Re: gEDA-user: opamp slew rate limiting

[Martin Maney <maney@xxxxxxxxx>]

On Tue, Nov 10, 2009 at 06:08:41AM -0500, gene glick wrote:
> SR definition is SR = 2 * pi * f * Vpk

Nope.  That's the maximum *signal* rate of change for a sinewave.  I
think you know that, but since there seems to be a lot of vague
misunderstanding about slew rate in this thread...

> So I need SR > 6.28 * 5000 cycles/sec * 15 Volts, or 471,000 
> Volts/Second.  If my math is right, that works out to 0.471 V/uS. Lots 
> of margin there too, the part can do 20.

That's about how it looks to me, too.  15V peak would be about 6db
overdrive (IIRC that the opamp is on +- 7.5 volt supplies).

> I agree about the linear/saturation description, but couldn't find any 
> hard literature on the subject.

If that refers to the comment I think it does, then it's half wrong. 
SR is the symptom of NONlinear operation: it's the output rate of
change observed when the input stage is saturated by an input signal
that changes so quickly that it outruns the feedback.  The saturated
stage is usually then charging or discharging an internal capacitance
with a more or less constant current, hence the relatively straight and
predictable (but *not* always symmetrical) output slopes.

Clipping is normally due to output stage saturation, and yes, there can
be a significant recovery time due to (once again) earlier stages
having gone into saturation trying to push the output to the
unreachable level that would satisfy the feedback loop.  There can be
some recovery after slew limiting as well, though that's rarely as
severe and is often insignificant.

> great point.  Maybe I should increase the gain to force the saturation 
> earlier into the cycle.  I was trying just the opposite.

Nah.  You should use a good comparator with controlled hysteresis.  An
opamp, any opamp, makes at best a mediocre comparator.  If phase jitter
is important, as I believe you said it was, then you should be
targeting a point very near the zero crossing for that critical
transition, as that is where the sinusoidal signal has its maximum rate
of change, and hence whatever noise is in the system will cause the
least jitter.

-- 
I didn't write a whole, free operating system, either.  I wrote some pieces
and invited other people to join me by writing other pieces.  So I set an
example.  I said, "I'm going in this direction.  Join me and we'll get
there." And enough people joined in that we got there.  -- R M Stallman



_______________________________________________
geda-user mailing list
geda-user@xxxxxxxxxxxxxx
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user