"Aggressive". Fair Criticism of Some Horn Speakers and Designs? If So, What Causes it?

caesar

Well-Known Member
May 31, 2010
3,550
273
538
Is it a fair criticism? For which models and brands?

And if so, is it the speaker design or the amps interacting with the speaker? Or something else? horns.jpg
 
Last edited by a moderator:

DaveC

Industry Expert
Nov 16, 2014
3,296
1,240
390
Part of it is certainly the how the horn effectively couples the diaphragm to the air, this is good... other times it can be metallic overtones caused by some kinds of diaphragms, this is not so good unless you're only using the driver for cymbals. ;)
 

morricab

Well-Known Member
Apr 25, 2014
4,748
1,277
388
Switzerland
Is it a fair criticism? For which models and brands?

And if so, is it the speaker design or the amps interacting with the speaker? Or something else?
Most SS need not apply...

The driver, diaphragm material and mating to the horn can result in distortion. Also, the horn profile itself can strongly influence the "horn" sound or if it sounds invisible. Modern profiles tend to minimize the resonances that result in unpleasant sounds. The material of the horn matters as well...wood seems to work very well, plastic is not bad and metal seems to need a fair bit of damping.

THere are so many variables that it is a bit of art mixed with science to get a transparent horn.
 

Duke LeJeune

[Industry Expert]/Member Sponsor
Jul 22, 2013
619
777
340
Princeton, Texas
I hear "aggressiveness" (or coloration or edginess or shoutiness or honk) in SOME horn loudspeakers.

Possible causes can be the internal shape of the horn itself causing a lot of internal reflections; the internal shape introducing a lot of diffraction which the ear interprets as "harshness" at high SPLs; the horn mouth itself (a sharp-edged mouth tends to cause coloration); a poor driver/horn combination; an inadequate crossover (sometimes in the name of "purity" an insufficiently capable highpass filter is used); and ringing in the material the horn is made from.

The compression driver itself can have issues which cause "aggressiveness". Some have significant resonances; some have underdamped break-up modes in the top octave or so; and some have features which trade off one characteristic for another, which may work well for some applications but not for others.

A narrow-pattern horn can result in an unusually high direct-to-reverberant sound ratio, which can result in a rather "dry" and/or "forward" presentation even if the horn itself isn't doing anything "wrong". Imo this falls more under the category of personal preference, but I suppose some might call it "aggressiveness".

And sometimes the horn and compression driver are both perfectly innocent yet are being blamed for a coloration which arises elsewhere in the loudspeaker system. I recall being asked to analyze a particular horn speaker which had a "cupped hands" type of coloration and the problem turned out to be a nasty cone break-up in the woofer at the top end of its range.

Imo it should go without saying that if the speaker manufacturer recommends a particular type of amplifier, there's probably a reason for it. The high efficiency of many horn speakers makes them potentially a good match for specialty tube amps, whose behavior into a varying impedance curve is very different from that of most solid state amps, and the speaker designer MAY have sought to capitalize on that. Also, the efficiency of horns can throw a spotlight on an amplifier's characteristics at very low power levels and on its noise floor, which probably would not have been problematic areas with medium-efficiency speakers.

Imo there are things good horns do well which make them worth the trouble.
 
Last edited:

DaveC

Industry Expert
Nov 16, 2014
3,296
1,240
390
I hear "aggressiveness" (or coloration or edginess or shoutiness or honk) in SOME horn loudspeakers.

I think we need the OP to better define "aggressiveness". I was thinking he was talking about the more dynamic and forward sound of horns in general, or colorations that accentuate that particular trait. In which case all horns would exhibit this to some degree because of efficiency and polar pattern, but that's a good thing... so I was thinking part of the issue is a positive trait, but other parts could be seen as a negative.
 

morricab

Well-Known Member
Apr 25, 2014
4,748
1,277
388
Switzerland
I think we need the OP to better define "aggressiveness". I was thinking he was talking about the more dynamic and forward sound of horns in general, or colorations that accentuate that particular trait. In which case all horns would exhibit this to some degree because of efficiency and polar pattern, but that's a good thing... so I was thinking part of the issue is a positive trait, but other parts could be seen as a negative.
A horn that cannot portray depth when it is on the recording is creating some kind of distortion that is preventing the proper development of space perception.
 

Audiophile Bill

Well-Known Member
Mar 23, 2015
3,134
1,826
420
UK
No Problem with that here....
View attachment 71409

Hi Arno,

Interesting speaker. Please can you tell us some more. Looks to be 3 way with back loaded Fostex driven horn, a tractrix upper mid and AMT ribbon tweeter? What crossover points etc?

Best.
 
  • Like
Reactions: christoph

ArnoFenn

Member
Oct 28, 2020
24
37
15
Hi Bill, you got a sharp eye I must say!

After many many years of developing normal dynamic loudspeakers, ESL (10 years) Horns were my ultimate goal, driven by 2 300B SET (all done via intensive listening, measuring...and thinking ;-) ). The main horn started as a 1-way (Lowther) with a simpel shelving filter to flatten the mids. The horn size had to be limited to a 40Hz design (and not stretching to 20Hz as I aimed for years before). After having the optimum and listening a long period, the curiosity hit me. If the main horn sounded that good and low on distortion...what would it do for the mids. I always enjoyed the low distortion of my ESL's so for me that was the way forward, simplicity, low moving mass, phase etc. Having said that, bougth some cheap but good second hand mid-horn (Fostex), installed, filter...and there was no way back. AMAZING! But work to do. What would be the best possible horn-mid (for me that is)-> ended up with a TD2001 driver (could have been larger than 1" but, I wanted to have the low mass...and Be diafragm). The last bit was the search for a tweeter (if any), so had the T900A fostex, JBL (slot-type) and it just added the "ambience" on top of the TD2001. BUT, thinking back to the low mass ESL foils......Ribbon is comparable without the struggle of keeping the ESL foil coating in top-condition. So I looked for the best and also for the best designer/knowledge...ended up with RAAL (lazy Ribbon). A good tweeter is not even "noticed". The RAAL is mounted on an adjustable slide for phase alignment wrt the TD2001. Unfortunately there was no (WAF) option to also phase align the low with Mid/Hi.
For the filtering there is a lot of going back and forth with measurements and listening to a lot of different music and keeping things simple. Theory vs reality. In the mean-time I could not reduce the mid-cone of the lowther without higher order filtering and the Fostex208 had a better response curve (no mid-cone) with comparable T&S parameters. The TD2001 started with a 3rd order high-pass and a second order low-pass, first of all to avoid overload in the lower F. Cross-over from low-mid: approx 800-900Hz. Mid-Hi: 8-10kHz. Low, low-pass second order.
AB switching with filter components was an eye-opener, blind tested with several people. Difference between capacitor types or inductors air-wire vs ribbon foil in wax etc.
Measurements ok, sound good...but was it the best?

Striving to "as simple as possible", looked for options to strip down the filtering and there are some tricks that are not easily to explain in text here that are not realy standard in normal designs. One of the things that made a big difference is to see if the TD2001 could pass at 600-700Hz with a lower order filter and looking at the specs it is possible IF the max power curve is not exceeded...so I did and it worked out great. The low is single order filtered now and so is the tweeter.

All units are aligned to the listening position with a laser by the way.

Now, how perfect was this. Not there yet. Some recordings sound too much in the mids (and partly this might relate to the actual topic) Measurements looked good...flatlining here at the listening position but this was not the correct approach. A flat line measured in an sound-proofed chamber is a good start but when putting that speaker in the living room there will be reflections that will "help" mids/highs. With this help on an already flat curve...well...it might be too much. Measurements in room, aim for 6-8dB drop-off from low to high frequencies. Modified a bit on adjusting mid/high selector switch and now I must say it's great!

What's next? Who knows, but I usually find another step ;-)

Feel free to ask any questions and I 'll be glad to answer them.
 

Audiophile Bill

Well-Known Member
Mar 23, 2015
3,134
1,826
420
UK
Hi Bill, you got a sharp eye I must say!

After many many years of developing normal dynamic loudspeakers, ESL (10 years) Horns were my ultimate goal, driven by 2 300B SET (all done via intensive listening, measuring...and thinking ;-) ). The main horn started as a 1-way (Lowther) with a simpel shelving filter to flatten the mids. The horn size had to be limited to a 40Hz design (and not stretching to 20Hz as I aimed for years before). After having the optimum and listening a long period, the curiosity hit me. If the main horn sounded that good and low on distortion...what would it do for the mids. I always enjoyed the low distortion of my ESL's so for me that was the way forward, simplicity, low moving mass, phase etc. Having said that, bougth some cheap but good second hand mid-horn (Fostex), installed, filter...and there was no way back. AMAZING! But work to do. What would be the best possible horn-mid (for me that is)-> ended up with a TD2001 driver (could have been larger than 1" but, I wanted to have the low mass...and Be diafragm). The last bit was the search for a tweeter (if any), so had the T900A fostex, JBL (slot-type) and it just added the "ambience" on top of the TD2001. BUT, thinking back to the low mass ESL foils......Ribbon is comparable without the struggle of keeping the ESL foil coating in top-condition. So I looked for the best and also for the best designer/knowledge...ended up with RAAL (lazy Ribbon). A good tweeter is not even "noticed". The RAAL is mounted on an adjustable slide for phase alignment wrt the TD2001. Unfortunately there was no (WAF) option to also phase align the low with Mid/Hi.
For the filtering there is a lot of going back and forth with measurements and listening to a lot of different music and keeping things simple. Theory vs reality. In the mean-time I could not reduce the mid-cone of the lowther without higher order filtering and the Fostex208 had a better response curve (no mid-cone) with comparable T&S parameters. The TD2001 started with a 3rd order high-pass and a second order low-pass, first of all to avoid overload in the lower F. Cross-over from low-mid: approx 800-900Hz. Mid-Hi: 8-10kHz. Low, low-pass second order.
AB switching with filter components was an eye-opener, blind tested with several people. Difference between capacitor types or inductors air-wire vs ribbon foil in wax etc.
Measurements ok, sound good...but was it the best?

Striving to "as simple as possible", looked for options to strip down the filtering and there are some tricks that are not easily to explain in text here that are not realy standard in normal designs. One of the things that made a big difference is to see if the TD2001 could pass at 600-700Hz with a lower order filter and looking at the specs it is possible IF the max power curve is not exceeded...so I did and it worked out great. The low is single order filtered now and so is the tweeter.

All units are aligned to the listening position with a laser by the way.

Now, how perfect was this. Not there yet. Some recordings sound too much in the mids (and partly this might relate to the actual topic) Measurements looked good...flatlining here at the listening position but this was not the correct approach. A flat line measured in an sound-proofed chamber is a good start but when putting that speaker in the living room there will be reflections that will "help" mids/highs. With this help on an already flat curve...well...it might be too much. Measurements in room, aim for 6-8dB drop-off from low to high frequencies. Modified a bit on adjusting mid/high selector switch and now I must say it's great!

What's next? Who knows, but I usually find another step ;-)

Feel free to ask any questions and I 'll be glad to answer them.

Thanks Arno. It is a very clever and well thought out solution I must say. Are you time aligning the back loaded horn with the mid and RAAL at all? If not I wondered how that was working out. Aries Cerat is similar solution from what I see in this respect - back loaded Fostex + mid + tweeter. I even think the Aries uses a ribbon in the tweeter horn.
Did you build the back loaded horn itself? Is there plans for this or did you design it?

Best wishes and very pleased you posted here and shared your system.
 

morricab

Well-Known Member
Apr 25, 2014
4,748
1,277
388
Switzerland
Thanks Arno. It is a very clever and well thought out solution I must say. Are you time aligning the back loaded horn with the mid and RAAL at all? If not I wondered how that was working out. Aries Cerat is similar solution from what I see in this respect - back loaded Fostex + mid + tweeter. I even think the Aries uses a ribbon in the tweeter horn.
Did you build the back loaded horn itself? Is there plans for this or did you design it?

Best wishes and very pleased you posted here and shared your system.
Yes, that’s right Bill. Stavros also time aligns the drivers and the backloaded horn is good to 30hz.
 
  • Like
Reactions: Audiophile Bill

Audiophile Bill

Well-Known Member
Mar 23, 2015
3,134
1,826
420
UK
Yes, that’s right Bill. Stavros also time aligns the drivers and the backloaded horn is good to 30hz.

Thanks Brad. How does Stavros time align the back loaded horn, which must be 3+ metres long?
 

ArnoFenn

Member
Oct 28, 2020
24
37
15
Thanks Arno. It is a very clever and well thought out solution I must say. Are you time aligning the back loaded horn with the mid and RAAL at all? If not I wondered how that was working out. Aries Cerat is similar solution from what I see in this respect - back loaded Fostex + mid + tweeter. I even think the Aries uses a ribbon in the tweeter horn.
Did you build the back loaded horn itself? Is there plans for this or did you design it?

Best wishes and very pleased you posted here and shared your system.
Hi Bill, there is no time-alignment between mid/low, but then again this part is not the most critical since the edge of an impulse is vastly made by the mid/highs. The BLH is home-made and it started with reading a lot of books, first calculations and aiming for the best size vs lowF trade-off....and Tractrix is a good choice. After that I used AJHorn software (bought, cheap but effective), played around a bit and in the end printed the 3.5m straight horn profile. Measured the ceiling height and figured out how the long horn should be cut and bended to fit in. Glued that design on card-board and cut the parts out. Glue and stuff to shape the mini proto (10cm high) to get the feel on how to build it in real size and how wood had to be processed etc. After that.....do it ;-) Test in, measure and since it was correlating to what I wanted....start to finish...2 wooden Ikea salad bowls for the housing etc etc
 

bonzo75

Member Sponsor
Feb 26, 2014
15,414
4,905
940
London
Thanks Arno. It is a very clever and well thought out solution I must say. Are you time aligning the back loaded horn with the mid and RAAL at all? If not I wondered how that was working out. Aries Cerat is similar solution from what I see in this respect - back loaded Fostex + mid + tweeter. I even think the Aries uses a ribbon in the tweeter horn.
Did you build the back loaded horn itself? Is there plans for this or did you design it?

Best wishes and very pleased you posted here and shared your system.

Yes Stavros uses a ribbon tweeter. Steve in UK uses the RAAL ribbon tweeter on a vitavox S2

Arno, the question I have is that 2001 is often used as a tweeter itself, and has great high end extension over 20k. I have heard it in a two way. What do you get or lose, if you keep it only to the 2001 vs adding a ribbon on top?
 

ArnoFenn

Member
Oct 28, 2020
24
37
15
Yes Stavros uses a ribbon tweeter. Steve in UK uses the RAAL ribbon tweeter on a vitavox S2

Arno, the question I have is that 2001 is often used as a tweeter itself, and has great high end extension over 20k. I have heard it in a two way. What do you get or lose, if you keep it only to the 2001 vs adding a ribbon on top?
Indeed, the 2001 is a good high F performer and the trick is that even though it sounds really good with only high pass 2001 filtering to blend in the actual tweeter and filter back the highest part of the 2001 since after all the higher the F the lower the moving mass has to be and here the very light Be diafragm can't win from the ribbon foil.

Reminding me of the actual opening topic/question. Usually I experience loudspeakers to sound aggressive when they distort (driven too hard, wrong cross-over etc). To put it the other way, ESL have a very low distortion and in cases of very low distortion the listener is tempted to easily increase the volume (apparently we push that brake when we start perceiving some level op distortion (THD/TIM/Phase)). Goin deaf is the easiest with a very low distortion set up ;-)
 

ArnoFenn

Member
Oct 28, 2020
24
37
15
Honestly...not sure how it is done...
Basically a good start is to get transducers lined up as good as possible but the actual finetuning needs to be done since also the exact acoustic center might be different and the filters can introduce additional shifts. What I do for the last step is to run clio pocket in a cycled measurement while observing the impulse response adjusting the tweeter position wrt mid-horn.
By the way the low speaker impuls-bump is also visible during that but I can't move this.
 

Audiophile Bill

Well-Known Member
Mar 23, 2015
3,134
1,826
420
UK
Basically a good start is to get transducers lined up as good as possible but the actual finetuning needs to be done since also the exact acoustic center might be different and the filters can introduce additional shifts. What I do for the last step is to run clio pocket in a cycled measurement while observing the impulse response adjusting the tweeter position wrt mid-horn.
By the way the low speaker impuls-bump is also visible during that but I can't move this.

Hi Arno,

This makes sense but my query was how Aries time align the back loaded 3m+ Fostex back loaded horn with their mid/tweeter. It can’t be physically time aligned (or else the mid / tweeter would need to be physically separated from the back loaded horn cabinet whereas they are attached and juxtaposed so I wondered whether they were achieving it electrically with the filter.
 

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