Good news on my RK05 drives

From: Tony Duell <ard_at_p850ug1.demon.co.uk>
Date: Wed Jul 28 18:17:09 2004

> I wrote a little test program to attempt to read random blocks and I
> watched the read/write head. It did move in and out like it should.
> If it was attempting to read a low numbered block, it moved just a
> little bit. If it was attempting to read a high numbered block, it
> moved farther toward the center of the platter. However, every

That sounds right. There should be a vernier scale on the positioner so
you can check it's going to the correct cylinder, but at this point I'd
assume it was not the problem.

> attempt to read or write resulted in an error. The head would
> position to the appropriate(?) position on the disk, sit there a
> short while, and then move back to cylinder 0.

This is presumably due to your program, or the RSTS driver. The hardware
should not do this. Can you execute seek commands from the front panel?
If so, I assume the head goes to a cylinder and stays there.

Can you execute a READ command from the panel? (You may have to do this
from a toggled-in program on the 11/34, as I think the processor doesn't
correclty handle NPR's when halted). If so, then look at the error
register to see what it thinks is wrong.

>
> I suppose I could swap boards with the good drive to verify the
> boards.

Or you could learn to fix it properly :-)

OK, it sounds like the spindle motor is fine, the positioner is fine. The
problem is probably on the Read/Write board (a less likely possibility is
on the tiny bit of logic between this board and the drive cable). COuld
also be the heads.

Start by inspecting the heads, If they look OK, I'd start by loading a
pack with something on it and then looking at the output of the read
circuit with a 'scope. Are you getting anything sane there?

>
> Any other ideas? I guess that some day soon I'm going to have
> to learn more low-level technical stuff and how to troubleshoot
> electronic components (I keep saying that!).

Indeed you should. It's not that difficult, and you have picked a good
machine to learn on. It's complicated enough to be worthwhile, but the
chips are all fairly simple, they're all documented (no big custom gate
arrays), and schematics/technical manuals exist.

-tony
Received on Wed Jul 28 2004 - 18:17:09 BST