Whitechapel MG-1 workstation questions

From: Tony Duell <ard_at_p850ug1.demon.co.uk>
Date: Sat Apr 24 16:34:15 2004

>
>
> A few initial questions after a couple of these beasties turned up at
> the museum:

I have one. I also have schematics, technical notes, and so on.

>
> Does anyone have OS install media for these? We've got the manuals, but

Yes, but on 3.5" (720K) disks. You can put a 3.5" drive in place of the
standard 5.25" 80 cylinder one, at least temporarily, to do an OS install.

> no floppies and I'm not sure what state the hard drives are in yet.
>
> Don't suppose anybody has schematics / service information?

Yes I do, inclduing schematics of the bits that weren't in the oficial
manuals...


>
> Predictably, the batteries inside the machines are toast and have taken
> half the circuitry with them (grr!). I'll clean everything up and then

Well known problem. The 0.1" header connector between the power
distribution board and the mainboard often suffers, BTW. Clean things up
with a dilute solution of citric acid, the battery electrolyte is alkaline.

The power distribution PCB is drilled to take the better-grade of AA cell
holders that RS components sell (sold?). Fit those and pop in normal
non-tagged AA NiCds. Easier to replace next time you need to do it.

> bypass the tracks which have been damaged / eaten away. Presumably
> there's a trick to starting these things after battery failure by
> feeding power straight to the internal relay - any ideas what voltage it

YEs, you connect a 9V PP3 battery to a connector on the power
distribution board. I can look up the exact details.

> needs though? And once running will I still need to keep the relay
> energised or will the PSU circuitry take over (even in the absence of
> batteries - I'm just going to remove the damn things completely)?

I think you need the batteries to act as a shunt regulator for the RTC
chip. You certainly do on a Torch XXX.

Once you've jump-started the machine, it will stay running even if the
NiCds are totally flat. You can disconnect the 9V battery once it's going.

>
>
> I've got one hard drive to spin up and become ready after dumping half a
> can of WD40 onto the bottom spindle bearing - it wouldn't even turn

WD40 is not really a lubricant!

> before that. Now it just sounds like a sick cat. :-/

Hmmm.. I prefer healthy cats...

> Hopefully it'll last long enough to get any useful data off it though.
>
> I notice what seems to be a SCSI connector on the system board - can I

No it's not. It's a 50 pin header that's connected to some of the signals
on the hard disk controller chip (NEC 7261 or something like that -- I
know I have some data on it).

It was going to be used for an SMD interface. Pull out some of the chips
on the mainboard (ST506 buffers, data separator, etc), connect an adapter
board to this connector. Problem was, the DMA controller couldn't keep up
with the SMD data rate, so the adapter was never produced.

Mind you this is the machine where early models have a graphics
'processor' that's bit-serial and actually slower than doing the
operations using the 32016. That's why later machines have lots of empty
IC locations on the mainboard.

> pull out the ST506 disks and just run a more modern (and hence reliable)
> SCSI disk from here? Or is the SCSI connector (if that's what it even
> is!) just designed to support a tape drive, and the machine always
> expects the boot drive to be an ST506 disk?

If you have the IBM slot adapter (which adds 3 8-bit ISA slots), then
presumanly you could put a tape controller in one of those slots and
write software to talk to it. I've never heard of it being done, and the
IBM slot adapters are not exactly common.

-tony
Received on Sat Apr 24 2004 - 16:34:15 BST

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:36:30 BST