Prev: PROM for Spartan 6 FPGA
Next: Ring Oscillator -> counter differences
From: Jonathan Bromley on 26 Mar 2010 16:45
On Thu, 25 Mar 2010 18:41:06 +0000 (UTC), glen herrmannsfeldt wrote:

>> Why do we need power planes? Are we trying to keep the "reference
>> rails" common between chips to a very high degree like we do with our
>> ground planes?
>
>Some drivers can pull up almost as much as down.

True, for sure. But in the olden days (and remember, I'm an
olden-days person) ground, but never power, was used as a
reference for inputs. Is that still true? My gut tells me
that CMOS inputs need both power and ground to be well-
behaved if they are to act sensibly. Does anyone know
different?
--
Jonathan Bromley
From: glen herrmannsfeldt on 26 Mar 2010 17:18
Jonathan Bromley <jonathan.bromley(a)mycompany.com> wrote:
(snip, I wrote)

>>Some drivers can pull up almost as much as down.

> True, for sure. But in the olden days (and remember, I'm an
> olden-days person) ground, but never power, was used as a
> reference for inputs. Is that still true? My gut tells me
> that CMOS inputs need both power and ground to be well-
> behaved if they are to act sensibly. Does anyone know
> different?

TTL is pretty much ground referenced, where the switch
point is related to forward bias on PN junctions.

The first CMOS had the switch point at (Vss+Vdd)/2,
symmetric as you would expect. Later, to be compatible
with TTL, CMOS circuits were changed such that the
range matched the TTL (0.8V and 2.0V) points.
It seems that might make it more sensitive to ground
bounce than Vdd bounce, but it isn't so obvious to me.

-- glen
First  |  Prev  | 
Pages: 1 2 3 4 5 6
Prev: PROM for Spartan 6 FPGA
Next: Ring Oscillator -> counter differences