Commodore 65 and DIN8 cable

From: Michał Pleban <>
Date: Thu, 28 Jan 2016 16:16:59 +0100
Message-ID: <>

I started a project to re-create the 1565 external floppy for the
Commodore 65. The floppy is supposed to be attached to the computer via
a mini-DIN8 cable.

So I purchased a cable and stuck it into the floppy port of the computer
(of course I had to solder a mini-DIN8 connector there first). The
internal floppy stopped working. When I remove the cable, the floppy
works again. Hmm.

I suspect that this happens because the "disk read data" and "disk write
data" signals are exposed directly on the floppy port, and they are
shared among all floppies. So it looks to me that some stray signals are
being induced on these lines in the cable, and they confuse the disk
controller enough to throw disk read errors. This would be similar to
Bil's story about the data bus of C116 being exposed directly on the
joystick port.

To confirm this, I bought a shorter cable. The original cable had 3m,
and the new one has 1m. The situation improved somewhat - now the floppy
works about 20% of the time. With a yet shorter cable I could probably
make it work even better, but that's the shortest cable I found on Ebay,
and besides it's not really a solution I'm looking for.

My extremely limited knowledge of analog electronics tells me that
perhaps I need termination on the end of cable. Is this a good guess?
And if yes, is there any way to find the correct terminating resistor
value apart from soldering a potentiometer between GND and DISK_RD and
patiently trying for the best value?

Also, at some point of course a floppy drive will be attached to the end
of cable - what should happen with the termination then? I suppose it
should be removed but will the floppy drive itself act as a good terminator?

And why this effect does not exhibit itself in PCs and other computers
where the floppy cables also tend to be long (or maybe it does)?


       Message was sent through the cbm-hackers mailing list
Received on 2016-01-28 16:00:10

Archive generated by hypermail 2.2.0.