Compaq Portable II - Restoration Problem, Help requested

Tony Duell ard at p850ug1.demon.co.uk
Sun Jan 3 16:25:07 CST 2010


> 
> On 3 Jan 2010 at 17:46, Tony Duell wrote:
> 
> > As an aside, I have a machine using standed '360K' 5.25" drives --
> > that is Tandon TM100s -- which does correclty implent a disk-changed
> > feature. How it does it is to have the 2 drives separately cabled to
> > the controller board and the drives always selected. It then looks for
> > a change-of-state of the write-protect line. This must change if a
> > disk is inserted or removed.
> > 
> > Anyone want to guess the machine? I have 3 of them, all different
> > models (one only has on disk drive).
> 
> Heh, nihil sub sole novum.
> 
> I implemented that on the Durango F85 in 1979, but I don't think 
> that's the machine you mean.  We had a terrible problem with 

No it's not. 

> customers simply changing diskettes whenever they wanted.  The drives 
> were cabled normally, just selected and the status of the write-
> protect line checked every few hundred milliseconds.  The FDC was a 
> WD1781, so WP status was available all of the time, regardless of the 
> state of the motor.  The drives were TM-1004Ms or Micropolis 1610s.

The machine I am thinking of uses a WD1793 or similar. The write-protect 
checking is done in hardware -- as I said both drives have separate 
cables back to the contrtoller PCB. The MX jumpers are fitted on the 
drives, causing them to drive the cable all the time, so the WP signal is 
always valid from the drive. There's a littlebit of locking involving a 
flip-flop,. etc that detects if the WP line changes state. (I would have 
to grab the schematics, it may be it only detects if the WP line goes to 
the asseted state, not the revers trasnition. That's certain to occur 
when a disk is inserted anyway).

> One problem we never did solve was that of the customer powering the 
> machine down, without exiting an application.  But that persisted on 
> all systems until someone invented the "soft power" button.

YEs. Another of my machines (an early form of 'unix for the masses' -- it 
has a graphical front end over Uniplus+, but you tend to end up in the 
shell anyway -- has a touch switch to turn the power on and off For power 
down, the service processor detects the touch on the switch and then 
tells the main processor to flush the caches, etc before turning off the 
main power relay.

The identity of the machines/ The first (WP used for disk change) is the 
HP9826 or HP9836, the latter is the Torch XXX (I think the Whtiechapel 
MG1 does something similar too).

-tony



More information about the cctalk mailing list