From: Sar Saloth (sarsaloth_at_yahoo.com)
Date: 2002-03-09 00:07:26
Hi Joerg,
OK, either I think you are agreeing or you have stumped me. I am impressed
that you went and did it, and also I have to give LTSPice a shot, but I am
not sure what to say or what you are saying or implying.
Don't you think that the output of the 2nd circuit looks much more stable
and also less likely to be perturbed by noise coming out of the A/D
converter? Don't you think that is worth the addition of few passives? ???
In as much as the simulation doesn't show continued oscillation, with the
original part, there might be a few points.
1. The problem I saw was with the L version of the part, I don't know if
that really should account for the difference.
2. I think that the models wouldn't necessarily show the proper behaviour
of capacitive loading of the output (for the older part that is), possibly
even in the small signal case. I am almost certain it wouldn't correctly
model when the current draw is high. Remember, there is a trade-off in
models for execution speed and accuracy.
3. I will repeat the test in real life (not simulation) next week (or the
week after, whenever I get my scope back).
4. Maybe I was wrong.
5. However, I don't like even severely under-damped responses as they seem
rather closer to unstable than I am comfortable with. Didn't your
simulation show an under-damped response?
<snip>
Looks like figure 14.25.c on page 946 "The Art of Electronics", Horowitz,
Hill, second edition, 1999, except for the component values.
Is that a recommended text? Or am I in trouble for plagiarizing the
circuit without giving due reference? Since you have the book, does it say
how to find the optimum values?
Great weekend everyone - I am not going to write anything more until Monday.
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com
This archive was generated by hypermail 2.1.4 : 2002-07-27 12:28:40 BST