Date   : Thu, 28 Oct 2010 21:33:13 +0100
From   : list-a_cloud9.bbc-micro@... (Theo Markettos)
Subject: Reversing the Tube ULA (destructively)

In article <2k0jc61uqdia4sfhhminpctqp7b6vviq6j@...> you wrote:
> 
> On Thu, 28 Oct 2010 15:00:00 +0100, Ed Spittles
> <percy.p.person@...> wrote:
> >I hope we'll only need one Tube ULA, but this is the reason I'd like to use
> >a broken one - they are in short enough supply as it is.
> 
> Not a good idea. The reason why a ULA 'broke' may well
> be something that smoked the die to a crisp as well
> (or at least part of it).

I wouldn't worry about that too much.  If it's dead, and it doesn't cost too
much to depackage, then you haven't lost much.  'Dead' is most likely to be
pad-ring stuff - static burning out the I/O drivers, or shorts on the power
supply rails.  The dead chip I was microscope faultfinding on had very clear
pits in the power grid where the short current had gone, but the rest was
quite visible.  That had a removable metal lid to the PGA so I could see the
die.

If it's somehow had a transistor short the power rails in the middle of the
die, the overheating might have cooked that area a bit but the transistor or
nearby metal will have gone pop eventually - the power grid is capable of
delivering a lot more power than a thin metal 1 wire can carry.  It's
possible heating affected a greater area, but it's more likely to be
electromigration effects than melting the surface.

But it's definitely worth getting some scrap contemporary chips (as near a
fab technology as you can get) and practicing on them first.

Theo