I'm looking for information about differential all-pass and notch filters. To me it seems like it's a very straight forward process to convert the single-ended versions to differential, but I don't want to make assumptions. The few circuits I have found were either non-nonsensical (to me at least) and others were balanced (ground referenced), not differential. I'd save myself the trouble and use single-ended circuits, however, the circuits are quite extensive. This includes an 8ᵀᴴ order active crossover(s) with over 300 parts, so along with all the other circuits, there is a significant signal path(s) length (I already have a headache thinking about all the resistors and capacitors that I'll have to match to tight tolerances).
Besides, the advantages of differential circuits such as harmonic cancellation and noise rejection make differential circuits too attractive to ignore!
Considering the topology of these filters, it appears I should be able to simply convert them to differential. The single-ended circuits have no parts going to ground (other than the op-amp's + input of course) and so even the part values would remain the same. Any experience, insight, or (helpful) comments would be great!
Thanks.
Besides, the advantages of differential circuits such as harmonic cancellation and noise rejection make differential circuits too attractive to ignore!
Considering the topology of these filters, it appears I should be able to simply convert them to differential. The single-ended circuits have no parts going to ground (other than the op-amp's + input of course) and so even the part values would remain the same. Any experience, insight, or (helpful) comments would be great!
Thanks.
Active. It's pretty much impossible to make a passive differential filter. Balanced yes, but there's not much point to that.
Handbook of active filters, D.E. Johnson, J.R. Johnson, H.P. Moore, 1980, Prentice-Hall.
I own a copy in my local language (ISBN 9062150586) which I presume is hard to transcribe through this platform to suit your needs. Everything is incorporated already. Best part is inverse Chebyshev filtering: lovely!
I own a copy in my local language (ISBN 9062150586) which I presume is hard to transcribe through this platform to suit your needs. Everything is incorporated already. Best part is inverse Chebyshev filtering: lovely!
I have a couple of books about filters and all of them are for single-ended. I converted a number of types of low, high, and band-pass to differential successfully and used them in active cross-overs. The main difference with both the all-pass and notch filters is the summing amplifiers. I already designed the filters I need and removed the summing amplifiers and used Vocm instead. One less op-amp is always a bonus! I'm going to make up a board and see how that works out.
Thanks for the book reference, I'll see if I can have a look at it before buying it, it's a bit on the expensive side (even though I could only find a used one so far).
Expensive but in use for almost 40 years now. True, almost no filters applied to differential circuits. And summing amplifiers cannot really deployed into these. X-crossing the feedbacks might work, but more networks are needed for the cross references. Post you idea's and sketches, and I'll ponder about the issue more directed.
The point would be to keep signal currents outside of the ground connection, A balanced filter driven with a push pull signal will send no unbalanced signal current to ground.
I found a used copy at a decent price and in pretty good condition considering it went from a colledge in England and then to a library in Germany (and now to me in Canada). Thanks for the book reference, I'm loving it!