Looking to build a 2-way to work with my existing active analog crossover

[JohnnyNG]

I realize I may be going about this bass-ackwards.

I've got a Sublime Acoustic K231 Stereo 3-Way Active Crossover which I'm using as a 2-way for sub/mains. Since it's a 3-way crossover, however, I was toying with the idea of building a 2-way which would work well with this basic kind of crossover. By "basic" I mean it's got Linkwitz-Riley filters, 4th order 24dB/Octave or optional 2nd order 12dB/Octave filter slope as well as Baffle Step Compensation, with adjustable gain and selectable baffle size ranges but can't do notch filters, etc.

So I'm thinking I'd need well-behaved drivers and am planning a sealed enclosure since I'd be high-passing to a sub anyway.

Any thoughts? I've been dreaming of a Purifi build but not sure how feasible that is. Maybe the Fiber Cone PTT6.5X04-NFA-01 6.5"? Seems not to demand the notch filter recommended for the Alu cone variant. Madisound suggests ".20 -.30 cf sealed for an F3 of 87Hz, F6 of 58Hz, F10 of 40Hz".

I'm coming from Revel M106 with which I have no complaints whatsoever so not sure why I'm bothering.

 

[Wolf11Man]

The first question should be: Do you have measuring gear. If no, first get a measuring microphone and learn how to use such a program. REW is free and one of the most common used. ARTA is a more capable, but harder to learn, free app.
If you want to buy a microphone, there are calibrated (better to use such a product) with USB connection. These are fine for a fast, plug&play measurement as you only need a laptop, but can not be used for time sensitive jobs without some complicated extra.
If you want a "pro like" hardware, get an XLR powered, calibrated microphone like the Sonarworks SoundID Reference Measurement Microphone and a matching USB interface with 48Volt phantom power.

 

[JohnnyNG]

The first question should be: Do you have measuring gear.

Thanks very much! Yes, I’ve got REW and a Umik-2.

 

[Wolf11Man]

You will run in some problems in the midrange, because your crossover has no option to correct the baffle step. When a certain frequency is reached, a driver doesn't distibute its energy (=sound pressure) all around the cabinet, but starts to beam it in one direction. So sitting in front of your speaker, you got a quite sudden increase in sound pressure.
The most simple way is to measure this and apply some kind of equalisation to it. A DSP based x-over does this with 3 mouse clicks.
With your analog x-over this is a little more complicated. You may use some passive components in the mid channel.
Another way is to get x-over frequency and baffle design in such a relation that it doesn't disturb the response.
This is the point where speaker building gets interesting. There are quite some simulations that predict the response, but they are only as exact as the data you feed them. Using factory data may be far off, so measuring your individual driver is much better. If you would like to play with baffles, my advice is to use XPS panels to build a dummy and measure it before you finalise your build and cut expensive wood.

 

[JohnnyNG]

Thanks again, I've got some work to do!

 

[Salt]

If the data of the crossover given are correct, you can choose between 24 and 12 bB. That's not bad and can, if drivers support it, be enough to align the drivers to a correct output.
Trial and Correct, as alway in speaker design

 

[a4eaudio]

You will run in some problems in the midrange, because your crossover has no option to correct the baffle step. .

It only has four knobs and one is literally labeled Baffle Step Compensation. I have no idea how good it works, but the statement that it has "no option" suggests you didn't look at the link.

 

[Heinrich]

... so not sure why I'm bothering.

If there is DSP available, it may work, but then it would be an option to do the x/o in the digital domain also. DASCs are very cheap today.

As you mentioned S.Linkwitz already, he designed a three way using analog filters on signal level. It is still found on his site. You'll see how much more is needed if there's no DSP available. And to begin with, you didn't mention the x/o parameters, that you have already in hardware.

If you're serious about the project, it is going to be a tough job for an expert even.

As far as I do not follow the Purify path, they have a tip on how to eliminate the resonant peak, and it should be done wih coil/cap passive for very good reason.

 

[JohnnyNG]

If the data of the crossover given are correct, you can choose between 24 and 12 bB. That's not bad and can, if drivers support it, be enough to align the drivers to a correct output.

Thanks! That was really the crux of my question - the drivers that can support it.

 

[JohnnyNG]

As far as I do not follow the Purify path, they have a tip on how to eliminate the resonant peak, and it should be done wih coil/cap passive for very good reason.

Understood! The Fiber Cone PTT6.5X04-NFA-01 6.5" I mentioned seems not to demand the notch filter recommended for the Alu cone variant, but perhaps wishful thinking.

 

[Heinrich]

... PTT6.5X04-NFA-01 6.5" ... perhaps wishful thinking.

O/k, you didn't reveal the hardware's features, which x/over is built in, the frequencies, tunable? I don't want to spoil your enthusiasm, this is not going to work without extra DSP upfront.

 

[Salt]

I realize I may be going about this bass-ackwards.

I've got a Sublime Acoustic K231 Stereo 3-Way Active Crossover which I'm using as a 2-way for sub/mains. Since it's a 3-way crossover, however, I was toying with the idea of building a 2-way which would work well with this basic kind of crossover. By "basic" I mean it's got Linkwitz-Riley filters, 4th order 24dB/Octave or optional 2nd order 12dB/Octave filter slope as well as Baffle Step Compensation, with adjustable gain and selectable baffle size ranges but can't do notch filters, etc.

So I'm thinking I'd need well-behaved drivers and am planning a sealed enclosure since I'd be high-passing to a sub anyway.

Any thoughts? I've been dreaming of a Purifi build but not sure how feasible that is. Maybe the Fiber Cone PTT6.5X04-NFA-01 6.5"? Seems not to demand the notch filter recommended for the Alu cone variant. Madisound suggests ".20 -.30 cf sealed for an F3 of 87Hz, F6 of 58Hz, F10 of 40Hz".

I'm coming from Revel M106 with which I have no complaints whatsoever so not sure why I'm bothering.

Just scrolled through the purify sites, and with 6,5" fiber cone the resonances above 5 kHz are a lot less than with Al cones.
With a crossover frequency at 2 kHz or below and 4th order even the resonance frequencies of the Alu cones should be attenuated sufficiently.
Going purely Purify with their new 33 mm tweater would enable a Xover almost down to 1 kHz

https://www.audiosciencereview.com/...6-prototype-is-here-with-a-ptt-tweeter.44523/

 

[JohnnyNG]

Just scrolled through the purify sites, and with 6,5" fiber cone the resonances above 5 kHz are a lot less than with Al cones.
With a crossover frequency at 2 kHz or below and 4th order even the resonance frequencies of the Alu cones should be attenuated sufficiently.
Going purely Purify with their new 33 mm tweater would anable a Xover almost down to 1 kHz

https://www.audiosciencereview.com/...6-prototype-is-here-with-a-ptt-tweeter.44523/

Thanks again!