A Universal Sallen-Key Filter Board

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: A Universal Sallen-Key Filter Board

Post by JR. »

Gold wrote:
JR. wrote: -6db/oct filters are real poles so have fixed Q. Cascading 3 one pole filters tuned for the same -3dB point would by definition be -9dB at tuning (ASSuming you put buffers between them to eliminate loading).
Or you could think of it as an over damped 3 pole filter. I don't think I've ever heard one. Getting perfect three pole filter response by cascading sections isn't what this box is about. This will be tuned by ear in design and use. I also don't think it will end up exactly like I imagined it would. I have never run across anything remotely like what I am proposing although the building blocks are straight forward. I have wanted this box for a long time. I don't imagine a designer would think of doing this as it would appeal to a very limited few. A lot of people seem to want tone boxes that are easy to use. I hate that. I wouldn't let a pultec clone in the same building.
Perhaps the reason you don't encounter cascaded real poles is because the transfer function is not very desirable, as most beginner filter designers discover after doing that.

For example cascading three real or natural poles in series tuned for the same -3dB @ say 50Hz, they would add to -9dB at tuning. Further it would be -3dB or half-power up at 2x tuning (100Hz). It would even be -0.3 dB at 10x 500 Hz. Not only that but cascading three passive poles without buffers will result in interaction between the pole tuning due to higher source impedance of previous stages.

In this type of specialty design if it sounds good it is good. As long as this is what you actually want, go for it, but there is a reason you probably won't find these filter alignments in professional products.

JR
Cancel the "cancel culture", do not support mob hatred.
Gold
Posts: 678
Joined: Sat Dec 01, 2007 5:20 pm

Re: A Universal Sallen-Key Filter Board

Post by Gold »

JR. wrote: Perhaps the reason you don't encounter cascaded real poles is because the transfer function is not very desirable,
Desirable for what? A textbook response? That's fine. Who cares?
For example cascading three real or natural poles in series tuned for the same -3dB @ say 50Hz, they would add to -9dB at tuning. Further it would be -3dB or half-power up at 2x tuning (100Hz). It would even be -0.3 dB at 10x 500 Hz. Not only that but cascading three passive poles without buffers will result in interaction between the pole tuning due to higher source impedance of previous stages.
That all sounds great to me and is what I expect to happen. We are talking about Sallen-Key not passive filters.
but there is a reason you probably won't find these filter alignments in professional products.
Most mastering engineers aren't designers and most designers are not mastering engineers.

JR[/quote]
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: A Universal Sallen-Key Filter Board

Post by JR. »

Gold wrote:
JR. wrote: Perhaps the reason you don't encounter cascaded real poles is because the transfer function is not very desirable,
Desirable for what? A textbook response? That's fine. Who cares?
No, practical experience gained from designing audio paths for several decades. Cascading multiple real poles results in an undesirable transfer function (for a flat or band-passed audio path).
For example cascading three real or natural poles in series tuned for the same -3dB @ say 50Hz, they would add to -9dB at tuning. Further it would be -3dB or half-power up at 2x tuning (100Hz). It would even be -0.3 dB at 10x 500 Hz. Not only that but cascading three passive poles without buffers will result in interaction between the pole tuning due to higher source impedance of previous stages.
That all sounds great to me and is what I expect to happen. We are talking about Sallen-Key not passive filters.
No you are talking about cascading -6dB/oct filters. AFAIK Sallen and Key are active 2-pole filters (-12dB/oct).

(according to wiki)
The Sallen–Key topology is an electronic filter topology used to implement second-order active filters that is particularly valued for its simplicity.[1] It is a degenerate form of a voltage-controlled voltage-source (VCVS) filter topology. A VCVS filter uses a unity gain voltage amplifier with practically infinite input impedance and zero output impedance to implement a 2-pole low-pass, high-pass, bandpass, bandstop, or allpass response . The unity-gain amplifier allows for very high Q factor and passband gain without the use of inductors. A Sallen–Key filter is a variation on a VCVS filter that uses a unity-gain amplifier (i.e., a pure buffer amplifier with 0 dB gain). It was introduced by R. P. Sallen and E. L. Key of MIT Lincoln Laboratory in 1955.[2]
but there is a reason you probably won't find these filter alignments in professional products.
JR
Most mastering engineers aren't designers and most designers are not mastering engineers.
I try not to argue with people about what "they" want, just trying to be helpful. You can breadboard something up and find out.

If you use 2 pole S&K sections the slopes will be steeper, and they will be self buffered by the op amps involved. Using under damped 2 pole sections that are less than -3dB at tuning might cascade a little better (speculation), but you need to try what you want.

JR

PS: If you want to use a one pole VCVS (Sallen and Key?) you might try one cap and one resistor and a non-inerting op amp stage. Image
Cancel the "cancel culture", do not support mob hatred.
Gold
Posts: 678
Joined: Sat Dec 01, 2007 5:20 pm

Re: A Universal Sallen-Key Filter Board

Post by Gold »

The 1 pole filter in the drawing is what I had in mind. It's the first topology discussed in the Lancaster book. Maybe it's not technically a Sallen-Key but it can be made with PCB's designed for Sallen-Key filters.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: A Universal Sallen-Key Filter Board

Post by JR. »

Gold wrote:The 1 pole filter in the drawing is what I had in mind. It's the first topology discussed in the Lancaster book. Maybe it's not technically a Sallen-Key but it can be made with PCB's designed for Sallen-Key filters.
I am not familiar with the Lancaster book (there are several I suspect).

That one pole VCVS is the real pole buffered by an op amp I mentioned several posts ago.

When cascaded It will suffer the transfer function I described (for better or worse).

If this is what you want, we look forward to your feedback after you build it.

JR
Cancel the "cancel culture", do not support mob hatred.
Gold
Posts: 678
Joined: Sat Dec 01, 2007 5:20 pm

Re: A Universal Sallen-Key Filter Board

Post by Gold »

If I can't scale the resistor value to not load the previous section I could use the next (previous) section on the PCB as a buffer. I think Wayne has taken this into account in the board design.

By having three single pole filters with eleven frequency choices in series you can go from barely audible to 'what happened' quickly. It's kind of like an exponential cut EQ.
Post Reply