First Order Crossovers - Snake Oil or difference?

[TLS Guy]

I'm certainly not a qualified speaker designer, but the difference in arrival times has got to be so short, in the range of a few cycles even at 15KHz-20KHz, I can't believe this is an audible effect. Have you run experiments between coherent and non-coherent alternatives and heard a difference?

Actually yes.

About 10 years ago, I was involved in a research project with my son over instrumental synthesis to avoid sound sampling. The problem with building music from sound sampling is that it precludes altering the dynamic in a musical way.

This is a long story, but essentially I recorded some live instruments in a time and phase coherent fashion. This included violins from Italian masters.

Then my son mathematically sorted out the the harmonic structure of the sounds with carefully written programs.

Now certain things came out as expected. The most complex harmonic structure belongs to the human voice and the violin. It is no accident that composers and musicians have regarded the violin as most akin to the human voice.

I and others have long known that it is fairly easy for a reproducer to do a good job of reproducing a French horn. It harmonic structure is not that complex.

The violin and the human voice are the very devil for reproduction systems to get right, or even passable for most.

Now we spent a lot of time focusing on the violin and its the relationship of its harmonic structure to the fundamental.

To cut a long story short, we found that Ted Jordan was in essence correct that maintaining all the correct relationships of the harmonics to the fundamental was vital to capturing and being able to even making any violin sound synthesis.

It is no accident that the violin has a very hard time making friends through recordings, but not in the recital hall. Listening to the the Bach violin sonatas on all but a very select few reproducers is an excruciating experience to many.

This project convinced me that this has a lot to do with the wave envelope and preserving it intact as much as possible is important.

As far as general audibility is concerned that is problematic because as I have pointed out previously, most commercial recordings are awash in violence to time and phase and how much further damage the speaker can do is mute.

However I have to stress there are bigger fish to fry in the total concept of a speaker than time and phase coherence. In fact you can do a lot of damage trying to achieve it.

However my experience with full range drivers is actually very revealing. I set up a demonstration with a full range driver system for the engineers at MISCO some months ago, and they were in many ways astonished. I did use carefully selected recordings. Obviously there are significant limitations to full range drivers, but they are revealing of the significant trespasses of crossovers.

In my designs I try and let acoustic roll offs do as much of the work as possible, with the minimum required from electrical crossovers commensurate with very smooth frequency response, lobing pattern and dispersion control.

To achieve the above goals using only first order crossovers is in my view an impossible task.

And by the way, it is a myth that first order crossovers have no time and phase shift. They do and its 90 degrees and a quarter cycle.

 

[GranteedEV]

I'd probably be of the position that inter-driver phase difference at the listening spot is a lot more important than overall phase rotation of the system.

This is a problem with ""first order"" crossoverss because the multiple driver outputs sum in phase quadrature - else you would have a 3db peak at the crossover frequency.

What that means?

It means that the drivers are 90 degrees out of phase with each other if you did the crossover correct. Small errors in the phase tracking or even your listening position and you might be sitting at 120 degree out-of-phase drivers. Then you've probably got problems in the speaker's subjective ""presentation"". I've also in other threads mentioned other countless issues with going first order and odd-order - the vertical lobing pattern, the horizontal polar response, the minimal supression of non-pistonic driver behavior, and tweeter over-excursion.

2nd and 4th order acoustic filters, or at least those of the linkwitz riley sort, are 6db down at the crossover point, and so if the drivers are completely in phase, even "missing the target" barely gives the listener 20, 30 degrees of phase difference between drivers. IOW, these filters inherently optimize a coherent forward wave. Going IE 4th order also helps keep voice coils from exciting cone breakup (so kevlar, aluminum, magnesium etc can be used), tweeters operating in their "happy" passband(ribbons can be used), allows multi-way speakers to sum more correctly (you don't have three or four different drivers all giving their own contribution to the sound at the LP - a recipe for disaster that led to the now primitive "decade rule" of keeping all crossovers a decade apart [something the Revel Salons, for example, certainly don't follow]), and lastly, it also keeps dispersion wide as beaming drivers make less contribution to the forward frequency response.

If someone insists on linear phase, they should do it as an after-the-fact correction. Even if the speaker as a whole uses 4th or higher order IIR filter, you can manually implement FIR all-pass filters over-top to correct for phase and even time. This can be done with a lot of computer-based listening setups thanks to advances in software.

Do I think that any one way is right? Dunno. But in general a 4th order filter has a lot going for it making it an effective sweet spot. Any higher and you're probably just going into the realm of diminishing returns. Any lower and you're running into other issues.

So are 1st order filters snake oil? No. But they introduce as many problems as they solve.

 

[Irvrobinson]

To cut a long story short, we found that Ted Jordan was in essence correct that maintaining all the correct relationships of the harmonics to the fundamental was vital to capturing and being able to even making any violin sound synthesis.

This may explain why I fell so hard for the Sound Lab electrostatics. I ultimately didn't choose them for a plethora of other reasons, not the least of which is their sheer hugeness, but on certain instruments, strings especially, they make for the most effective audio presentation I've ever heard.

 

[Irvrobinson]

So are 1st order filters snake oil? No. But they introduce as many problems as they solve.

I'm surprised at this closing comment, because you seemed to be making a very effective argument that in the realm of most consumer products 1st order crossovers *were* snake oil.

As a perspective speaker buyer, I don't think I'd ever completely discount a speaker just because it used 1st order crossovers and tried hard to be phase and time coherent, but I'm more skeptical of them. Certainly I seriously considered the Dunlavys, and John's designs did sound awesome within the limitations of the drivers he could find while he was alive. Nonetheless, the Dunlavys are the only really interesting design of this genre I'm aware of. (I've never understood the fascination with Thiel or Vandersteen, though I have listened several times.)

 

[AcuDefTechGuy]

Wow, looks like their spec'd measurements on my previous link aren't really much to go by.

Exactly.

I'm not going to say which sounds better. That's pure subjective opinion.

But from an objective retrospective point of view, 4th order XO speakers have been shown to have much better on-axis and off-axis frequency responses. And the proof is in the measurements on Stereophile and NRC, not somebody's opinion about what makes sense or not.

Andrew Jones of TAD said in an interview once that you can't be afraid to use complex crossover designs to achieve your goal. Siegfried Linkwitz would agree. And who am I to disagree ?

 

[GranteedEV]

I'm surprised at this closing comment, because you seemed to be making a very effective argument that in the realm of most consumer products 1st order crossovers *were* snake oil.

well: All crossovers haves pros and cons; some outweighing others. But "snake oil" would imply that audible differences are only imagined. 1st order crossovers definitely have aspects some might prefer, just like a SET amp or a loudness EQ or a poorly braced cabinet with a mistuned vent. Not our place to disagree with preferences. On paper though these are distortions but it's not my place to tell someone not to like what ever they like, as long as it is REAL.

Now if they imagine they like something that can't be measured - or doesn't exist - then that is snake oil.

If it can be measured, but audibility is dubious, then it's just "unverified" - IE the differences between good amplifiers or even crossover components. You can measure differences between those things, and maybe some can hear that. But I put this as "Beyond the fundamentals" and am essentially uninterested.

Linear phase is in that category. I'd love to know, but 2K on a DEQX just to find out is just a tad bit steep for my part time income...I'm sure things would measure better, just like you adding two or three more subs probably would. Just not convinced it's worth it from a reliable audibility perspective.

I'm all about that first 95%, not the last 5%. Snake Oil products do not exist in this spectrum at all.

 

[j_garcia]

well, I'll say this;

All crossovers haves pros and cons; some outweighing others. But snake oil would imply that the differences are only imagined.

1st order crossovers definitely have aspects some might prefer, just like a SET amp or a loudness EQ or a poorly braced cabinet with a mistuned vent. Not our place to disagree with preferences. On paper though these are distortions but it's not my place to tell someone not to like what ever they like, as long as it is REAL.

Now if they imagine they like something that can't be measired - or doesn't exist - then that is snake oil.

I was going to comment somewhat to this effect as well. I've actually owned some inexpensive speakers that utilized 1st order x-overs and thought they sounded great (Mordaunt Short 902s) Not because they had 1st order x-overs, but because the overall speaker was well designed, and that is the key. While I may look at such details as the x-over point, type used, overall response, on and off axis performance and other factors, but what it really comes down to is: does the speaker sound good? If yes, then sweating the details is less of an issue to me.

As mentioned, first order crossovers are real and have a real effect. It is not a shady marketing "product" trying to make you believe whatever they are selling is having an effect. It is a design choice of the speaker designer. Also, as mentioned by TLS earlier, ALL crossovers affect the resulting sound, period.

 

[GranteedEV]

I've actually owned some inexpensive speakers that utilized 1st order x-overs and thought they sounded great (Mordaunt Short 902s)

I know you probably know this, but for some others reading I just want to point out that while in expensive speakers might claim to be first order, it may be true only for barely an octave - I doubt that it's first order over a wide bandwidth, even if a 1st order electrical filter is used. Tweeters will have an inherent 2nd order rolloff as you go down in frequency, and midwoofers with voice coils will have inductive and mass based rolloffs of their own.

For inexpensive speakers claiming first order slopes, this may only be true for a very narrow bandwidth, after which it can transition to 3rd or even higher order rolloffs.

 

[ridikas]

None of you even know what a first order crossover is. So how can this thread be three pages long?

First challenge:

Show me a speaker which has a first order crossover and then prove it with measurements. I challenge YOU!

This is the most ridiculous thread ever. Obviously made by someone who hasn't got the slightest idea on what their talking about.

 

[mike c]

This is the most ridiculous thread ever. Obviously made by someone who hasn't got the slightest idea on what their talking about.

From the creator of ...

http://forums.audioholics.com/forums/search.php?searchid=5785819

"Top 10 things that can get you a time out"

 

[TLS Guy]

None of you even know what a first order crossover is. So how can this thread be three pages long?

First challenge:

Show me a speaker which has a first order crossover and then prove it with measurements. I challenge YOU!

This is the most ridiculous thread ever. Obviously made by someone who hasn't got the slightest idea on what their talking about.

Really! I have designed quite a few, have you?

 

[ridikas]

Awesome! We have an engineer who has designed first order crossovers. Please give us a quick description of the drivers that you've chose, the crossover, and please post the measurements showing each individual driver's roll off with the crossover in place. Thanks!

 

[ridikas]

OK, since no one accepted the challenge, I will answer OP's question:

**THERE ARE NO SPEAKERS ON THE MARKET, NOR HAVE THERE EVER BEEN, WHICH USE FIRST ORDER CROSSOVERS!**

The drivers which would be required to achieve these slopes have not been invented yet.

The holy grail of speaker design today would be extraordinarily carefully chosen drivers and implemented with a second order ACOUSTICAL slope. The tweeter would have to be physically offset to account for delay.

If the drivers are on the same plane/baffle, the crossover would require a complex lattice network on the woofer. See the more expensive Dynaudio speakers as an example of this style of network/design/speaker. Unfortunately, this completely negates the benefits of a second order crossover.

Most drivers are manufactured (with natural rolloff) to work well with LR4 slopes. In this style of design, the crossover is asymmetrical, with the woofer's slope more shallow to account for the acoustic offset between the drivers.

The complexity works in reverse. LR4 is the simplest crossover to build, i.e. usually requiring 2 parts on the tweeter. Forget about electrical and only think acoustical. Yes, a 2 part crossover on the tweeter would be considered a second order electrical, but in reality combined with the tweeter's natural rolloff, the resulting acoustical slope is LR4.

A second order acoustical crossover will be extremely complex. Most certainly LCR networks, etc. would have to be used on the tweeter.

A first order acoustical crossover is impossible with the drivers available today. Even if you can hammer the drivers into these slopes using 100+ crossover parts, the resulting speaker will have severe distortion and limited output.

 

[Matt34]

None of you even know what a first order crossover is. So how can this thread be three pages long?

It's one page if you have the correct settings
First challenge:

Show me a speaker which has a first order crossover and then prove it with measurements. I challenge YOU!

This isn't your thread, OP asked a very specific question. Learning is very much part of the reasoning behind forums like these.

This is the most ridiculous thread ever. Obviously made by someone who hasn't got the slightest idea on what their talking about.

This is your very last warning about trolling/insulting members.

 

[ridikas]

Trolling? I was the ONLY member to answer OP's question.

P.S. I've made a mistake in my original post, the lattice network would have to be implemented on the tweeter, not the woofer.

 

[Irvrobinson]

Trolling? I was the ONLY member to answer OP's question.

That's incorrect. GranteedEV and TLS Guy both offered technical comments. Especially GranteedEV, who mentioned that most claimed 1st order crossovers might only behave that way for a narrow frequency range. I actually found that quite interesting and thought provoking. I also found TLS Guy's empirical findings to be interesting (though not definitive). You, on the other hand, have offered nothing here except veiled personal attacks, assertions without qualifications, and show-me-the-data demands that are rude for a forum like this unless you present some yourself. Now you have positioned yourself as an expert beyond all others on this thread, and you have driven away most everyone else from what was a fun and informative thread.

I'm not sure you're a troll, but you're not a very good teacher, if you are indeed an expert, and you're mostly frustrating.