|
Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment. |
|
Thread Tools |
25th Nov 2020, 2:27 pm | #1 |
Dekatron
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,191
|
Computest 3020
This could possibly go under test equipment, but it's a vintage computer-controlled device.
Has anyone come across a unit called a Computest 3020? It's a luggable case with fold-down keyboard revealing a monochrome (green screen) monitor and a pair of 5.25" floppy drives. Inside there's the obvious monitor, disk drives and PSUs which I'll say no more about. A processor board slides in from the back. It contains a Z80 processor, 128K DRAM, a 4K byte boot ROM (of which not much is actually used) video display circuitry (Intel 8276 cbip), disk controller (Western Digital 2797), a couple of RS232 serial ports, etc. On top of that is the 'measurement system'. This seems to be a multi-channel ohmmeter (each switch board, and you can have up to 8 of them, adds 64 pins, it can I think measure the resistance between any pair of pins) and a bit more, including some kind of frequency/pulse width counter. What was it used for? I have one, it seems to work but all it does is complain that it can't boot. Not surprising as I don't have any disks for it. I guess I could try to port CP/M to it but I am not a programmer... |
25th Nov 2020, 2:32 pm | #2 |
Octode
Join Date: Mar 2011
Location: North Yorkshire, UK.
Posts: 1,087
|
Re: Computest 3020
Looks familiar. Kaypro?
|
25th Nov 2020, 2:51 pm | #3 |
Dekatron
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,191
|
Re: Computest 3020
It doesn't seem identical to any of the common CP/M luggables, alas.
One oddity is that the keyboard (but nothing else) is electronically identical to that on a Nicolet 800 logic analyser. |
25th Nov 2020, 3:33 pm | #4 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,265
|
Re: Computest 3020
With that many resistance measurements available it could be an ICT, in circuit tester, for use with a bed of nails test fixture.
Strange that it would be in a luggable rather than rack mount case if that was its purpose. Is it four wire resistance measurements? Perhaps with ability to drive adjacent test nodes to the same voltage as the sense wires to isolate resistance within networks. |
25th Nov 2020, 3:42 pm | #5 |
Dekatron
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,191
|
Re: Computest 3020
Yes, the fact that's it's luggable with so many external connections puzzles me too. Maybe it was intended to be moved between test rigs or something. I'm hoping somebody says 'Oh I remember that thing, we used it for <foo>'
It's 4 wire up to the switching circuits (4052 analogue multiplexers) but they're wired so the current source and voltage measurement circuits end up on the same pins. |
25th Nov 2020, 3:47 pm | #6 |
Hexode
Join Date: Jun 2018
Location: Buckinghamshire, UK.
Posts: 382
|
Re: Computest 3020
|
25th Nov 2020, 3:52 pm | #7 |
Tetrode
Join Date: Feb 2007
Location: Liverpool, Merseyside, UK.
Posts: 99
|
Re: Computest 3020
It's a Manufacturing Defect Analyzer. The attached image is from the 'New Products' section (page 98) of a 1985 issue of Design&Test magazine.
|
25th Nov 2020, 3:58 pm | #8 |
Dekatron
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,191
|
Re: Computest 3020
Thank you. So I guess it did link to a bed-of-nailstype fixture.
At least I know what it was used for now, and my deduction that it is, in part, a mult-channel ohmmeter was not too wide of the mark. Quite what I do with it is another matter. As I said,porting CP/M (or FIG forth) seems a possibilty if I can manage to write the necessary code. |
25th Nov 2020, 5:27 pm | #9 |
Octode
Join Date: Mar 2011
Location: North Yorkshire, UK.
Posts: 1,087
|
Re: Computest 3020
You may well find that underneath the resistance test software it is CP/M
|
25th Nov 2020, 5:32 pm | #10 |
Dekatron
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,191
|
Re: Computest 3020
Alas I have no disks for it at all.
|
26th Nov 2020, 4:21 am | #11 |
Heptode
Join Date: Aug 2009
Location: Worthing, Sussex, UK.
Posts: 661
|
Re: Computest 3020
When you get to powering it on does it display any messages. Also an image/disassembly of the boot rom would yeild a lot of infomation.
If it is capable of running CP/M the one thing you will need to work out what to do in code is to disable/remap the boot rom as CP/M needs RAM at address 0. Other than than you could be lucky and find code in the boot rom to do console i/o and also disk read. Disk write will be an easy modification of disk read with a datasheet of the controller. Any messages it shows on power up may also aid identification. Also a list of the "BIG" chips would also give some idea. eg is there a second processor for the screen/keyboard etc |
26th Nov 2020, 7:54 am | #12 |
Dekatron
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,191
|
Re: Computest 3020
The only message is something like '*** No Boot...timeout on track 0 sector read' No copyright/manufacturer's name.
I've dumped the boot ROM but haven't got round to disassembling it yet. It's a 2732 (4K bytes) of which less than 1K is used (the rest is all FF's) There's a small PROM (also dumped) which acts as the memory address decoder. It allows the boot ROM to be disabled and selects between the 2 banks of DRAM. As for the large chips, they are : Z80B-CPU (although clocked at only 4MHz) 8237 DMA Controller. 4 channels, 1 not used, one for the floppy controller, one for the video (a repeating transfer is set up on this channel for the video chip to load its line buffer from RAM), the last is available on the bus connector but is not used by the measurement system that connects there. 9519 Interrupt controller 8276 Video chip 8255 parallel interface, port A is used (mode 1, thus with part of port C) for the keyboard, port B for floppy drive selects, etc. WD2797 floppy controller 2 off 8251 USART for the serial ports 8254 counter/timer for the baud rate clocks and a heartbeat interrupt. The keyboard outputs a 7 bit code (I assume ASCII) to the 8255. There's no second processor for I/O I've traced out the schematic (too large to post here) so I know the port addresses, etc. |