How Do Horn Speakers Get Their Gorgeous Life-like Tonality?

DaveC

Industry Expert
Nov 16, 2014
2,518
371
83
Trying to align it so impulse/step-response (same thing if you're measuring off the speaker alone) is lined up is just playing with phase. You win some and lose some by trying to set different drivers at slightly different timing through DSP. Physically doing it gains the most claims for advantages but there are other things happening that aren't accounted for so "time alignment" is awarded the prize. If it was all such a huge deal Magico wouldn't be heralded as a good company in any way. They have tweeters in front of the woofers on every speaker but their ultimate.

If you want a cleaner response you basically are just making sure your box doesn't hold a bunch of energy. To do this it may not have anything to do with moving drivers forward and back, even if that could possibly help. It isn't alignment, it's just making sure everything dissipates in a relatively good manner so drivers are not singing well past the signal, while others are not.
You have to consider crossover delays, slope and frequency ime.

I do agree it's an overblown marketing issue but I did the math by hand, tested it and measured it and there are situations where time/phase are very important.
 

Folsom

VIP/Donor
Oct 26, 2015
3,594
289
83
Eastern WA
Have you find out all yourself by trying and testing it? I mean not just place a tweeter a bit behind so it is align but fully adjustable.
Chicken or egg. Is placing the tweeter farther back better because of time alignment or other factors? That's what I've been saying from the start, and that it's other factors.
 

morricab

Well-Known Member
Apr 25, 2014
3,268
429
83
Switzerland
Trying to align it so impulse/step-response (same thing if you're measuring off the speaker alone) is lined up is just playing with phase. You win some and lose some by trying to set different drivers at slightly different timing through DSP. Physically doing it gains the most claims for advantages but there are other things happening that aren't accounted for so "time alignment" is awarded the prize. If it was all such a huge deal Magico wouldn't be heralded as a good company in any way. They have tweeters in front of the woofers on every speaker but their ultimate.

If you want a cleaner response you basically are just making sure your box doesn't hold a bunch of energy. To do this it may not have anything to do with moving drivers forward and back, even if that could possibly help. It isn't alignment, it's just making sure everything dissipates in a relatively good manner so drivers are not singing well past the signal, while others are not.
Just getting the phase right will not guarantee time-alignment. All drivers can be in correct phase but the arrival of sounds is not at the same time due either to physical distance from the listeners ears or group delay, like one would get from a 4th order LR filter. It’s in phase but two drivers are delayed in time.
 

morricab

Well-Known Member
Apr 25, 2014
3,268
429
83
Switzerland
Chicken or egg. Is placing the tweeter farther back better because of time alignment or other factors? That's what I've been saying from the start, and that it's other factors.
You can test this digitally, which leaves all drivers in a fixed position. Then correct phase and delay so all sounds arrive at the same time. I have done this at home and heard demos with digital corrections and it matters.
 

Folsom

VIP/Donor
Oct 26, 2015
3,594
289
83
Eastern WA
You cannot emmulate what moving a driver does with software.

And trying to correct with delay still leads to the fact you are going to have the situation where changes in frequency arrive closer in time but then then decay late. It may sound different but that isn’t the same as obviously better. It hasn’t gained much popularity over the years for a reason. And manufacurers continue to prove physical alignment isn’t paramount.
 

morricab

Well-Known Member
Apr 25, 2014
3,268
429
83
Switzerland
You cannot emmulate what moving a driver does with software.

And trying to correct with delay still leads to the fact you are going to have the situation where changes in frequency arrive closer in time but then then decay late. It may sound different but that isn’t the same as obviously better. It hasn’t gained much popularity over the years for a reason. And manufacurers continue to prove physical alignment isn’t paramount.
No, this is not how digital time delay works, at least not when incorporated with a digital crossover. The time of all frequencies going to a particular driver are shifted in time so that they lineup with the frequencies generated by the other drivers. If you achieve a perfect step response then all frequencies, regardless of whether or not it is an initial arrival or the decay, will arrive at the correct time.

The situation for a passive speaker is worse where you will have a woofer that is delivering, say 400Hz at one time and the midrange is delivering the same 400Hz at a slightly different time to the ear...even if they are in-phase due to the different path lengths the wavefronts have to travel in space. This smearing in time is sometimes audible, particularly with large group delays and with higher frequencies.

They prove this how? Do they make a version of the speaker available that is with time-alignment vs. one without? No, they do not. Many designers simply don't think it is important and their agenda is different like controlling frequency response on and off-axis, reducing cabinet and driver distortions or whatever. They think because the ear/brain works as an integrator that it will take the scrambled mess that comes out of most speakers and make sense of it...and to a large extent they are right. However, once one becomes accustomed to a time coherent speaker and how it allows an easier time with listening, then one often doesn't want to go back to a non-coherent speaker. Not everyone is sensitive to it but if you are then you hear the lack of coherence.
 

bonzo75

Member Sponsor
Feb 26, 2014
11,397
1,352
113
London
Oh yes, deep indeed! I heard a WE demo...actually it was Line Magnetic WE clones but it was awesome and I heard it twice. The AcousticPlan manufacturer was demoing them and was importing them into Germany...was a great 40K system with one of the midsized snail horns (15A maybe) the little horn tweeter (can't remember now) that doesn't really go that high and then a big Jensen field coil bass driver with a short horn front side and some kind of box (sealed?) behind. The mid driver was a 555 copy I think and the tweeter driver was also a copy of a famous WE driver. The power supply for all these field coils was a huge tube powered thing. Awesome.
What Acoustiplan has are a big 16a in the middle, and two smaller 22a by the sides. Both cross over to WE subs, you can play either of them. If you want to do it yourself, the 16a horn itself (replica from South Korea) used to be 7500 euro three years ago. The LM drivers are quite cheap and can be found used. The cost is then to install as getting the power supplies right etc requires someone who knows. And what he also did is, to get the woofers right, he had to buy 4 to 5 pairs to find a matched one. That search process itself will cost a lot.
 

morricab

Well-Known Member
Apr 25, 2014
3,268
429
83
Switzerland
What Acoustiplan has are a big 16a in the middle, and two smaller 22a by the sides. Both cross over to WE subs, you can play either of them. If you want to do it yourself, the 16a horn itself (replica from South Korea) used to be 7500 euro three years ago. The LM drivers are quite cheap and can be found used. The cost is then to install as getting the power supplies right etc requires someone who knows. And what he also did is, to get the woofers right, he had to buy 4 to 5 pairs to find a matched one. That search process itself will cost a lot.
Yes, that sounds right...the big one in the middle and the smaller snails one per channel. The woofers though he told me were Jensen and not WE and not replicas.
 
Likes: christoph

Folsom

VIP/Donor
Oct 26, 2015
3,594
289
83
Eastern WA
Either information is left out, or sped up at some point. DSP/crossovers is not smart enough to perfect choice on when notes are played or such. This is a reality because the group delay can't be removed, just moved.

For your 400hz example to be true they would have to be beaming if they're in phase at the speaker. As far as audibility there's a lot more factors that will contribute than these distances that mean nothing.

With horns there are other concerns related to size, shape, and wavelength. So yes, they need to "align" you might say. Otherwise they would multiply the problem greatly compared to a dynamic speaker where tens of thousands of a second don't mean a lot.
 
Likes: christoph

morricab

Well-Known Member
Apr 25, 2014
3,268
429
83
Switzerland
Either information is left out, or sped up at some point. DSP/crossovers is not smart enough to perfect choice on when notes are played or such. This is a reality because the group delay can't be removed, just moved.

For your 400hz example to be true they would have to be beaming if they're in phase at the speaker. As far as audibility there's a lot more factors that will contribute than these distances that mean nothing.

With horns there are other concerns related to size, shape, and wavelength. So yes, they need to "align" you might say. Otherwise they would multiply the problem greatly compared to a dynamic speaker where tens of thousands of a second don't mean a lot.
You have a funny notion about what is going on with DSP, I have to say. What you are saying really doesn't make sense with reality. What does being smart enough and notes have to do with it? These are all just frequencies that get processed in time, notes or decay etc. has nothing to do with it. The DSP knows what frequencies it is sending to which driver and the signal that has been divided to go to that driver is delayed in time. That delayed signal is fed into the driver then based on a given delay. It is up to you to decide what is the right delay (unless the system has an automated way of doing this...many now do). If a signal is being fed to a passive speaker, then based on measurement, it can figure out which frequencies in time need shifting based on the group delay in the speaker then the SIGNAL is delayed based on the frequencies until they are arriving to within a certain limitation of the measurment hardware (probably a few microseconds). This then compensates for phase shift, and driver offset from the listening position...these are the two things that contribute to group delay.

Most drivers are not "in-phase" at the speaker other than at the crossover frequency so that they sum flat. However, a crossover is not an infinite slope (at least not in the passive world) and there is significant overlap that is not necessarily in-phase. This has nothing to do with beaming, although off-axis is usually significant contributor to mis-alignment of drivers in phase (and why off-axis often starts to look pretty strange with many speakers).

To claim these effects are not important or audible tells me you have never really tried the experiment of taking a non-coherent speaker and running software that will time alter the signal so that the output of the speaker is time aligned at the listening position. It is most definitely audible. Now, you might ask then why not just get any old speaker and correct it? Well, the correction does not compensate for non-linear distortions that have a big effect on the speaker character. What I am talking about is all things being equal, ie. the same speaker with and without digital time correction. The only problem I ever had with this is that I could hear the digital fingerprint that was stamped on the sound from the processor (you could hear this also in the uncorrected when a flat curve (ie. DSP engaged with no actual correction) was applied...again to make apples to apples comparisons.
 
Likes: audioquattr

Robh3606

Active Member
Aug 25, 2010
1,000
20
38
Destiny
'Well, we are talking about horns here, and most of the horn companies give importance to alignment. It has caught up with horn manufacturers.'


Well that may be true but unless they are using a DSP based active crossover there is no way they can be time aligned with the different path lengths in the horns themselves. You can physically offset them but that really doesn't solve the problem. You have to realize that virtually all of the classic horn systems are not even close to being time aligned. Any modern horn that uses a passive crossover is typically not as well. I really don't worry about it as long as it is under the B/L criteria for the audibility Group Delay.

Rob:)
 
Last edited:
Sep 7, 2016
250
260
63
Netherlands
....
They prove this how? Do they make a version of the speaker available that is with time-alignment vs. one without? No, they do not. Many designers simply don't think it is important and their agenda is different like controlling frequency response on and off-axis, reducing cabinet and driver distortions or whatever. They think because the ear/brain works as an integrator that it will take the scrambled mess that comes out of most speakers and make sense of it...and to a large extent they are right. However, once one becomes accustomed to a time coherent speaker and how it allows an easier time with listening, then one often doesn't want to go back to a non-coherent speaker. Not everyone is sensitive to it but if you are then you hear the lack of coherence.
I not allways agree with you but in this topic and this reply i do.
 
Last edited:
Sep 7, 2016
250
260
63
Netherlands
[QUOTE="morricab, post: 541452, member: 5671"

To claim these effects are not important or audible tells me you have never really tried the experiment of taking a non-coherent speaker and running software that will time alter the signal so that the output of the speaker is time aligned at the listening position. It is most definitely audible. Now, you might ask then why not just get any old speaker and correct it? Well, the correction does not compensate for non-linear distortions that have a big effect on the speaker character. What I am talking about is all things being equal, ie. the same speaker with and without digital time correction. The only problem I ever had with this is that I could hear the digital fingerprint that was stamped on the sound from the processor (you could hear this also in the uncorrected when a flat curve (ie. DSP engaged with no actual correction) was applied...again to make apples to apples comparisons.[/QUOTE]
I also have tried/experiment a lot with delays and corrections with dsp . And meassuring everything , i did and still can learn a lot.
Imo aligning does a lot , not everything for fr but for soundstage and first atack of cymbal, piano or string it does a lot.
But physical align i prefer much much more than dsp.
I can’t listen to dsp in my big system only in my office system and than it’s only good (imo) under about 150hz or under room-Schroeder frequency. So for me dsp only for testing/learning and for subs.
 

Audiophile Bill

Well-Known Member
Mar 23, 2015
2,313
689
113
UK
[QUOTE="morricab, post: 541452, member: 5671"

To claim these effects are not important or audible tells me you have never really tried the experiment of taking a non-coherent speaker and running software that will time alter the signal so that the output of the speaker is time aligned at the listening position. It is most definitely audible. Now, you might ask then why not just get any old speaker and correct it? Well, the correction does not compensate for non-linear distortions that have a big effect on the speaker character. What I am talking about is all things being equal, ie. the same speaker with and without digital time correction. The only problem I ever had with this is that I could hear the digital fingerprint that was stamped on the sound from the processor (you could hear this also in the uncorrected when a flat curve (ie. DSP engaged with no actual correction) was applied...again to make apples to apples comparisons.
I also have tried/experiment a lot with delays and corrections with dsp . And meassuring everything , i did and still can learn a lot.
Imo aligning does a lot , not everything for fr but for soundstage and first atack of cymbal, piano or string it does a lot.
But physical align i prefer much much more than dsp.
I can’t listen to dsp in my big system only in my office system and than it’s only good (imo) under about 150hz or under room-Schroeder frequency. So for me dsp only for testing/learning and for subs.[/QUOTE]

Yes I personally can’t fathom why someone would choose a dsp based crossover if musical pleasure is the goal. I never heard one that sounded natural and with life / vibrancy. The dsp has the uncanny ability to rob the music of the life and soul. Ymmv but imho.
 

kodomo

Well-Known Member
Apr 26, 2017
352
270
63
'

Well that may be true but unless they are using a DSP based active crossover there is no way they can be time aligned with the different path lengths in the horns themselves. You can physically offset them but that really doesn't solve the problem. You have to realize that virtually all of the classic horn systems are not even close to being time aligned. Any modern horn that uses a passive crossover is typically not as well. I really don't worry about it as long as it is under the B/L criteria for the audibility Group Delay.

Rob:)
No they do not have to do it via dsp, it can be a multiway, different path length horns and aligned with a passive crossover like my system. I have explained how to do it before so I won't repeat again.

Yes, the old systems were not aligned but there are people now aligning them via dsp (for ex. hornsolutions) and saying they have a good result. People like Yamamura also did dsp based horn systems and I remember Gian writing that system was the most life like. For hornsolutions dsp based we system, Sven likes it and told me so does Thomas Schick and some other people who have heard it...
 
Likes: christoph

Robh3606

Active Member
Aug 25, 2010
1,000
20
38
Destiny
"No they do not have to do it via dsp, it can be a multiway, different path length horns and aligned with a passive crossover like my system. I have explained how to do it before so I won't repeat again."

You have optimized it for your listening position and that's about all you can do. Even with DSP it's a one point in space solution. You cannot adjust out the physical distance between the horns in 3 dimensions. Even "time aligned" systems are aligned on their listening axis typically the tweeter as soon as you vary height it goes out the window because the physical offsets change between the listener and the individual drivers. Any doubt about it take a look at your step response , impulse doesn't cut it, at different heights and listening positions in your room as you undoubtedly know.

DSP can do it better you are limited using passive networks.

Rob:)
 

Attachments

kodomo

Well-Known Member
Apr 26, 2017
352
270
63
"No they do not have to do it via dsp, it can be a multiway, different path length horns and aligned with a passive crossover like my system. I have explained how to do it before so I won't repeat again."

You have optimized it for your listening position and that's about all you can do. Even with DSP it's a one point in space solution. You cannot adjust out the physical distance between the horns in 3 dimensions. Even "time aligned" systems are aligned on their listening axis typically the tweeter as soon as you vary height it goes out the window because the physical offsets change between the listener and the individual drivers. Any doubt about it take a look at your step response , impulse doesn't cut it, at different heights and listening positions in your room as you undoubtedly know.

DSP can do it better you are limited using passive networks.

Rob:)
Varying height has the worse effects, varying depth does not sound too bad... At your listening point though, it is great and that is what I was aiming for. What can dsp offer more than this, I really won't get into what dsp takes away, what does it offer? A better step response, how, I genuinely would like to learn.
 
May 30, 2010
15,309
595
113
Portugal
Apr 29, 2018
51
21
8
Isn't this whole hobby driven by insensible decisions? ;)
I could visualize them in our living room, but then again the Universums are really extremely good :cool:
How much is the retail price of the Pnoes anyways? :oops:
How good are the Universum ?
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. A place where audiophiles and audio companies discuss existing and new audio products, music servers, music streamers and computer audio, digital to audio convertors (DACS), turntables, phono stages, cartridges, reel to reel, speakers, headphones, tube amplifiers and solid state amplification. Founded in 2010 What's Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals we enjoy learning about new things and meeting new people and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing