Active crossovers measurably better?

[Thomas savage]

I use a Sony STR-DA1200ES for one pair of three-way speakers. Going price on eBay seems to be about 30 - 150 euros.

Great, we love a cheapskate at ASR

 

[oivavoi]

FWIW, some time back I was a very intolerant active supremacist. After going for active monitors myself, and becoming very suprised by the jump in sound quality and clarity I experienced, I became convinced that all passive speakers were horrible. Which annoyed some traditional audiophile acquaintances of mine. But after having done some more listening, I had to moderate this view. There are clearly many passive speakers that sound amazing. The Giyas for example, I heard them recently and they're stunning. I'm not convinced that there are that many actives which beat out the giyas in their category (forward-firing electrodynamic "box" speakers).

But, my experience is still very clear when it comes to cheaper speakers. I know of very few cheap passives which really get it right. There might be some exceptions, but overall I feel that the difference between a JBL 305/308 and an average cheap passive small box is rather substantial. I know of setups with actives in the 2000 USD price range that can easily beat passive setups in the 10000 USD price range. Etc. But when venturing higher up the ladder, it seems to me that one can pull it off with passive crossovers as well. Except, of course, that possibilities are much better for tinkering with and adjusting the speaker when going active. EQ of the anomalies in the speaker is easier, one can delay drivers, etc.

 

[Purité Audio]

Unless you can compare speakers in your own room it is very difficult to have a really valid opinion, it is difficult enough when they are both in your room!
Keith

 

[DonH56]

A passive crossover can present a better ("nicer") load to an amplifier than using an active crossover before the amplifiers. The raw input impedance of drivers and box can be pretty ugly (not that crossovers always make it better). That and reduced cost and complexity (for the user) are the usual pros of passives that I recall. Obviously those issues do not apply to a complete active loudspeaker system (i.e. a box to which the user need only connect a line-level input from his source).

 

[sergeauckland]

A passive crossover can present a better ("nicer") load to an amplifier than using an active crossover before the amplifiers. The raw input impedance of drivers and box can be pretty ugly (not that crossovers always make it better). That and reduced cost and complexity (for the user) are the usual pros of passives that I recall. Obviously those issues do not apply to a complete active loudspeaker system (i.e. a box to which the user need only connect a line-level input from his source).

Really? The raw impedance of mid and HF drivers is mostly resistance with a small amount of inductance, so very easy. LF drivers have impedance changes at resonance, but those are generally an increase in impedance, and so pretty benign, and above resonance again tend to be largely resistive. Consequently, if anything, an amplifier connected directly to a driver will have an easier time than driving a complete passive system through a crossover.
S.

 

[oivavoi]

Unless you can compare speakers in your own room it is very difficult to have a really valid opinion, it is difficult enough when they are both in your room!
Keith

True. So let me rephrase: Some of the most subjectively convincing acoustic events that have been rendered to my brain from a stereo setup, have come from speakers with passive crossovers. This may not only be about the speakers, of course, it can also be connected to room acoustics, and totally non-objective things like bias for expensive shiny boxes, the amount of alcohol in my blood, etc. But nevertheless, some of my greatest stereo experiences have come from passives. That doesn't mean that these passives couldn't be improved further with active implementation. I have no way of knowing that! And other equally great experiences have been from active setups.

My point is simply that really great sound can be had from passive speakers. Even though it's probably more cumbersome, difficult and expensive to achieve great sound through a passive crossover than through active implementation.

 

[DonH56]

Really? The raw impedance of mid and HF drivers is mostly resistance with a small amount of inductance, so very easy. LF drivers have impedance changes at resonance, but those are generally an increase in impedance, and so pretty benign, and above resonance again tend to be largely resistive. Consequently, if anything, an amplifier connected directly to a driver will have an easier time than driving a complete passive system through a crossover.
S.

I do not claim to be an expert regarding speakers. For some drivers that is true, though I can guess we could debate what is the definition of a "small" value of inductance, but for others their interaction with the box causes severe impedance dips at certain frequencies. I agree that is more true with woofers than mid/tweeter drivers, though some designs do port the mid-bass and/or midrange drivers. The problem with woofers is usually below resonance; for other drivers it is for a signal above their response range (which I agree the crossover, of any flavor, should prevent). Crossovers also often include resistors to reduce the output ("tone down") the HF output of things like ESL and ribbon tweeters that would otherwise present very low impedance to the amplifier if directly connected. The attenuation resistor also serves to raise the load impedance. I said "can be" based upon my experience and measurements in the past, would certainly not say "is".

 

[Cosmik]

Really? The raw impedance of mid and HF drivers is mostly resistance with a small amount of inductance, so very easy. LF drivers have impedance changes at resonance, but those are generally an increase in impedance, and so pretty benign, and above resonance again tend to be largely resistive. Consequently, if anything, an amplifier connected directly to a driver will have an easier time than driving a complete passive system through a crossover.

I was going to say something similar but less authoritative. Stereophile speaker reviews often show really difficult loads presented by passive crossovers, whereas individual drivers are relatively benign and, of course, require much less power from the amp in the first place. Active systems really are win-win in every technical respect.

 

[March Audio]

If I could do what @BE718 has done and build some fantastic actives I would. Whenever I see him posts the measurements and pictures of the speakers I get tinges of jealously

I can send you the cad file then all you have to do is find some local firm with a CNC machine to cut the MDF. You glue it all together and get an auto paint sprayer to finish to your taste

Simples

 

[Thomas savage]

I can send you the cad file then all you have to do is find some local firm with a CNC machine to cut the MDF. You glue it all together and get an auto paint sprayer to finish to your taste

Simples

View attachment 10228

But then there’s the amp building and crossover work.. it’s ok, if it’s more complicated than putting gypsum mud on a wall I get a bit intimidated

 

[stunta]

There’s all sorts of active and passive crossovers. In fact, one could make an active crossover behave exactly the same as a passive crossover. So you’ll need to be more specific.

Ok. I think I understand.

At the very least, I was hoping someone would have converted a passive speaker to active and measured the before/after result and also done a listening test.

ATC makes passive and active versions of the same speakers and given they believe active is better, I am surprised they still make passive versions. I can understand if that was limited to the consumer series (for folks who want to switch amps) but they do this even for the pro series. Why not make external crossovers?

In software, we have shared libraries that are well established optimal implementations of common functionality. It is expensive and stupid to reinvent the wheel. There may be different implementations for different platforms but the underlying logic is typically the same and there aren't that many platform specific implementations. I don't see this pattern in hifi speakers. I would expect to have crossovers in software that I can configure depending on speaker specifications and a standardized external module splits the signal.

I understand this is effectively what is already happening inside DSP based active speakers but it's not componentized. I find this very frustrating. As a hobbyist I want to try different speakers but they are monolithic. It's quite bizarre and needs to be addressed.

 

[mitchco]

An older article, but still applicable: http://sound.whsites.net/biamp-vs-passive.htm

And it is a bit more than just about active vs passive: http://www.acourate.com/XOWhitePaper.pdf

While the entire article is excellent, this section: http://sound.whsites.net/ptd.htm#s2

There are measurements in each article that demonstrate active (e.g. digital) XO (and a very particular type at that) objectively measures better than passive wrt to not causing a multitude of issues, as addressed in each article.

 

[March Audio]

Ok. I think I understand.

At the very least, I was hoping someone would have converted a passive speaker to active and measured the before/after result and also done a listening test.

ATC makes passive and active versions of the same speakers and given they believe active is better, I am surprised they still make passive versions. I can understand if that was limited to the consumer series (for folks who want to switch amps) but they do this even for the pro series. Why not make external crossovers?

In software, we have shared libraries that are well established optimal implementations of common functionality. It is expensive and stupid to reinvent the wheel. There may be different implementations for different platforms but the underlying logic is typically the same and there aren't that many platform specific implementations. I don't see this pattern in hifi speakers. I would expect to have crossovers in software that I can configure depending on speaker specifications and a standardized external module splits the signal.

I understand this is effectively what is already happening inside DSP based active speakers but it's not componentized. I find this very frustrating. As a hobbyist I want to try different speakers but they are monolithic. It's quite bizarre and needs to be addressed.

Well thats pretty much what I did, and as mentioned both the technical and subjective performance was significantly improved. This is of course a sample of one and statistically meaningless . However the technical improvements of active dsp I dont think are in question. So why shouldnt it lead to an improvement in subjective performance?

As already mentioned, there will be inertia to overcome with this type of speaker in commercial designs. It doesnt fit well with certain audiophile rationales, the product will inevitibly be more expensive than a passive box, it doesnt allow audiophiles to tweak amps, cables etc.... Its currently a small part of the market for commercial reasons and not performance reasons. That will change.

There is no question in my mind that active dsp xo is a superior solution. That does not mean that there are not excellent passive speakers out there, far from it, but there are technical improvements you can make with active dsp that simply cant be acheived passive.

 

[svart-hvitt]

The fact that ATC and other producers haven’t gone full active; could it be lack of integrity?

To clarify: Would a doctor recommend patients to continue smoking so as not to risk losing his patients to doctors who recommend whatever the patient wishes?

 

[March Audio]

The fact that ATC and other producers haven’t gone full active; could it be lack of integrity?

To clarify: Would a doctor recommend patients to continue smoking so as not to risk losing his patients to doctors who recommend whatever the patient wishes?

Its not lack of integrity, its a reasonable commercial imperitive. At the end of the day its a business that needs to sell product. Its a poor business decision that cuts of a large segment of its market - which it is still serving well with good and more cost effective products.

I dont think there is any comparison to a moral medical analogy.

 

[svart-hvitt]

Its not lack of integrity, its a reasonable commercial imperitive. At the end of the day its a business that needs to sell product. Its a poor business decision that cuts of a large segment of its market.

Its not lack of integrity, its a reasonable commercial imperitive. At the end of the day its a business that needs to sell product. Its a poor business decision that cuts of a large segment of its market.

I dont think there is any comparison to a moral medical analogy.

I think the «pill pusher» analogy is relevant.

In the end I think a firm’s serviceability will be more important than its vendibility. The world is awash in products. And I think (hope...) people are about to wake up to that fact.

 

[March Audio]

Naa, there is a world of difference between recommending someone continue doing something harmful to their health for commercial gain and continuing to sell products, which although may not deliver the ultimate in performance, are perfectly good and fit a real commercial demographic.

Should a company not sell a cheaper 2 way speaker because the bigger more expensive 3 way is better? Of course not . There is a genuine market for both.

Also, ref your example, a small company like ATC probably needs a range of products that fit different price points to survive. If they just made one top end active speaker they may not be around very much longer.

 

[Cosmik]

At the very least, I was hoping someone would have converted a passive speaker to active and measured the before/after result and also done a listening test.

But this would tell only half the story. Going active gives the designer more choice in the selection of drivers, and other aspects, so the difference should be more 'holistic' than just a straight swap for the passive crossover.

Here's a challenge: can you create a true multi-way speaker where you can switch between two completely different crossover setups (radically different frequencies, slopes) and hear no difference while playing music? With a DSP-based, driver-corrected, time aligned, linear phase three-way system, this demonstration is easily accomplished - in fact the final choice of crossover frequencies and slopes is almost arbitrary. The passive designer would be sweating over their version for months and could never achieve it: their version must always produce artefacts that are perhaps not obvious when static, but would be shown up with an instantaneous changeover.

 

[March Audio]

But this would tell only half the story. Going active gives the designer more choice in the selection of drivers, and other aspects, so the difference should be more 'holistic' than just a straight swap for the passive crossover.

Here's a challenge: can you create a true multi-way speaker where you can switch between two completely different crossover setups (radically different frequencies, slopes) and hear no difference while playing music? With a DSP-based, driver-corrected, time aligned, linear phase three-way system, this demonstration is easily accomplished - in fact the final choice of crossover frequencies and slopes is almost arbitrary. The passive designer would be sweating over their version for months and could never achieve it: their version must always produce artefacts that are perhaps not obvious when static, but would be shown up with an instantaneous changeover.

OMG how true, optimizing my passive XO was a true, expensive and time consuming PITA!

 

[sergeauckland]

Ok. I think I understand.

At the very least, I was hoping someone would have converted a passive speaker to active and measured the before/after result and also done a listening test.

ATC makes passive and active versions of the same speakers and given they believe active is better, I am surprised they still make passive versions. I can understand if that was limited to the consumer series (for folks who want to switch amps) but they do this even for the pro series. Why not make external crossovers?

In software, we have shared libraries that are well established optimal implementations of common functionality. It is expensive and stupid to reinvent the wheel. There may be different implementations for different platforms but the underlying logic is typically the same and there aren't that many platform specific implementations. I don't see this pattern in hifi speakers. I would expect to have crossovers in software that I can configure depending on speaker specifications and a standardized external module splits the signal.

I understand this is effectively what is already happening inside DSP based active speakers but it's not componentized. I find this very frustrating. As a hobbyist I want to try different speakers but they are monolithic. It's quite bizarre and needs to be addressed.

That's exactly what I did with my B&W 801Fs.
I have a description of the process and the graphs on my web site. https://sites.google.com/site/audiopages/
Unfortunately, it's not really possible to compare directly the before and after, as this compares 30 year old drivers and a 30 year old passive crossover, with 30 year old drivers, but new tweeters and a modern DSP-based crossover and equaliser.

It's probably more instructive to compare the original specification with my results, and the active version was clearly an improvement. The frequency response of the originals was +-2dB, still pretty good for a passive 'speakers, my version after equalisation was +-1dB. Harmonic Distortion was somewhat but not hugely better. Pair matching was closer, but again wasn't bad on the originals. Although the 801s were claimed to be mechanically time-aligned, using the DSP crossover, that could be improved, but again, not by much.

My conclusion is that a very good passive loudspeaker will be made better by going active, but not hugely so, whilst more modest passives will be much more improved.

When I was in the business, albeit over 30 years ago, I sold Meridian and John Bowers actives, much to the surprise of fellow dealers who wouldn't touch them. Their reasoning was that audiophiles liked to fiddle with cables, amplifiers etc, and would upgrade the cables, the amplifiers, do stupid things like passive biamping etc. All gave the dealer additional selling opportunities. Selling actives firstly made the sale a much bigger ticket item, so reducing the opportunities, and secondly, once sold that's it. No further upselling/upgrading.

I suspect that's behind ATC and PMCs continuing with passive versions of their loudspeakers, dealer pressure.
S.