Sounds like the PIC got into latch-up somehow, and the Ni-Cad's low 
     impedance help the PIC into oblivion. I've no idea what impedance a 
     latched up PIC has, but it must be sub-ohm. What size Nicads were 
     they? 
     
        Steve Lawther


______________________________ Reply Separator _________________________________
Subject:      Re: Getting a PIC really HOT!! or What will make a PIC 
Author:  MIME:anick@IVERSOFT.COM at INTERNET-HUSKY
Date:    21/01/98 08:26


Well The PIC in this case go so hot that the case is missing around the 
die area. You can actually see the the die or whats left of it.
I forgot to take the board to work so I could get a picture of it. I'll 
try and do it tomorrow.
     
     
Richterkessing, Frank H (GEA, 055132) wrote: 
>
> Alan Nickerson Wrote:
>
> > I received one of our units back from a customer of ours abd the pic 
> > was
> > completly destroyed. The location where the die is/was was all burned 
> > and such. It is suspected of being put in a mircowave oven.
> >
> To give you an alternate view on that failure mode, I have an old and
> used digital panel meter from a batch that I picked up at a hamfest some 
> years back. There's a 7474 on the front PCB with the LEDs. As I was
> testing them out I noticed that this one would work fine for about five 
> minutes or so and then call it quits. The die area of the 7474 gets so 
> hot that it actually blows a bubble on the plastic package! Let it cool
> down and it will work again. It seems odd that it would work at all with 
> some sort of defect that would get the chip that hot. Too much trouble
> to get the thing apart to find the problem. I haven't blown a bubble 
> with a PIC, but I have had one (due to an oops of a wiring error) get 
> hot enough that I managed to burn my finger on it (badly). After I
> corrected my oops and it cooled down, I continued to use it as my 
> development PIC without any problems.
>
> -Frank
>
> Frank Richterkessing
>
> FRANK.RICHTERKESSING@APPL.GE.COM