ICL OPD

[RWAP]

On the subject of programs - does anyone have any information / ROM cartridge images for any sofware that provides serial interface for the ICL OPD?

I know there was the 64180 based CPM module which supported disks and a serial port, I also know there was a unit based on a 80C31 which provided a DB25 serial interface, but I remember reading somewhere that there was a serial card / ROM slot based that provided a serial port - all be it with slow speeds by todays standards - Like 9600.

There was an Asynchronus Comms unit which provides a DB25 interface running at 9600 baud.

[stephen_usher]

Just for the record, here is my Mk.I disk interface board as a KiCAD project.

Note that the pins on the connector to the ROMpack are the wrong way around so the right-angled socket has to go on the "solder side" of the board.

Also, the clock circuit doesn't work so instead of a capacitor and crystal it needs an 8MHz crystal oscillator connected to the 74LS04 side capacitor pad and powered from the large power capacitor pads.

If you want to use a larger EPROM, link pins 1 and 27 of the ROM socket to ground.

[Pr0f]

Trying to understand the circuit around the crystal, it looks like a 1pF capacitor in the feedback loop with the crystal?

[Pr0f]

Actually - I just noticed it's marked all your capacitors as 1pF. I think I get the circuit now, and I think I know why it might not be oscillating - the 2 74LS04's that are wired around the crystal should have some feedback resistors between input and output - this will bias them into the linear mode - that should then create the oscillator circuit you need. The 3rd gate tied to the output of this oscillator circuit will 'square' up the clock and provide a cleaner signal to the WD1772 clock input. Of course all 3 of those gates could be removed and just use a crystal oscillator can.

[Pr0f]

Many thanks for sharing this and the hard work in reverse engineering it.

Do you have the ROM code available?

[stephen_usher]

I copied the crystal circuit from the PCB traces. The capacitor values were not specified on the documentation I had and the cap on the crystal circuit was small and ceramic in the pictures of the board. Anyway, a crystal oscillator is far better, as per my hack.

The design philosophy for this mark 1 board was to replicate the original circuit exactly. I could then improve it in later versions. I can reduce the chip count by at least one, maybe two using a GAL for a start.

[Pr0f]

I was thinking about the possibility of reduction - I think you could get away with:

Crystal oscilator
WD1772
A GAL - possibly even just 16V8 size
An LS240 or equival(for buffering disk drive signals)
An Eprom / Flash
An LS174 / LS175 latch

It's teasing my brain now

[stephen_usher]

Here's the ROM image.

[Pr0f]

Thanks Stephen - I shall download that later and have a look at the code.

I noticed that in circuit - the DDEN pin is grounded on the WD1772, in QL disk interface designs - this is often fed from a latch so the density can be controlled - do you know off hand what density this interace uses - is it 360K or 720K ? I see in the various pictures it's using a 3.5" floppy - so presumably a double density 3.5"?

[stephen_usher]

It's fixed at 720K, as per the Apricot floppy interface.