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

[Keith_W]

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.

I am under the impression that whether you like wide directivity or narrow directivity is a matter of preference and your room, and that what is important is that the off-axis response remains smooth. For e.g. if you have a narrow room, wide directivity would mean a lot of early reflections, so a narrow directivity speaker would be better. Is this incorrect?

 

[Bjorn]

I am under the impression that whether you like wide directivity or narrow directivity is a matter of preference and your room, and that what is important is that the off-axis response remains smooth. For e.g. if you have a narrow room, wide directivity would mean a lot of early reflections, so a narrow directivity speaker would be better. Is this incorrect?

This is not really about narrow vs wide direcitivty. It's more about constant or uniform vs something that's non uniform. The latter will alter the spectral content because the reflected energy changes with frequency. Obviously it depends on how the beaming is. If it only starts to narrow very high in frequency and the rest of the frequency area is constant down to 500 Hz or lower that's an ok compromise for one listener. Listeners outside sweetspot will not get good a treble though. However, the ealier the beaming starts, the less uniform is the directivity and the more reflected energy deviates. And this is how many horns behave.

 

[Hear Here]

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.

It is impossible to demonstrate horns at their best at audio shows. If properly set up for best sound quality, there will be a small sweet spot (but an exceptionally sweet one) but this isn't suitable at shows where there may be 20+ people listening outside this sweet spot. The speakers will be likely faced more directly forward, thus allowing "good" sound to far more listeners but no "great" sound anywhere. I've been there when I loaned my horn speakers to an amp builder at a UK show.

The best horn speakers (if imaging is important to you) will be ideally set up with a small sweet spot. If imaging is of little importance and you need a wider sweet spot, then consider alternative types of speaker, as you are not taking full advantage of the biggest benefit of horns. The best horns are the ones with the most "lifelike" presentation ie you can close your eyes and point unambiguously at each instrument and they sound like they did at the live performance. Unless properly set up you won't achieve this.

 

[rank]

The only horn-loaded speakers I've heard sounding more or less correct were all DIY. With developments that took years.
I see absolutely no interest in them for any kind of domestic listening given their very strong directivity.
If high efficiency is sought, there are excellent dynamic loudspeakers in all three registers, from 96 to over 100 dB (real) of efficiency. And cheap.

 

[Newman]

Is it me?

It could be bad luck. There are good and bad non-horns, and good and bad horns.

Am I biased?

Alan, IME of you, you are only biased against bad science, badly implemented. The science behind horns is solid, and applicable to home audio if well implemented.

Heard the wrong horns?

Almost certainly. But it's not your fault, nor surprising, because the home hobby audio world has done to horns exactly what it has done to some other exotic audio technologies: mythologise all the wrong stuff. Examples: seashell-like spirals, "legendary classics" eg Altec or old JBL or Sato or classic Klipsch, exponential or tractrix flares, long horns, sharp or too tight mouth terminations, old tech compression drivers, truncated backhorns on full range drivers, low order crossovers, no time alignment, no EQ.

Some rave about horns but it's lost on me.

I would say that, as a class, horns have a smaller percentage of extremely good implementations for home audio than almost any other audio tech. Look for constant directivity waveguide horns, time alignment, steep digital XO, plenty of EQ, non-truncated horns....then look for well done.

 

[Brian Hall]

Some people get carried away with horns:

 

[Sal1950]

Some people get carried away with horns:

WOW, but there's worse. LOL
I'm wondering what the 2 small orange (drivers?) are on the top shelves and again on the third? Supertweeters ???
That is carzy and still he had to fall back to 2 large stacks of direct radiator subwoofers.
I've loved horn speakers for many decades but like most everything else there's never moderation, the pendulum always needs to be swung from one extreme to the next.
Although my La Scala's had some FR irregularities, when properly setup I've never heard a speaker that could better them when it comes to laser sharp imaging focus.
Let the flames begin.

 

[Bjorn]

It is impossible to demonstrate horns at their best at audio shows. If properly set up for best sound quality, there will be a small sweet spot (but an exceptionally sweet one) but this isn't suitable at shows where there may be 20+ people listening outside this sweet spot. The speakers will be likely faced more directly forward, thus allowing "good" sound to far more listeners but no "great" sound anywhere. I've been there when I loaned my horn speakers to an amp builder at a UK show.

The best horn speakers (if imaging is important to you) will be ideally set up with a small sweet spot. If imaging is of little importance and you need a wider sweet spot, then consider alternative types of speaker, as you are not taking full advantage of the biggest benefit of horns. The best horns are the ones with the most "lifelike" presentation ie you can close your eyes and point unambiguously at each instrument and they sound like they did at the live performance. Unless properly set up you won't achieve this.

It seems like you either didn't see the graph I posted or didn't understand it!
It clearly refutes why you are saying here by showing a uniform directivity that will not only cover several seats, but could cover the whole width of the room if the speakers are placed correctly.

It seems like you have yet to experience a truly constant directivity horn speaker

 

[Hear Here]

It seems like you either didn't see the graph I posted or didn't understand it!
It clearly refutes why you are saying here by showing a uniform directivity that will not only cover several seats, but could cover the whole width of the room if the speakers are placed correctly.

It seems like you have yet to experience a truly constant directivity horn speaker

I'm less interested in graphs than the music I listen to.

With horns, at least my own ones, the ideal placement is such that the listener in the sweet spot gets a much greater impression of "being there" at a live performance than with other types of speaker. If the angle of the speaker is changed a degree or two, this imaging accuracy is seriously diminished - whatever your graph may suggest!

The ideal placement of horns for best imaging leaves a small sweet spot (easily demonstrated) and as I suggested earlier, this is not ideal for audio show demo rooms where lots of people want to listen. At such shows, imaging has to be set aside and the clarity and detail that good horns can deliver is emphasised.

 

[Duke]

I'm less interested in graphs than the music I listen to.

With horns, at least my own ones, the ideal placement is such that the listener in the sweet spot gets a much greater impression of "being there" at a live performance than with other types of speaker. If the angle of the speaker is changed a degree or two, this imaging accuracy is seriously diminished - whatever your graph may suggest!

The ideal placement of horns for best imaging leaves a small sweet spot (easily demonstrated) and as I suggested earlier, this is not ideal for audio show demo rooms where lots of people want to listen. At such shows, imaging has to be set aside and the clarity and detail that good horns can deliver is emphasised.

@Bjorn is right.

There are types of horns that give very uniform coverage over a wide enough angle to include a large listening area. Yes the imaging is going to be most precise in the most ideal listening position, but there will still be an enjoyable soundstage from a fairly wide range of listening positions, and the timbre will be correct throughout the listening area. This may not have been your experience because the types of horns that can do this are the exception rather than the rule, and their tradeoffs have to be minimized.

When it comes to horn design, and to system design using horns, the specifics matter a great deal.

 

[Paffi]

There are types of horns that give very uniform coverage over a wide enough angle to include a large listening area.

Do you have recommendations for Horns like that? My impression was, that JBLs can do that.

 

[gnarly]

I'm less interested in graphs than the music I listen to.

With horns, at least my own ones, the ideal placement is such that the listener in the sweet spot gets a much greater impression of "being there" at a live performance than with other types of speaker. If the angle of the speaker is changed a degree or two, this imaging accuracy is seriously diminished - whatever your graph may suggest!

The ideal placement of horns for best imaging leaves a small sweet spot (easily demonstrated) and as I suggested earlier, this is not ideal for audio show demo rooms where lots of people want to listen. At such shows, imaging has to be set aside and the clarity and detail that good horns can deliver is emphasised.

Yes, you have horns that have a narrow focusing pattern, and also multiple horns that must focused together. Net result a very small, perhaps tiny sweet spot.
Truly one person, sit exactly here, type listening. ime. If that's what you like, amen enjoy. (I was into that kind of listening for a long time, albeit with electrostats.)

Make no mistake, constant directivity horns (which I'm now into) do allow a much larger sweet area. And certain types of constant directivity horns such as unity/synergy designs, can use one horn only for the entire spectrum (apart from low sub duty). This is also an advantage over multiple horns, both in terms of enlarging the sweet area, and getting closer to the sonic benefits of "point source".

 

[Duke]

Do you have recommendations for Horns like that? My impression was, that JBLs can do that.

The big 15" JBLs (M2 and 4367) can, and the 4430 of yesteryear could. Earl Geddes' designs could, but they are no longer in production. The horn speakers in PBN's M-series can (not positive about the one with the giant butt cheeks horn; I'm not skeptical of it; just don't know enough about it). Synergy (Tom Danley) horns presumably can. I'm quite confident @Bjorn's speakers can. Probably the Dutch & Dutch 8c. The PiSpeakers 4Pi can. I'm sure there are others that don't come to mind offhand. I uses Earl Geddes' design principles (and sometimes Earl Geddes' math), so I claim that mine can.

 

[robh36062]

The big 15" JBLs (M2 and 4367) can, and the 4430 of yesteryear could.

Yes and so do the Array 1400 and it's siblings. It' all about the horn geometries and it really started at EV before DB Keele came to JBL. He was instrumental developing the 2344 Check out his website! JBL's horns forever changed after. DB Keele's 1975 paper.

4430 https://www.audioheritage.org/vbull...2-Improvements-in-Monitor-Loudspeaker-Systems

 

[Duke]

Yes and so do the Array 1400 and it's siblings. It' all about the horn geometries and it really started at EV before DB Keele came to JBL. He was instrumental developing the 2344 Check out his website! JBL's horns forever changed after. DB Keele's 1975 paper.

4430 https://www.audioheritage.org/vbull...2-Improvements-in-Monitor-Loudspeaker-Systems

Back in 2001 I had what I thought was a brilliant and original idea: Combine a 15" woofer with a large constant-directivity horn, with their patterns matching in the crossover region. Little did I know at the time that JBL's Model 4430 was already doing exactly that twenty years earlier!

My quest for a suitable horn led me to Earl Geddes, and the project he and I started eventually became the GedLee Summa (my role was smaller than that sentence makes it sound).

Anyway up until the time he built his first set of Summas, guess what speakers Earl used? Yup. The JBL 4430. His Summa might be called an update of the 4430 concept.

 

[robh36062]

Back in 2001 I had what I thought was a brilliant and original idea: Combine a 15" woofer with a large constant-directivity horn, with their patterns matching in the crossover region. Little did I know at the time that JBL's Model 4430 was already doing exactly that twenty years earlier!

My quest for a suitable horn led me to Earl Geddes, and the project he and I started eventually became the GedLee Summa (my role was smaller than that sentence makes it sound).

Anyway up until the time he built his first set of Summas, guess what speakers Earl used? Yup. The JBL 4430. His Summa might be called an update of the 4430 concept.

Hello Duke

I had no idea you you were involved with Earl. Yes somewhere I have pictures of his blind testing evaluation with the Sumas, 4430 and can't remember the third. I wasn't there grabbed them off the net. It's really something how things have evolved! The newer waveguides Earls included are really good and well worth a listen. They break the mold on old expectations!

Rob

 

[Duke]

Hello Duke

I had no idea you you were involved with Earl. Yes somewhere I have pictures of his blind testing evaluation with the Sumas, 4430 and can't remember the third. I wasn't there grabbed them off the net. It's really something how things have evolved! The newer waveguides Earls included are really good and well worth a listen.

Rob

The other speaker was the Gradient Revolution. I brought it to Earl's house in Michigan from my home showroom in New Orleans. The intrepid speaker-shuffler crouched behind the curtain for that blind listening test (Summa with TAD 2001; Summa with B&C DE25; JBL 4430; and the Gradient) was me.

Not too long after I first contacted Earl, he said to me (paraphrasing): "Duke, this speaker you want to build is something I've wanted to do for about ten years. Let's do this as a joint venture." The arrangement was that I'd pay for the stuff we needed and Earl would provide the engineering. Well I ran out of money had to drop out, so Earl finished the development of the Summa on his own. Then at one point I was assembling Summas for him at my home in New Orleans, but had to relocate unexpectedly, and no longer had a space where I could assemble and paint them.

I used the DDS ENG 1-90 waveguide until it became unobtainium, and a few years ago made my first 10" diameter Oblate Spheroid, with a 1" throat. The math took me days (I'm among the five-out-of-four people who suck at math) but the results were encouraging so I wanted to try it with a 1.4" throat and a bit narrower coverage angle.

Over the course of the past year or two I have been using Earl's math to design 1.4" throat Oblate Spheroid waveguides and try them with various compression drivers. Since I believe in matching the entry angle of the waveguide with the exit angle of the compression driver, and since the math for the entire waveguide is affected by the entry angle, it means a new waveguide is needed for each different compression driver exit angle in order to make "apples to apples" measurement comparisons. If I live long enough, eventually there will be a product or two resulting from all of this.

 

[robh36062]

Over the course of the past year or two I have been using Earl's math to design 1.4" throat Oblate Spheroid waveguides and try them with various compression drivers. Since I believe in matching the entry angle of the waveguide with the exit angle of the compression driver, and since the math for the entire waveguide is affected by the entry angle, it means a new waveguide is needed for each different compression driver exit angle in order to make "apples to apples" measurement comparisons.

Hello Duke

Have you looked at the large format JBL Coherent Wave Phase plug drivers 1.5" versions are throatless lot's of them available.2450SL 2451 2452 2453 As far as I know being throatless they don't have the exit angle issue. That's a pain hey an idea.

Could you do a waveguide with a factory or user insertable sleave kind of like the old 2440 throats to match the drivers?

How many options are there? I don't see why you could not look at the range and take a look in say a cad program for feasibility, depending on the throats overall OD vs ID.

Food for thought good luck on your project.

Rob

 

[Duke]

Have you looked at the JBL Coherent Wave Phase plug drivers 1.5" versions are throatless lot's of them available. As far as I know being throatless they don't have the exit angle issue. Here is a cross-section Same with the 2431/2435's but short phase plug very finicky response vs mounted horn/waveguide.

Rob

It's been years since I looked into JBL compression drivers but my recollection is that they no longer sell their best stuff to "the public", which would include me. Yeah theoretically I could source drivers as "replacement parts", but that's kinda like climbing in through a window instead of ringing the doorbell, unless they gave me permission.

But I DO like the idea of the phase plug ending right at the exit of the driver. Theoretically that should work with any horn or waveguide entry angle. In practice, there are still "better" and "worse" entry angles, in my limited experience. For example, I would want to use a horn or waveguide entry angle of zero degrees with the compression driver in that diagram.

 

[robh36062]

Yeah no it would be used or the newer dual rings if you could get them, they are not dependable as a driver source. I updated my previous post.

Rob