The sonic benefits of an active crossover. A discussion.

[JonFo]

Wow, lots of activity this week.

On replacing existing passives with an active, it's perfectly doable on most two-way speakers. The science is actually getting a good transfer function for the original, then modeling something close in the DSP unit, then (and this is the important part) measuring and tuning using a calibrated measurement rig. Final touch is to tune for optimized in-room performance.

It is this last step where I believe a good-quality DSP-based active beats any passive, as you are now fine-tuning all parameters (that are tunable) to your exact situation.

Since the thread is about Sonic benefits, I realized I had not qualified my impressions of the multi-year progression from passive to top-grade actives. Again, same room, same amps, same speakers.

The clearest benefit was a dramatic reduction in bass distortion. Even though on the first pass, I had an analog active with no delay adjustments, the total cohesion of panel to woofer integration was way better.
From there, the various iterations of active XOs just continued to benefit from my slowly improving skills at acoustic measurement and crossover fine tuning. The results were:
- Imaging improvements once the drivers time-alignment was perfected
- Even tighter mid-bass performance
- Amps are able to deliver full power into demanding ESL, so reduced amp distortions
- Dynamic capability and power curve of system improved, sounds the same at mid or high volumes.

The current setup is miles and miles ahead of the same core set of drivers and amps and original passive. With imaging and a frequency balance rarely heard on other high-systems.

Even compared to a very nice 100K+ Meridian Digital theater system with DSP8000's all around, my setup has a cleaner and broader sound-stage. And that Meridian is a pretty awesome active-speaker setup.

 

[Groucho]

It is this last step where I believe a good-quality DSP-based active beats any passive, as you are now fine-tuning all parameters (that are tunable) to your exact situation.

My own experiments with DSP have been such a long road, that I shudder to think where I would be if I was messing around with passive crossovers only. I would only have been able to do a fraction of the experiments I have done with DSP, and wouldn't be in a position to know what I was listening for. If I thought my speakers were any good, I would be deluding myself, that's for sure.

In my case I have been writing the code myself and adding more and more tweakable parameters as I have gone along (baffle step compensation, phase correction, amplitude correction and so on) and I can now tweak to my heart's content in real time. With a luddite approach, I would really have just been fumbling around in the dark. My ability (or is it a curse?) to hear defects has improved as time has gone on, and I can honestly say that it is only with a fairly rigorous approach to measurement-based correction, yet with an element of 'black art' hand-modification of software - later to be realised as tweakable parameters - that I have homed in on a speaker that I can listen to for hours without tiring. The thought of then trying to realise it in coils, caps and resistors doesn't bear thinking about.

 

[LL21]

That's exactly what I did. I didn't have to rip out the old, I just disconnected the passive crossover between the midranges a tweeters and added an analog active crossover and more power amps. (In the IRS-V there was already an active crossover between the woofer and midrange) I use a PassLabs XVR1 three-way active crossover. I initially set the crossover points and slopes the same as the original passive crossover as a starting point and then made slight adjustments in the crossover points between the woofer and midrange. That took a fantastic speaker and made it fantasticker.

As Robert Frost said, "Two roads diverged in a wood and I took the one less traveled, and that has made all the difference."

Great minds, one can only hope... Exactly what I did, disconnected the internal crossovers and hooked up straight to the drivers, and started with the original crossover frequencies and slopes to get things going. "Rip out" was a bit of artistic writer's license...

Gents...i have spoken with an engineer about my X1s. He suggested that i remove the bridge on the back of the X1s connecting upper/mids and woofer to the speaker terminals and remove/bypass the passive woofer crossover. And then use a Krell KBX as an active crossover from my CJ GAT. Keep the Gryphon Colosseum driving uppers/mids and use a Gryphon Antileon, Krell FPB to drive the bass. Does this make sense to you? What would happen in your opinion in terms of the general benefits of active crossovers?

 

[edorr]

Gents...i have spoken with an engineer about my X1s. He suggested that i remove the bridge on the back of the X1s connecting upper/mids and woofer to the speaker terminals and remove/bypass the passive woofer crossover. And then use a Krell KBX as an active crossover from my CJ GAT. Keep the Gryphon Colosseum driving uppers/mids and use a Gryphon Antileon, Krell FPB to drive the bass. Does this make sense to you? What would happen in your opinion in terms of the general benefits of active crossovers?

Is the Krell an analog or digital X-over with A/D conversion? Do you have all digital sources or also analog?

 

[LL21]

Is the Krell an analog or digital X-over with A/D conversion? Do you have all digital sources or also analog?

I have both analog and digital sources, and i believe the Krell is an old analog crossover. BTW, if people have superior recommendations for active crossover, i an open minded.

But first, perhaps i should assess whether this is even a worthwhile upgrade to consider or not.

 

[edorr]

I have both analog and digital sources, and i believe the Krell is an old analog crossover. BTW, if people have superior recommendations for active crossover, i an open minded.

But first, perhaps i should assess whether this is even a worthwhile upgrade to consider or not.

Cannot comment on the approach. Just thinking in an all digital system a digital crossover would be the way to implement this architecture, but with analog sources you would need A/D conversion which is typically considered a non starter among analog buffs.

 

[LL21]

What do people think of this article on active crossovers?

http://sound.westhost.com/biamp-vs-passive.htm

 

[opus111]

It starts out fine, pointing out how passive XOs are a compromise in some ways. I agree. But then lower down we reach this piece of propaganda:

In this day and age, amplifiers and active crossovers can be built for (almost) peanuts - ok, not great amplifiers perhaps, but when used in an active system they can still outperform a megabuck top-of-the-line amp driving the same loudspeaker drivers through a passive crossover network.

That's where I gave up reading because its just pulled out of the air without evidence or reasoned justification.

 

[LL21]

It starts out fine, pointing out how passive XOs are a compromise in some ways. I agree. But then lower down we reach this piece of propaganda:

In this day and age, amplifiers and active crossovers can be built for (almost) peanuts - ok, not great amplifiers perhaps, but when used in an active system they can still outperform a megabuck top-of-the-line amp driving the same loudspeaker drivers through a passive crossover network.

That's where I gave up reading because its just pulled out of the air without evidence or reasoned justification.

Thanks Opus. If executed properly (always the question!)...is it fair to say that the active crossover inserted between preamp and amp adds another link in the chain (potential distortion)...but that removing the passive crossover between amp and speaker cones removes a greater potential for distortion?

Again...just curious about what happens if you take a Krell KBX with its internal program card set specifically for the bass crossover of the X1...and remove/bypass the X1 crossover (which can be done) and connect a bass amp directly to the bass module of the X1?

 

[opus111]

Harmonic distortion (THD) is not the primary SQ problem in audio, rather IMD is. But not the IMD as its currently measured. However its a very common, very pervasive myth that harmonic distortion is the thing to be avoided.

Bearing this in mind, passive crossover components (specifically inductors, because they come with series resistance) do introduce harmonic distortion, but this turns out not to be particularly audible. Active crossovers can be designed with vanishingly low harmonic distortion (just use opamps) but it turns out that their IMD contributions are often audible. Then there's the extra interconnections to consider - a source of noise and potentially also noise modulation. if the active XO was built into the amplifier and both didn't introduce significant IMD then this would be the ideal solution and would probably beat out the passive solution.

To answer your question - in a high end system, I'd be wary of introducing an extra box into the signal chain. It can't ever improve the sound - removal of a passive XO might only make a small improvement in SQ whereas the extra box and set of ICs has the potential to make a major reduction in SQ.

 

[LL21]

Harmonic distortion (THD) is not the primary SQ problem in audio, rather IMD is. But not the IMD as its currently measured. However its a very common, very pervasive myth that harmonic distortion is the thing to be avoided.

Bearing this in mind, passive crossover components (specifically inductors, because they come with series resistance) do introduce harmonic distortion, but this turns out not to be particularly audible. Active crossovers can be designed with vanishingly low harmonic distortion (just use opamps) but it turns out that their IMD contributions are often audible. Then there's the extra interconnections to consider - a source of noise and potentially also noise modulation. if the active XO was built into the amplifier and both didn't introduce significant IMD then this would be the ideal solution and would probably beat out the passive solution.

To answer your question - in a high end system, I'd be wary of introducing an extra box into the signal chain. It can't ever improve the sound - removal of a passive XO might only make a small improvement in SQ whereas the extra box and set of ICs has the potential to make a major reduction in SQ.

Thanks...in other words, if the active crossver is designed by the speaker manufacturer...like in the Rockport Arrakis II...that is one thing (vs the Arrakis II passive version) and is apparently quite a positive. But if going with a third party active crossover...be wary, even if the third party has all the crossover settings of your speaker.

 

[opus111]

Not quite what I was getting at, no

I'm saying the ideal solution, SQ-wise is a passive line-level XO, fitted inside your power amp or inside your preamp so no extra power supply and no extra ICs are required. Whether having a separate active XO box is an improvement is debatable. It could be, depending on the particular amps you use with it. Offloading the bass onto a separate amp can have a positive impact on the sound of your MF/HF amp for example if the amp in question has an underspecified PSU (most that I've seen do). So a lot depends on your particular set-up, no hard-and-fast rules here.

 

[LL21]

Not quite what I was getting at, no

I'm saying the ideal solution, SQ-wise is a passive line-level XO, fitted inside your power amp or inside your preamp so no extra power supply and no extra ICs are required. Whether having a separate active XO box is an improvement is debatable. It could be, depending on the particular amps you use with it. Offloading the bass onto a separate amp can have a positive impact on the sound of your MF/HF amp for example if the amp in question has an underspecified PSU (most that I've seen do). So a lot depends on your particular set-up, no hard-and-fast rules here.

Thank you! I just PM'd you.

 

[Phelonious Ponk]

Harmonic distortion (THD) is not the primary SQ problem in audio, rather IMD is. But not the IMD as its currently measured. However its a very common, very pervasive myth that harmonic distortion is the thing to be avoided.

Bearing this in mind, passive crossover components (specifically inductors, because they come with series resistance) do introduce harmonic distortion, but this turns out not to be particularly audible. Active crossovers can be designed with vanishingly low harmonic distortion (just use opamps) but it turns out that their IMD contributions are often audible. Then there's the extra interconnections to consider - a source of noise and potentially also noise modulation. if the active XO was built into the amplifier and both didn't introduce significant IMD then this would be the ideal solution and would probably beat out the passive solution.

To answer your question - in a high end system, I'd be wary of introducing an extra box into the signal chain. It can't ever improve the sound - removal of a passive XO might only make a small improvement in SQ whereas the extra box and set of ICs has the potential to make a major reduction in SQ.

Yeah. That's why high-end systems are always so minimalist, with as few boxes as possible.

Tim

 

[opus111]

I thought the high-end dealers all loved extra boxes because the more boxes, the more ICs and the ICs go for monstrous margins compared to the electronics?

 

[Groucho]

I'm saying the ideal solution, SQ-wise is a passive line-level XO...

Do you see no advantage in the 'fancy' things that can be done with DSP, like time alignment etc.? My own bugbear has been that once I realised that these things could be done, but weren't in passive or analogue-active speakers, I was never going to be happy until I got them. I'm not even saying I could definitely tell the difference every time, just that I'd know, so it would spoil my enjoyment! But these things are real and clearly measurable. If you have heard them, did you not find them to be something worth pursuing?

 

[Keith_W]

To answer your question - in a high end system, I'd be wary of introducing an extra box into the signal chain. It can't ever improve the sound - removal of a passive XO might only make a small improvement in SQ whereas the extra box and set of ICs has the potential to make a major reduction in SQ.

With any change in your system you gain some and you lose some. Having converted my speakers from passive to active in a stepwise fashion, I have seen first hand exactly what gains and what losses there are. To say that it "can't ever improve the sound" comes across as a dogmatic statement. How do you know? Have you tried? I can tell you that the biggest downside is the hit to the back pocket. Overall, it is a sonic gain.

 

[opus111]

Sure, I do see advantage in lots of 'fancy' things DSP can do. I wasn't in my statement poo-pooing digital crossovers, rather I had taken it as read that the signal was already an analog one. I haven't as yet found an ADC/DAC combination that I could put up against a passive line-level XO though for transparency. I think its got to be possible, just haven't seen a practical implementation as almost all ADCs in digital products nowadays are of S-D architecture.

 

[opus111]

To say that it "can't ever improve the sound" comes across as a dogmatic statement. How do you know? Have you tried?

I think I didn't express myself clearly enough so thanks for picking up on this. I'm not saying that in the whole system context that adding the extra box can't improve the sound overall, no I think it can. But that would be down to stuff like the amps PSUs working less hard as a result of only feeding a single driver. My statement was a philosophical one - that I know of no way a box could add additional information from the original recording - so the signal coming out cannot be better (purer, higher fidelity) than the signal going in. At best there can be zero degradation, never enhancement. If you know of a way that any box can increase the fidelity, input to output, I'm all ears to learn of that and I'll eat a double portion of humble pie to boot

 

[Keith_W]

I think I didn't express myself clearly enough so thanks for picking up on this. I'm not saying that in the whole system context that adding the extra box can't improve the sound overall, no I think it can.

It isn't really adding another box to your system. You are removing one crossover and replacing it with another. In the context of your system, you are simply moving the crossover from one location to another. There are multiple documented issues with passive crossovers, most of which are either ameliorated or minimized with an active crossover.