Horn Speakers - Is it me or.......

[Ktacos]

This is the speaker I built, anything in the data giveaway that it will sound like a horn?

MTG Designs - VBS-10.2

Design description and goals: Bigger Brother to the VBS-6.2 this is another design that delivers a very high level of bang for your buck. A 10" 2-way with a horn loaded compression driver that does not sacrifice sound quality in its ability to deliver high output for very reasonable costs. Ideally

www.mtg-designs.com

 

[Sal1950]

If you're ever in the Dallas-Fort Worth area, drop in for a listen.

Thanks for the invite, I just might do that.
Belle Klipsch surrounds, got to love it.

 

[Hear Here]

So there is no "sweet spot" that you talk about, other than the limitations brought by stereo itself, and which is completely mitigated by a 5.1 array setup, in which you can stand, sit, and walk around and hear the musicians playing as if they were playing live in-room:

Not sure what you are getting at when talking about "no sweet spot". Horn speakers, when properly set up, offer a smaller sweet spot (best listening position) than other types, but it is "sweeter" in that the imaging (you may call it the stereo effect) is far greater with a good horn than most other speakers. The imaging of my speakers is better than any other system I have heard, closely followed by a friend's excellently set up Quad electrostatic system.

Your description of "walking around the room and hearing musicians playing" isn't a feature of horns but of omni-directional speakers in a 2-channel environment. I have recently considered omnis and visited dealers for demos of the 2 top omni brands. Good as omnis may be in delivering good sound to any listening position in the room, their imaging ability is dire compared with horns for the person in their sweet spot. Despite omnis being in theory the best type for my room's characteristics (semi-circular with centrally positioned speakers with main listening position in front, but dining and kitchen areas behind the speakers), there's no way I'd sacrifice the imaging benefits of horns in favour of the advantages of omnis. Adding more speakers isn't the answer for music, though it may be audio-visual / home theatre applications - but of no interest to me!

It's the type of horns used in your mentioned loudspeaker brand that are of the "old design" and that have serious polar coverage issues. I certainly wouldn't use those as a reference in terms of sound quality in-room.

Whether you classify Avantgarde speakers as "old design" or not, their performance is about as good as it can get. If AG could improve on their speakers, they would - and indeed they do on a roughly 5 year basis. Their latest range has the same well-regarded and successful mid horn (though the driver has been upgraded), but the top range horn has been significantly redesigned for this latest series with a deeper horn and again an upgraded driver. The bass drivers (not horn loaded in the Duo GT) are still twin 12" but the voice coil is significantly larger in diameter and this reduces distortion even further than the XD Duos that I have.

If you're ever in the Dallas-Fort Worth area, drop in for a listen.

Sadly, being 4729 miles from me, your kind invitation is unlikely to be taken advantage of. Likewise, if you'd like to hear horns of very high quality and pretty well set up (I'm not as expert at this as some), you'd be welcome to visit Portsmouth next time you're in the area!

PS - Looking at your room, I wouldn't recommend Avantagrdes. AG's need far more space to perform at their best.

 

[hex168]

This is the speaker I built, anything in the data giveaway that it will sound like a horn?

MTG Designs - VBS-10.2

Design description and goals: Bigger Brother to the VBS-6.2 this is another design that delivers a very high level of bang for your buck. A 10" 2-way with a horn loaded compression driver that does not sacrifice sound quality in its ability to deliver high output for very reasonable costs. Ideally

www.mtg-designs.com

Might want to try some 30PPI reticulated foam completely filling the horns (should extend slightly past the mouth) to reduce what Earl Geddes called Higher Order Modes:

https://www.swisstropicals.com/filtration-shop/poret-foam-shop/

It cuts cleanly with a hot wire or an electric carving knife.

Minor crossover adjustment probably needed, generally around +2 dB at 10K. But try it without first to see if it is along the right track.

An interesting thread for background:

The HOMster! (or How I Learned How to Fix a Horn)

There's been a great deal of discussion about diffraction and higher order modes on diyaudio lately. In a nutshell, it appears that diffraction may be more offensive than harmonic distortion. At the very least, diffraction muddies the soundstage by creating unnatural soundstage cues. I have...

www.diyaudio.com

 

[Purité Audio]

The Danley ( formerly known as) Hyperions are the only truly uncoloured horns I have heard.
Keith

 

[Chris A]

The Danley ( formerly known as) Hyperions are the only truly uncoloured horns I have heard.

As I understand it, those are actually SH-60s on top of custom direct radiating woofers and integral DSP amplifiers. The horns are typical Danley MEH style with dual flare (straight-sided) walls. Except for the hard chine break at the secondary flare, they perform similar to the HF horns in the front three of my setup.

Horn speakers, when properly set up, offer a smaller sweet spot (best listening position) than other types...

Perhaps that's a limitation of your setup (and I can certainly see why that is). It certainly isn't in the setups that Bjorn and I are talking about. The soundstage is the size of the entire wall, and you can move about freely without affecting anything except the "phantom stereo center" imaging, which of course is a limitation of stereo itself.

Go listen to a pair of well-set-up Klipsch Heritage Jubilees or a Danley Synergy home hi-fi set up (at least a pair of SH-96s or perhaps SH-60s/HRE1's). You will hear what I'm talking about. (Sorry about how much you had to spend on those Avantgardes.)

Chris

 

[Chris A]

Thanks for the invite, I just might do that.
Belle Klipsch surrounds, got to love it.

I'm about a half hour drive from DFW International--in far west Arlington by the lake. Note that the Belles are modified to bypass the tweeter/midrange top hats, in favor of bi-amping using dual-stacked AMT-1s with a miniDSP 2x4 HD, dialed in. It actually was a pretty big step up in terms of sound quality.

 

[Sal1950]

The Danley ( formerly known as) Hyperions are the only truly uncoloured horns I have heard.
Keith

Well either you may have missed hearing things like JBL Synthesis line of horns such as the M2, etc.
Of you have simply fallen back to the fact there are no truly uncolored speakers in the universe.
They all have some imperfections.
BTW, I corrected the spelling of "uncolored" for you. LOL

 

[Chris A]

I believe that JBL did some fairly extensive BEM work 15-20 years ago to arrive at their answer to the "new controlled directivity" horn efforts, and quite successfully, like the HF horn in the M2. JBL's solution gets around a Klipsch design by having those deeper creases in the corners of their rectangular horns that sport straight-sided design influences.

Klipsch patented the "mumps" modification in ~2010 to eliminate polar waistbanding problems at lower frequencies in their smaller K-510 horn design--about an octave above effective fc. (The larger K-402 horn design does not experience polar waistbanding.)

Chris

 

[Duke]

This is the speaker I built, anything in the data giveaway that it will sound like a horn?

MTG Designs - VBS-10.2

Design description and goals: Bigger Brother to the VBS-6.2 this is another design that delivers a very high level of bang for your buck. A 10" 2-way with a horn loaded compression driver that does not sacrifice sound quality in its ability to deliver high output for very reasonable costs. Ideally

www.mtg-designs.com

Nothing in the measurements stands out to me as throwing up a red flag.

It might be a resonance in the plastic body of the horn, which you can address by putting Dynamat on the back side of the horn.

Another possibility is this: There might be a significant mis-match between the exit angle of the compression driver and the entry angle of the horn. This discontinuity will cause an undesirable reflection, part of which will propagate back down into the compression driver and bounce off the diaphragm. This bounced-back-off-the-diaphragm energy will then be reflected yet again when it encounters the discontinuity on its way out. So some of this reflection sort of ping-pongs between the diaphragm and the discontinuity as it decays. This kind of problem causes too small of a frequency response aberration to make its presence obvious on a frequency response graph, but the ear notices it because it's a distortion that happens later in time than the main signal (and therefore the ear/brain system's masking characteristic cannot mask its presence). The ear tends to interpret this kind of distortion as "harshness", and it becomes increasingly audible and objectionable as the volume level is turned up.

A few years ago a client hired me to design a crossover for that horn and a compression driver which they specified. The end result met their measurement criteria but (to my ears) did not sound as good as the competition they were trying to beat. I later became aware of a significant mis-match between the exit angle of the specified compression driver and the entry angle of that horn. I cannot say for sure this was the issue, but to my ears (and evidently to theirs as well, as it never went into production) that particular compression driver and horn combination failed to sound as good as the measurements led me to expect.

Ime that particular horn works well when used with a screw-on compression driver and an adaptor, but that's a rather complicated and counter-intuitive solution, and would require a crossover re-design.

 

[Hear Here]

BTW, I corrected the spelling of "uncolored" for you. LOL

No - Uncoloured is the correct spelling in the civilised (not civilized, of course) world!

 

[Newman]

Another possibility is this: There might be a significant mis-match between the exit angle of the compression driver and the entry angle of the horn. This discontinuity will cause an undesirable reflection, part of which will propagate back down into the compression driver and bounce off the diaphragm. This bounced-back-off-the-diaphragm energy will then be reflected yet again when it encounters the discontinuity on its way out. So some of this reflection sort of ping-pongs between the diaphragm and the discontinuity as it decays.

I fully agree that it is a mechanism. But see further down for my view on its relevance as an explanation.

This kind of problem causes too small of a frequency response aberration to make its presence obvious on a frequency response graph,

As Toole would say, the FR plot needs to be of the appropriate resolution, and all will be revealed.

but the ear notices it because it's a distortion that happens later in time than the main signal (and therefore the ear/brain system's masking characteristic cannot mask its presence). The ear tends to interpret this kind of distortion as "harshness", and it becomes increasingly audible and objectionable as the volume level is turned up.

One would expect the exact same mechanism on non-horn speakers, with even longer time components, eg at the lip of a cone or the mounting plate of a dome.

Back in the days of long-throat compression drivers, they had a standardised flare rate, IIRC 18 degrees, to which bolt-on horns would match. It was important because the mechanism you described can screw things up. I'm not sure if that standard has been maintained into the modern era, because I don't think it is relevant with a modern ultra-short-throat horn, because the wavelength of the resonance would be above 20 kHz. Indeed, that is one of the advantages of an ultra short throat.

I remember attending lectures about 20 years ago at my local chapter of the AES, given by the Chief of Design at a commercial pro audio horn loudspeaker manufacturer, where he would throw onto the screen cutaway diagrams of commercial compression drivers, and show all the little pathways that the sound waves can take into various nooks and crannies in the driver housing. He would then show his modelling of the acoustics of the driver, and the effect of these pathways on the frequency response. He would then show measurements confirming this. It was always FR plots that showed it up, one just had to know where to look (which is where the modelling comes into its own and adds value).

When I combine the fact that non-horn speakers have the same mechanism, plus the fact that a horn has to have design issues for it to occur, I concur with you that any one horn might have it, but I don't think that this particular mechanism would explain a 'universal harsh horn sound'. (Which I don't think it true today, BTW)

cheers

 

[Duke]

I don't think that this particular mechanism would explain a 'universal harsh horn sound'. (Which I don't think it true today, BTW)

Neither do I, and if that's the impression I gave, then I communicated poorly.

I remember attending lectures about 20 years ago at my local chapter of the AES, given by the Chief of Design at a commercial pro audio horn loudspeaker manufacturer, where he would throw onto the screen cutaway diagrams of commercial compression drivers, and show all the little pathways that the sound waves can take into various nooks and crannies in the driver housing. He would then show his modelling of the acoustics of the driver, and the effect of these pathways on the frequency response. He would then show measurements confirming this. It was always FR plots that showed it up, one just had to know where to look (which is where the modelling comes into its own and adds value).

Very interesting! I wish I could have seen that lecture. Do you recall who gave it? I'd like to see if there are any articles by him on the topic online.

Back in the days of long-throat compression drivers, they had a standardised flare rate, IIRC 18 degrees, to which bolt-on horns would match. It was important because the mechanism you described can screw things up.

I didn't know that it used to be standardized. That was very smart. The only large-format oldschool compression driver I've worked with was the TAD 4001, which has an internal throat length of about 75 mm.

I'm not sure if that standard has been maintained into the modern era,

It has not. I've seen throat exit angles ranging from negative (converging) 21 degrees to positive (diverging) 34.5 degrees.

I don't think [the exit angle] is relevant with a modern ultra-short-throat horn, because the wavelength of the resonance would be above 20 kHz. Indeed, that is one of the advantages of an ultra short throat.

The reflection path for a discontinuity at the junction of the compression driver's exit and the horn's entrance is from there to the surface of the diaphragm, so it is considerably longer than the throat because its path length includes the phase plug.

I am of the opinion that it matters, but my life would be a bit easier if I didn't think so; I'm actually in the middle of designing a horn whose entrance angle will be an exact match to a particular compression driver's exit angle.

Some modern compression drivers have the end of the phase plug coincide with the driver's exit, this is called a "combined exit". In effect there is no throat, or at least no throat length.

 

[Sal1950]

No - Uncoloured is the correct spelling in the civilised (not civilized, of course) world!

Not according to my spell checker, both yours are wrong. Shame you missed proper English spelling.

 

[Hear Here]

Not according to my spell checker, both yours are wrong. Shame you missed proper English spelling.

View attachment 388129

Ha Ha! English is the language developed and used by the English for centuries - before America was even "discovered"! America uses "American", a language akin to English - but with many differences. An English spell checker will underline color and civilize - because they are not correct English spellings.

 

[ahofer]

Ha Ha! English is the language developed and used by the English for centuries - before America was even "discovered"! America uses "American", a language akin to English - but with many differences. An English spell checker will underline color and civilize - because they are not correct English spellings.

Two nations divided by a common language.

 

[Tim Link]

This is the speaker I built, anything in the data giveaway that it will sound like a horn?

MTG Designs - VBS-10.2

Design description and goals: Bigger Brother to the VBS-6.2 this is another design that delivers a very high level of bang for your buck. A 10" 2-way with a horn loaded compression driver that does not sacrifice sound quality in its ability to deliver high output for very reasonable costs. Ideally

www.mtg-designs.com

My best guess is it probably has something to do with the directivity. I love horns, especially ones with tighter directivity control. But, they do have a sort of cupped sound even to me, especially when I first turn them on. After a few minutes I adjust comfortably. Wide dispersion speakers sound very open to me at first, but then confused and fatiguing within about 20 minutes. Sometimes sooner. I seem to have an unstable perception of wide dispersion in listening rooms. Depending on my brain's state I might think it sounds really good, or not so good.

 

[AudioJester]

My best guess is it probably has something to do with the directivity. I love horns, especially ones with tighter directivity control. But, they do have a sort of cupped sound even to me, especially when I first turn them on. After a few minutes I adjust comfortably. Wide dispersion speakers sound very open to me at first, but then confused and fatiguing within about 20 minutes. Sometimes sooner. I seem to have an unstable perception of wide dispersion in listening rooms. Depending on my brain's state I might think it sounds really good, or not so good.

Yeah, we need more speakers where we can adjust/change directivity to match mood.

 

[Tim Link]

Yeah, we need more speakers where we can adjust/change directivity to match mood.

My setup right now can do something sort of like that. I'm running a left, center, and right tweeter right now from 900Hz and up. Depending on what signals I mix into the side tweeters, and time delay on the center tweeter, I can reinforce or partially cancel the direct sound that reaches my ears, which changes the direct to reflected ratio significantly because not only is the direct quieter when I do that, but the overall energy in the room is actually higher. This only applies to above 900 Hz. The effect is interesting, and I think pretty useful. I'm finding myself changing the settings to best match whatever I'm listening to. I've been trying to make an adjustable slider that lets me change the direct to reflected ratio as I'm listening. It's sort of working, but I've got something wrong with the software routing because the image is overall wider and better when I directly re-route the signals vs trying to use an adjustable bypass block to slide the effect one way or the other.

 

[Bjorn]

Secondly, he's dead right, horns do have a small sweet spot because they are very directional.

Not necessarily. It depends on the design and beamwidth. But typically commercial horns beam a lot. Meaning the get gradually narrower and narrower the higher you go and frequency. The result is then a very small sweet spot which you typically hear at audio shows. These are poorly designed horns IMO.

But if the horn has sufficiently wide enough directivity and constant, the sound can be great anywhere in the room correctly set up. This is actually where a horn speaker will shine compared to most traditional speaker that doesn't have such broadband constant directivity. And also why they are a great choice combined with other attributes for churches, theaters etc. Everyone can experience great sound quality combined with high SPL and low distortion.