[sdiy] op amp based filter

[cheater cheater]

Another question: what are the favourite externally compensated op amps?
What are the favourite programmable gain op amps with low GBW?

What are the best ways to make a normal op amp's gain programmable
with CV? The only thing I was able to find was that you can have the
resistor in the negative feedback loop replaced with a JFET (and have
it in parallel with another resistor to make sure there's always some
negative feedback at least, i.e. prevent too high gain from
happening). I don't know if that's the best idea, but I hadn't seen
anything better. One problem is with the linearity of this circuit
with regards to the input voltage, the response is mostly Tanh style
it would seem. This would make it desirable as a soft clipper all
right, but not necessarily as a filter with linear frequency control.

Cheers,
D.

On Sun, Jul 25, 2010 at 09:54, JH. <jhaible at debitel.net> wrote:
>
>
>> Sure, check here: http://www.national.com/ds/LM/LM101A.pdf
>
> Oh, I thought there was some extra application note you were referring to.
> The data sheet was the first place I looked. But where is that Eventide LPF?

The only use of the LM301 in your SOST, so I assume that's what you've
taken from the Eventide, is with a couple capacitors between pins 1
and 6. That's on page 7, top right, labeled "two pole compensation".

>> Given that there are two compensation inputs, I wonder if you couldn't
>> even make a 4-pole filter. Just a silly idea.
>
> How many poles do you think the Eventide filter has?

If that's the same thing that you're using in the SOST then it's
two-pole, according to the datasheet.

> That was the whole point of my mentioning it: How steep a response they
> squeezed out of a single-opamp filter, combining two-pole compensation and
> an external circuitry reminiscent of a 3-pole LPF topology.

I hadn't looked at the external filter, I just mean the use of the op amp.

> My PSpice istn't running anymore, and I don't recall if it was closer to
> 4pole or 5pole response, but it sure was impressive.

With a 3-pole filter, you get 2+3=5 ;-) so I can easily believe that
it's steep :)