A non text book transform based on state variable filters.
Initial fo, Qo and target fp, Qp parameters values can be set independently.
https://jipihorn.files.wordpress.com/2013/06/linkwitz-variable-state.pdf
Initial fo, Qo and target fp, Qp parameters values can be set independently.
https://jipihorn.files.wordpress.com/2013/06/linkwitz-variable-state.pdf
State-variable is indeed a valid solution to transform an arbitrary 2nd order transfer function into desired 2nd order function (within reason of course).
The Linkwitz transform circuit does the same but is restricted in terms of frequencies and Q values. But where it is useable it is definitley the simplest possibility.
Sometimes quite simple approximations can be used that don't need as many OP-AMPs as the state-variable filter shown in the linked document. But that has to be decided case-by case.
Regards
Charles
The Linkwitz transform circuit does the same but is restricted in terms of frequencies and Q values. But where it is useable it is definitley the simplest possibility.
Sometimes quite simple approximations can be used that don't need as many OP-AMPs as the state-variable filter shown in the linked document. But that has to be decided case-by case.
Regards
Charles
Ok, So far I have bought the Douglas Self book and read most of it. I have also downloaded the TI project tutorial and their software.
Here's where I am so far..
I like the TI guide. It seems simple and relatively easy to understand. I also like their software but I worry that it's over simplifying things.
Douglas Self's book is WAY too comprehensive for my purposes. I would appreciate an abridged version that focuses only on best practices and updates the op amp models. I'm also confused because he seems to constantly be making trade-offs for cost. Not everyone is on a budget, and I would really like to better understand the impacts of these decisions.
Here's where I am so far..
I like the TI guide. It seems simple and relatively easy to understand. I also like their software but I worry that it's over simplifying things.
Douglas Self's book is WAY too comprehensive for my purposes. I would appreciate an abridged version that focuses only on best practices and updates the op amp models. I'm also confused because he seems to constantly be making trade-offs for cost. Not everyone is on a budget, and I would really like to better understand the impacts of these decisions.
Just as you learned about passive crossovers and the effects of various component value changes, so you will need to experiment with actives once you have a basic understanding of their circuitry. The filter principles are the same, except one uses the amplifier's output power and driver impedance to shape the filter, the other uses a fixed power supply and fixed resistor impedance to shape the filter.
Of course, there are plenty of "tricks" that can be learned and done with passives, but nowhere that I'm aware of has ever delved into the incorporation of such manipulations into active filters, other than SL.
I suggest you buy a basic active xover kit and begin experimenting. Once you have a decent power supply and basic pcb, the cost of filter components is miniscule compared to passive parts. ESP's P09 is how I learned the most, which isn't much! It's not especially flexible except for filter orders and knees, but it does provide a platform for learning. If you want to implement any of SL's fancy circuitry, you'll need one of his expensive boards or try to breadboard or dead bug the stuff.
Peace,
Tom E
Of course, there are plenty of "tricks" that can be learned and done with passives, but nowhere that I'm aware of has ever delved into the incorporation of such manipulations into active filters, other than SL.
I suggest you buy a basic active xover kit and begin experimenting. Once you have a decent power supply and basic pcb, the cost of filter components is miniscule compared to passive parts. ESP's P09 is how I learned the most, which isn't much! It's not especially flexible except for filter orders and knees, but it does provide a platform for learning. If you want to implement any of SL's fancy circuitry, you'll need one of his expensive boards or try to breadboard or dead bug the stuff.
Peace,
Tom E
I am actually less worried about getting the slopes I'm looking for and more worried about impedance mismatches to specific models of op amps and adding distortion. My Clio system is a far cry from an Audio Precision system that's designed for this kind of stuff. As the components are cheap, I may just place an order at digikey for a variety of components and start playing.
Any of you guys have any experience with using modern op-amps for filtering? What models lend themselves to these applications. I would rather not be using designs that date back to the 70's. TI used the 1604, but stated that this was a cost decision. My considerations will be ease of use and sound quality. Cost isn't a concern as I'm not putting this into production.... I would just use the book, but it seems that a TON of new models have come out since. Also , I have SMD equipment so it doesn't need to be available in DIP.
Lastly, it seems that everything is SMD now. Is there a reason to use old large poly caps, and through hole resistors, instead of modern SMD stuff?
Any of you guys have any experience with using modern op-amps for filtering? What models lend themselves to these applications. I would rather not be using designs that date back to the 70's. TI used the 1604, but stated that this was a cost decision. My considerations will be ease of use and sound quality. Cost isn't a concern as I'm not putting this into production.... I would just use the book, but it seems that a TON of new models have come out since. Also , I have SMD equipment so it doesn't need to be available in DIP.
Lastly, it seems that everything is SMD now. Is there a reason to use old large poly caps, and through hole resistors, instead of modern SMD stuff?
Last edited:
I would say that, in something as simple as a filter, the advantages of SMD are few, while it's always nice to use those big TH poly caps. I would expect a "passive" guy to appreciate that. I would worry more about layout, PS bypassing, and preventing oscillation of high speed opamps than I would about impedances.
Regarding opamps, I have found that discrete units sound noticeably better than the IC's, at least in my system. I've used OPA2134, which seemed so smooth on top as to be dull but had full bass; LME49720, which brought out plenty of detail to the point of being artificial and a bit fatiguing but did present a more spacious sound; OPA627, which struck an almost perfect balance tonally between those two but still lacked some depth and air; and finally a Sparkos Labs discrete, which possesses all the good qualities of the 627 but also provides a realistic layering of the sound to form a truly three dimensional acoustic source hanging between and around the speakers. Expensive, but I use only one in my xovers.
Some builders prefer simple discrete buffers over opamps, and that makes sense, but I've never been able to figure out which parts and circuit are suitable, and I'm not aware of any available platform for construction. Details given here are always somewhat sketchy, as though the posters do not want to divulge too much, or perhaps they assume we know a lot more.
Peace,
Tom E
Regarding opamps, I have found that discrete units sound noticeably better than the IC's, at least in my system. I've used OPA2134, which seemed so smooth on top as to be dull but had full bass; LME49720, which brought out plenty of detail to the point of being artificial and a bit fatiguing but did present a more spacious sound; OPA627, which struck an almost perfect balance tonally between those two but still lacked some depth and air; and finally a Sparkos Labs discrete, which possesses all the good qualities of the 627 but also provides a realistic layering of the sound to form a truly three dimensional acoustic source hanging between and around the speakers. Expensive, but I use only one in my xovers.
Some builders prefer simple discrete buffers over opamps, and that makes sense, but I've never been able to figure out which parts and circuit are suitable, and I'm not aware of any available platform for construction. Details given here are always somewhat sketchy, as though the posters do not want to divulge too much, or perhaps they assume we know a lot more.
Peace,
Tom E
For unity gain current buffer I like ST Micro 33078/79 opamp. TI 4562 and derivatives sound Brystonesque (closed in, suffocating) when used in unity gain, but they open up wonderfully with 10dB of gain. Implementation is everything. 5532 is good. Don't worry about old vs new, for this application any good opamp will work fine, even the "old ones." You might find the old ones, like 5532 were designed specifically for audio by superb analog enegineers who knew how to make music sound good. Back in the days when music was worth investing in and Sansui and Pioneer needed ICs that sounded very good.
I use Vishay MKT-1813 caps and Dale CMF55 resistors for active crossovers. The sound and price is very good and small size.
Rod Elliot has written a lot of articles about active filters and crossovers. He perked my interest in it long ago. Worth reading his educational articles.
I use Vishay MKT-1813 caps and Dale CMF55 resistors for active crossovers. The sound and price is very good and small size.
Rod Elliot has written a lot of articles about active filters and crossovers. He perked my interest in it long ago. Worth reading his educational articles.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.