits are there, you just have to decide how deep you want to dig to satisfy your curiosity!


Regards,
Arjen.