----- Original Message -----
From: "Tony Duell" <ard_at_p850ug1.demon.co.uk>
To: <classiccmp_at_classiccmp.org>
Sent: Sunday, February 17, 2002 5:54 AM
Subject: Re: Destructive charities (was: STAPLES STORES...)
> > Indeed. I am in the process of making a pc controlled pan/tilt head for
a
> > webcam using bits gleaned entirely from 5.25" drives.
>
> Indeed. All the more reason that old computers shouldn't just be treated
> as scrap metal. There are useful parts in there.
Yes, either for experimentation/reuse or to fix other machines.
> > By using the stepper controller and some of the sensors that already
exist
> > in the disk electronics it can interface directly to a parallel port
that I
>
> This is quite easy to do on older drives, but more modern (particularly
> 3.5") ones have all sorts of 'interlocks'. The head moves back and forth
> at power on to check the track 0 sensor is doing the right thing and so
> on. If it doesn't then the drive won't respond to interface signals.
The older Chinons aren't too bad, though some of the control lines do not
function like the old tandons.
The NMOTORON which is supposed to toggle the spindle motor does nada, there
is a IR photo interrupter that
starts the motor when a disk is inserted. Can't get a pulse from Write
Protect or Track 0 at the interface, though oddball things happen if I
remove the sensors. Got around NMOTORON by removing the IR PI and tieing
the line to the pin on the port directly. The Track 0 sensor upsets the
stepper if you remove it, basically makes it stop after a half turn. Using
the WP sensor which seems to work with visible (fluoro) or the IR led
source, to pull down an input on the port, though still tinkering with
getting the computer to recognise the toggled bit. Using the input side,
(Hex 0379) but getting no response, possibly the cheapskate port on the
diskless box is not bidirectional, so might try the status register instead.
> So you may not be able to use much of the control electronics. It depends
> on the drive. If there's a microcontroller surrounded by buffer chips
> (say mid 1980's designs) then you can probably make use of (at least) the
> buffers and surrounding components. If there's on ASIC then the control
> board may not be much use to you.
Yes, I have taken apart a 3.5" drive that does a head seek and stops. The
method I'm using with the chinons is to leave any sensors that cause trouble
where they are, and tie others directly to the proper pins.
> > control with software written in my (classic? must be close.) 'old but
good'
> > Turbo Pascal 6 for DOS.
>
> I think it is a classic. I got my first PC in 1991 (a true-blue XT that's
> still in use). I am pretty sure TP v 6 was available then.
I have a complete XT, keyboard and monitor (true blue) in my collection too.
I started playing with tp around V5 around 1990.
> > By judicious use of the drive select lines, it's possible to control
several
> > 'drives' from the one port so a pan and tilt plus (say) focus or
something
>
> Let's see. You need step and direction. 2 signals. Maybe a 'track 0'
> (home position) sensor input to the PC (you can feed that in on the busy
> line or something).
Yup.
> That's 2 bits, so you could use the other 6 data
> lines to select 6 'drives' with no extra electronics. Add a couple of TTL
> chips ('138s are useful) and you could go beyond that.
yep no big deal at all. Numerous possibilities.
> > similar is possible. Got plenty of half dead 5" drives, prefer the
older
> > Chinons as they have huskier stepper motors with conventional shafts,
the
>
> I'd probably go for the old Tandons (mainly because I have a box of dead
> ones...).
I'm looking but yet to find any, they are far easier to interface to.
> Nice big stepper motors, the electronics is all simple TTL, etc
> (so there are no 'features' that prevent you using just the stepper
> drivers),
Yes exactly.
> and the schematics are easy to get.
Oh? Now that would be a help, where from?
> Obviously I prefer to repair such drives when possible (I do still _use_
> 5.25" disks here),
Me too, but have plenty of spares that are beaten up, broken heads etc.
> One of my printers (a Sanders 12/7) has a stepper motor to drive the
> platten (direct drive -- the platten is roll-pinned to the motor shaft)
> and another one in the optional tractor feed unit. Both are wired in
> parallel to the same driver circuit If you turn the tractors with the
> machine switched off, the stepper motor generates enough voltage to deive
> the platten motor
yeah sounds like a relative. Big sucker.
> > it I can get better than 1 degree resolution. Will need to build an
> > appropriate driver board, it would toast a floppy drive stepper
controller.
>
> Did you save any of the printer's electronics? At least the output stages
> should be useable.
Got the entire board. Including all the pin and stepper drive transistors
and the psu.
(Print head was stuffed in it)
> > from the 60w psu. Powering it from a separate supply creates earth
looping
> > issues that confuse the hell out of the logic by lifting the pulled down
>
> You have linked the -ve rail of your external PSU to the ground on the
> computer's parallel port I hope...
Yes, if I hadn't that would explain it, but tried it with/without and
several variants inbetween, still getting .4v pull up that stops the logic
recognising a low. Working around it at the moment, but will come back to
it when I have the interface and control stuff working properly. I have a
theory that it may actually be due to a minor fault in the drive
electronics, so I will test that later in the week with a substitute drive
board. I have compiled a test program that spins the motor both ways,
toggles the spindle motor on or off, and another section that lets me toggle
the direction of the stepper and alter the speed by varying the delay
between the step pulses.
Cheers
Geoff in Oz
Received on Sun Feb 17 2002 - 02:06:11 GMT