Re: [N8VEM: 3912] Re: [!! SPAM] [N8VEM: 3908] Use of bus termination cards?

[Michael Haardt <mic...@moria.de>]

> In the S-100 bus realm we used an active termination - I believe it was 
> 470 ohm resistors to an approximately 2.0 volt stiff DC source.  
> Termination only on one end of the board.  This forces all boards to 
> actively drive both ones and zeros but did match the impedance of the 20 
> inch long S-100 bus lines.  It really cleaned up the signals especially 
> with the higher speed boards.

If the termination card is at one end, what's at the other?

The other thing I wonder about is the "stiff" DC source.  Doesn't it have
to be a very fast regulator for being able to improve things?  The design
I am looking at uses an opamp and two complimentary darlington transistors
with the second transistors being BD139/BD140 - low frequency transistors.
Five 10uF and four 10nF capacitors are supposed to help there and the
layout shows the autor was worried about this.

I am aware of the fact that active termination usually draws less power,
but that's just "usually".  I wouldn't want things to get unstable if
I have extreme bus signal patterns.

The design uses a voltage of 2.6V, connected with 270 Ohm to all bus
lines.  Poor bus drivers, that almost pushes them to the limit.

Finally: I believe these cards do help, but did anybody ever look at the
signals? Could it be that this is called termination, but in fact only
improves signals to a certain extent, but does not really terminate the
backplane ends?

Michael