Whitechapel MG-1 : dead power supply ?

From: Tony Duell <ard_at_p850ug1.demon.co.uk>
Date: Thu Jan 15 16:48:34 2004

> Looking at the block diagram for the PSU (pg. 4-3) I'm guessing shorting

I don't have the offiical service manual to hand -- I forget where I put
it, becuase it's so useless. But I do remember that you have to be
careful with the block diagrams in that manual -- the one for the
mainboard (Stickleback) is plain wrong -- it's the video memory that's
shared between the main processor and the service processor. As an aside,
the Service processor copies a routine from the single EPROM (8 bits
wide) into the video memory, which appears as a 16 bit wide memory area
to the 68K. The 68K then executes this program, which gets the whole
thing going.

> across the switched relay contacts so that it's always on would do it;
> the service processor should start up and the battery should get the
> appropriate charge voltage.
> Presumably power-down via the touch switch should work then too.

You can't power the machine down with the relay contacts shorted -- the
service processor turns off the relay to turn off the PSU. But the
service processor will notice the touch switch operation, and will cause
the 68K to flush buffers, etc.

> Whether the machine will power up normally in this state I don't know;

It will.

> the block diagram implies that it should though as power-up logic for
> the machine as a whole is only handled when the service processor sees
> proper mains-derived regulated voltages from the PSU, not the +5V from a
> healthy battery. (Invalid NVRAM at that point is another matter, but

The battery must not be open-circuit, or the NVRAM will get +12V and will
promptly expire (!). There's a kludge I have for fitting 3 AA primary
cells and a few passives (including a zener diode to clamp that +12V down
to something sane). I find it more reliable than the NiCd (did I mention
that I hate rechargeable batteries?)

Should I add the XXX schematics to the list of things I need to copy for you?

Received on Thu Jan 15 2004 - 16:48:34 GMT

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:36:46 BST