Subwoofer Crossover Advice

[andreasmaaan]

Have a look at the following short thread.
https://devialetchat.com/Thread-PREOUT-and-SPEAKERS-filters-type-solved-BESSEL-filters

Ok, it's slightly different with Bessel filters, but that short thread is somewhat oversimplifying/misleading.

I know others here are far sharper than me when it comes to filters, but to summarise the situation with Bessel crossovers: while the group delay of the summed response exhibits less severe peaking compared to a Butterworth or L-R crossover, this does not result in the summed output of a 2nd or higher order Bessel crossover having linear phase. So with Bessel crossovers, the overall degree of group delay will be comparable to other crossover topologies of an equivalent order.

In other words, Bessel filters behave more or less the same as other non linear-phase crossover topologies; that is, a 2nd order Bessel will require opposite driver polarities to sum correctly. However, it will have a less severe peak in the group delay than a Butterworth or L-R crossover of the same order, the trade-off being a slightly less flat amplitude response.

There is a pretty good explanation of this here.

 

[andreasmaaan]

To illustrate:

2nd order Bessel, both drivers with positive polarity (blue = subwoofer, red = woofer, black = sum):

2nd order Bessel, drivers with opposite polarities:

 

[Roen]

I'm a big fan of Rythmik subs for being accurate and blending well. I also had Power Sound Audio subs in the past, but much prefer the Rythmik.

Also agree on 80 Hz, based on my personal unscientific tweaking.

 

[watchnerd]

When I run the mains full range with no sub I used SAM.

When I use a sub using the Devialet to cross over, so HP to mains, and mono LP to sub I continue to use SAM on the mains. Theoretically this should make the roll off much more predicable for the mains.

You should not use SAM if you are using the high level into the sub from the speaker sockets, like REL recommend, as the SAM correction will also go to the sub, which will be all kinds of wrong.

If using stereo (non REL), as opposed to mono, I suppose the same applies?

 

[Soniclife]

If using stereo (non REL), as opposed to mono, I suppose the same applies?

Yes, if you are connecting your sub to the pre-outs SAM is not sent there. And easy to test, just turn the SAM % up and down and see what happens.

 

[watchnerd]

I'm a big fan of Rythmik subs for being accurate and blending well. I also had Power Sound Audio subs in the past, but much prefer the Rythmik.

Also agree on 80 Hz, based on my personal unscientific tweaking.

Do you know if Rythmik have any wireless or auto-EQ /DSP models? At a quick glance, they all look fairly old school. Nothing wrong with that, per se, just clarifying.

 

[watchnerd]

You should not use SAM if you are using the high level into the sub from the speaker sockets, like REL recommend, as the SAM correction will also go to the sub, which will be all kinds of wrong.

BTW, I don't understand how the REL approach could be a best practice.

Unless I'm misunderstanding, it seems like I'm either:

1. Running a high level signal through an analog crossover, with the potential addition of noise that active components have, plus the phase shifts

or

2. Running a high level signal through a DSP which just seems un-necessary compared to doing it all in the "preamp"

Am I missing something?

Nobody else seems to do it the REL way....so it seems eccentric to me.

 

[watchnerd]

To illustrate:

2nd order Bessel, both drivers with positive polarity (blue = subwoofer, red = woofer, black = sum):

View attachment 23799

2nd order Bessel, drivers with opposite polarities:

View attachment 23800

Does this imply I should always flip the polarity switch on a subwoofer if I'm using 2nd order Bessel filters?

It would seem the answer would be "not necessarily", as we're dealing with acoustic vs electrical summing, right?

 

[andreasmaaan]

Does this imply I should always flip the polarity switch on a subwoofer if I'm using 2nd order Bessel filters?

It would seem the answer would be "not necessarily", as we're dealing with acoustic vs electrical summing, right?

That would probably be my answer, or perhaps more of an “in most cases”

 

[Roen]

Do you know if Rythmik have any wireless or auto-EQ /DSP models? At a quick glance, they all look fairly old school. Nothing wrong with that, per se, just clarifying.

Not sure, but my gut instinct is no.

 

[DonH56]

Do you know if Rythmik have any wireless or auto-EQ /DSP models? At a quick glance, they all look fairly old school. Nothing wrong with that, per se, just clarifying.

No. Most models include several crossover selections, phase control, and single PEQ. You can use those to manually dial it in, use your AVR's room correction program, or an external device like a miniDSP aor Antimode. I'm OK with that; prefer they focus on the subs and not the bells and whistles most folk won't need or use.

 

[Soniclife]

In other words, Bessel filters behave more or less the same as other non linear-phase crossover topologies; that is, a 2nd order Bessel will require opposite driver polarities to sum correctly. However, it will have a less severe peak in the group delay than a Butterworth or L-R crossover of the same order, the trade-off being a slightly less flat amplitude response.

There is a pretty good explanation of this here.

Rane's Bessel crossover in Halogen is set for phase match between low-pass and high-pass. This minimizes lobing due to driver separation, and also results in a pretty flat combined response.

It was following along until the above bit, which suggests the only way to know in this case is to measure. Which is always best anyway.

 

[Soniclife]

Do you know if Rythmik have any wireless or auto-EQ /DSP models?

You can buy wireless modules from other sub manufactures, and EQ from antimode etc, but once you start adding those extras on it changes the price and complexity somewhat.

 

[Soniclife]

BTW, I don't understand how the REL approach could be a best practice.

Unless I'm misunderstanding, it seems like I'm either:

1. Running a high level signal through an analog crossover, with the potential addition of noise that active components have, plus the phase shifts

or

2. Running a high level signal through a DSP which just seems un-necessary compared to doing it all in the "preamp"

Am I missing something?

Nobody else seems to do it the REL way....so it seems eccentric to me.

I've never bought into the logic of what REL say, it mainly sounds like pandering to audiophiles who would need a lie down in a dark room if someone added a high pass filter to their main speakers, especially if it was done in the digital domain. Plus they are targeting people who probably do not have a simple way of doing sub crossovers in their pure 2ch system, so buying a REL and using as they suggest does not need any new boxes other than the sub.

 

[Soniclife]

If using stereo (non REL), as opposed to mono, I suppose the same applies?

I can think of no reason stereo or mono makes any difference here.

 

[andreasmaaan]

It was following along until the above bit, which suggests the only way to know in this case is to measure. Which is always best anyway.

By “phase match” they mean the filters are optimised for matched phase at the crossover point (as opposed to optimising for flattest possible amplitude). This will mean inverting one driver’s polarity in the case of a 2nd order crossover (since otherwise there would be a 180-degtre phase mismatch).

And yeh, measurement is always best!