Hello again,
because you have indicated your interest: As outlined in my answer #13, we always have two different alternatives for a 4th-order lowpass (example):
(a) cascade design (based on a series of 2nd.order stages)
(b) "direct" design (mostly based on active realisation of passive RLC topologies).

Al always in our live, the two alternatives have some advantages and some disadvantages.
* Alternative (a) is easy to calculate and to fine tune - that means: Easy to design. But it is rather sensitive to parts tolerances
* Alternative (b) has opposite properties (involved calculations; better: Tabulated normalized parts values) but it is much less sensible to parts tolerances.

Examples for (b): FDNR- technique, Leapfrog topology, FLF (Follow-the Leader-Feedback), PRB (Primary-Resonator-Block).

Comment 1: These methods (in particular: Leapfrog) are the starting point for inegrated Switched-Capacitor (SC) realizations.

Comment 2: The filter design programm from NuhertzTechnologies (www.filter-solutions.com) allows - amomg many other design alternatives - also a design based on "leaprog"-structures.

Hi again,

Thanks again for the clarification.

I have to say that i thought we were talking analog only, and i also did not realize that you wanted to include inductors. If we include inductors in the design it is clear that we can produce probably any order we want without using a single op amp.

Let me just point out a few i have worked with in the past.
First, the highest order was probably 7 or 8, one was multi LC the other i remember was strickly RC with amplification stage.
I've done 4th order with two op amps, but one i remember best was only 2nd order BUT it HAD to use two op amps stages because we were using it as a sort of analog computer, where it would be used on the output of a synthesized sine converter to simulate the output leakage inductance and filter cap and load resistor. The idea was to create a design that allowed the change of just ONE resistor in the filter that would simulate changing the output load resistance and thus we could measure the distortion with different loads, without actually connecting different loads. This was still kind of a theoretical idea with the assumption of linearity, but the main point was that it could not be done with a single op amp due to the requirement of being able to change just ONE resistor and get the required filter response change.
That's a special case yes, but illustrates how much things can vary in the real world of engineering.

Anyway back to the op amp and resistors and capacitors. I think for this type we always need two op amps for a sharp response unless we do include inductors which in many cases we are trying to avoid.

 

Hi again,

I am afraid there was a misunderstanding between us. Of course, I do not recommend to use (wired) inductors.
When I speak about "active realisations" of passive RLC structures, I refer of course to (a) active inductor realizations or (b) FDNR blocks.
(FDNR=FrequencyDependent Negative Resistors).
In both cases, we make use of the versatile GIC-units (Generalized Impedance Converter).
As a result - a passive ladder structure is transferred into an active chain containing opamps, resistors and capacitors only.
It was shown elsewhere that GIC based filter circuits have the best properties as far as active sensitivity is concerned (influence of non-ideal opamp data).

By the way: In my examples for "direct realizations" I forgot to mention that GIC-based circuits - in particular FDNR-techiques - are very popular in the world of active filters of higher order.

 

Hi again,

I am afraid there was a misunderstanding between us. Of course, I do not recommend to use (wired) inductors.
When I speak about "active realisations" of passive RLC structures, I refer of course to (a) active inductor realizations or (b) FDNR blocks.
(FDNR=FrequencyDependent Negative Resistors).
In both cases, we make use of the versatile GIC-units (Generalized Impedance Converter).
As a result - a passive ladder structure is transferred into an active chain containing opamps, resistors and capacitors only.
It was shown elsewhere that GIC based filter circuits have the best properties as far as active sensitivity is concerned (influence of non-ideal opamp data).

By the way: In my examples for "direct realizations" I forgot to mention that GIC-based circuits - in particular FDNR-techiques - are very popular in the world of active filters of higher order.

Hi,

Doesnt that take another op amp though?
No big deal though, thanks for elaborating.

 

Hello,

Since there has been no responce from the TS, I will close this thread.

Bertus