My evil modular plans

[Tony Clark]

> The point is, grounds aren't a "given."

   Well poor design and considerations on the part of telephone companies 
aside...

> >  The more important aspect of doing this 
> >particular chore would be to have a clean power source....
> 
> Neither is more important. A clean power source does one NO good if
> there's several uA of noise current flowing through a ground trace in
> the wrong area. There's little one can do about this in a board that's
> already built, but it's crucial if one is designing a new card.

   Well if there IS noise on the ground trace then you ARE screwed from 
the get go.  I'll agree with that for sure.  That certainly is something 
that I do a lot of testing on with new products.  Plug it into a "clean" 
power source then plug it into the computer.  See how the product rates 
from best case scenario to worst case scenario.  Then adjust circuitry to 
minimize the effects of the worst case.
   It isn't that hard to produce clean analog signals from crappy 
computers.  My company's new product puts out less than 5mV of noise 
even after amplifying super small signals with gains in excess of 1000.  
It's unfortunate though that when the signal returns to the computer that 
the A/D converters add a pile of noise to it.
   To be honest, I wouldn't trust any sort of "hi-fi" circuitry "inside" 
my computer.  I think Paul is better off abandoning the PC-Card scenario and 
using an external box. 

   Tony

-------------------------------
I can't drive (my Moog) 55!
-------------------------------
Tony Clark -- clark at andrews.edu 
http://www2.andrews.edu/~clark
-------------------------------