Subject: Center Speaker Recommendations for Main Horn Speakers

[Zeotrope]

I would expect Bill Woods' horns to work extremely well in a "time/intensity trading" configuration in phantom center mode, with the speaker axes criss-crossing in front of the central sweet spot.

Thanks. Indeed, the soundstage is wide, deep, and very focused. You could have a fake center speaker and listeners would never believe it’s not working.
My view of surround is less is more. Apply the same budget to fewer speakers and amplifiers and you’ll enjoy better sound.

You don’t need 10+ channels in most rooms. With a Trinnov, it will route all of the channels to the speakers you have, including creating virtual output channels.

 

[Duke LeJeune]

Thanks. Indeed, the soundstage is wide, deep, and very focused. You could have a fake center speaker and listeners would never believe it’s not working.

Yes, a good constant-directivity horn can do that!

Recently one of my clients had a friend who works for a fairly well known Southern California company that makes audio processors over for a visit. They watched a couple of show-off-my-system movie clips and some music videos. When they were done and the screen retracted towards the ceiling, the friend had a moment of cognitive dissonance, eyes searching the wall behind where the screen had been in disbelief, when there was no center-channel speaker hidden behind the screen. I've had several clients sell their rather expensive center-channel speakers because they preferred the sound quality and spatial quality in phantom center mode.

Just curious, which Bill Woods horns are you using? And, if you happen to know, what is the coverage angle of their radiation pattern?

Thanks!

 

[Francesco tube fan]

I like the Kondo Souga!
What horn is this for the center?

@Duke LeJeune how would you expect conical (straight sided) horns like the Bill Woods designs to perform in terms of radiation patterns as a phantom center?

avantgarde colibrì c2

 

[Zeotrope]

Yes, a good constant-directivity horn can do that!

Recently one of my clients had a friend who works for a fairly well known Southern California company that makes audio processors over for a visit. They watched a couple of show-off-my-system movie clips and some music videos. When they were done and the screen retracted towards the ceiling, the friend had a moment of cognitive dissonance, eyes searching the wall behind where the screen had been in disbelief, when there was no center-channel speaker hidden behind the screen. I've had several clients sell their rather expensive center-channel speakers because they preferred the sound quality and spatial quality in phantom center mode.

Just curious, which Bill Woods horns are you using? And, if you happen to know, what is the coverage angle of their radiation pattern?

Thanks!

Interesting!
I am using the Bill Woods AH!300 horn. Attaching some info from his old website.

Bill was going to design a complete speaker for me, but then OMA's Jonathan Weiss shut him down in a Trumpian fit because I asked too many questions about OMA and wanted to prevent me from having a Bill Woods' horn. (Not to digress, but I think Fremer just experienced the kind of person Weiss is, and I have a much better system than anything from OMA, so all's well.)

 

[Zeotrope]

I would expect Bill Woods' horns to work extremely well in a "time/intensity trading" configuration in phantom center mode, with the speaker axes criss-crossing in front of the central sweet spot.

Duke, can you elaborate more on the recommendation to criss-cross in front of the central sweet spot?
I just did a deep dive on channel balance, using tracks like the Roger Waters Q-Sound Late Home Tonight Pt. 1, and tracks with a solo female vocalist centered, such as Patricia Barber’s This Town.
When you nail balance, wow, does it make a difference! It’s not just L-R localization but the size of the vocalist is more realistic, and everything just snaps into focus.

What I noticed however is that head position makes a difference as well. This could be because my conical midrange are aimed at the ears, not crossed in front of the listening position? I sit about 3m away, so it’s a near field setup, with the central axis (i.e., a line drawn from the throat of the midrange) lined up to the ear.

 

[Francesco tube fan]

I like the Kondo Souga!
What horn is this for the center?

@Duke LeJeune how would you expect conical (straight sided) horns like the Bill Woods designs to perform in terms of radiation patterns as a phantom center?

Avantgarde colibrì c2

 

[ajant]

I have used time/intensity trading with constant-directivity horns and phantom center mode for many years.

Thanks for your reply. But aside from any center channel speaker questions, I am trying to select a main horn for a two or three-way system. Might there be one or more constant directivity horns which would work very well with these midwoofers for 1.5 or 2" compression drivers?

Altec 416-8B in 100L Sealed

In this blog post I test the Altec 416-8B in a 100L sealed enclosure. Testing was done outdoors for the frequency response and then brought indoors for the distortion testing. The woofers are new from Great Planes Audio. A customer has sent me the finished cabinets with drivers to integrate...

josephcrowe.com

 

[Duke LeJeune]

Thanks for your reply. But aside from any center channel speaker questions, I am trying to select a main horn for a two or three-way system. Might there be one or more constant directivity horns which would work very well with these midwoofers for 1.5 or 2" compression drivers?

Altec 416-8B in 100L Sealed

In this blog post I test the Altec 416-8B in a 100L sealed enclosure. Testing was done outdoors for the frequency response and then brought indoors for the distortion testing. The woofers are new from Great Planes Audio. A customer has sent me the finished cabinets with drivers to integrate...

josephcrowe.com

First, my apologies for being so slow to respond. I haven't been here in the past couple of weeks.

Second, that's a really good question! Unfortunately, I do not have a really good answer.

Imo two of the best off-the-shelf large-format horns are the 18Sound XT1464 and the Faital LTH142. Personally I'd like a bit wider pattern; of course that's a matter of personal preference. Ime the XT1464 measures a bit better, but I think a pair of LTH142's image a bit better. I wish the B&C ME90 had a larger-radius mouth round-over.

Now this may not matter to everyone, but imo it's a really good idea to get a good match between the exit angle of the compression driver and the entry angle of the horn. Imo a discontinuity at this junction can be a source of coloration, and of a type that cannot readily be fixed by EQ.

So (in my opinion at least) the search becomes a quest for "the best horn + compression driver combination". About a year ago I was tasked with designing a fairly high-end speaker and we ended up having custom 12" diameter horns made, at over ten times the price of the above-mentiond off-the-shelf horns.

If Joseph Crowe has a horn that he says works well with your compression driver, and if he says that horn/compression driver combination works will with the Altec 416-8b woofer, I'd suggest that over any off-the-shelf combination I'm aware of.

(Not that I've tried every promising-looking large-format off-the-shelf horn/compression driver combination, but I have yet to find one that I'm really thrilled with.)

 

[ajant]

If Joseph Crowe has a horn that he says works well with your compression driver, and if he says that horn/compression driver combination works will with the Altec 416-8b woofer, I'd suggest that over any off-the-shelf combination I'm aware of.

Thanks for your reply and suggestions. Cask05, a highly accomplished diyer, posting at diyaudio, audiokarma, Klipsch and other forums, also strongly agues that constant directivity horns sound more natural than the relatively beamy JMLC-or even Joseph Crowe’s largest exponential horn, which he also considers beamy towards HF, which
I would otherwise be inclined to go with. See sonogram here. https://josephcrowe.com/products/3d-cad-plans-for-es-290-biradial-horn-horn-no-1670

Chris’s system shown here with Klipsch K402 and/or one of the Synergy horns. https://www.diyaudio.com/community/members/cask05.73439/#about

Synergy Horn Family of Loudspeakers | Danley Sound Labs

Delivering full range audio thanks to our proprietary single source technology, the SH family of loudspeakers offers targeted, crisp sound that can cover a whole range of needs.

www.danleysoundlabs.com

SM Family of Loudspeakers | Danley Sound Labs

This technology is present in all our Synergy Horn™ Family. And Danley’s SM family of loudspeakers is simply a smaller, lighter version of our Synergy Horn family.

www.danleysoundlabs.com

https://community.klipsch.com/topic/161404-a-k-402-based-full-range-multiple-entry-horn/page/40/#findComment-2495708

https://community.klipsch.com/topic/161404-a-k-402-based-full-range-multiple-entry-horn/

Chris replying to my inquiry:

Generally, the Synergy Horns (SH, SM series) are what I was referring to, but there are others within Danley's offerings that are also useful. I recommend 90 degree x 60 degree coverage horns (like the SH-96 or SM-96 series) for a reasonable sized listening rooms. As the room gets smaller, then horn coverage angles likewise should get smaller.

But note: using smaller coverage horns (width and height) reaches a practical limit quite quickly as the room size gets smaller--and the horns must get fairly large in order to control their polar coverage to a low enough frequency. The practical limit is something like the SH-60 sized horn (full range horn loading--not direct radiating woofers).

Since you've already made the decision to buy direct radiating woofers, you've already compromised the ideal coverage at frequencies below ~500-800 Hz. That's a trade that candidly I wouldn't have made, preferring instead to employ full-range horns that control their polar coverage down to ~100-200 Hz. This means full-range horn loading (i.e., horns about 1 m wide having ~90 degrees coverage angles).

Thus, along with the K402 footprint and at least most of the Danley SH horns I’ve found (Danley doesn’t reply to my emails) appearing too big to safely sit atop my midwoofer cabinets, my use direct radiating woofers seems to have largely ruled out the practical use of such horns, as implied.

Then what constant directivity horn alternatives for my midwoofers? You recommended the 18Sound XT1464 and the Faital LTH142. But how well would their x and y coverage angles of those horns, above that of my midwoofers, work with my room’s dimensions? Compared to the sonogram of the K402 above, might they be too small for my room, as Chris implied above? See polar plots.

Eighteen Sound - Professional loudspeakers

www.eighteensound.it

https://faitalpro.com/en/products/HF_Horns/product_details/datasheet.php?id=903020105

See attached room sketch.

But how might you explain Noah117’s issues with them?.

18sound xt1464 vs faital pro lth142, can only pick one .

Very interested in trying one of these horns Looking for feedback or data to help, Looking at doing a 15" 2way Jeremy

www.avsforum.com

Yes, the B&C ME90’s polar plots are amazing. https://bcspeakers.com/en/products/horn/1-4/0/me90

But are you saying that because it has no round overed mouth-unlike the beamy JMLC AH425 horns, for example-where might the ME90’s response suffer due to consequent mouth edge diffraction?

Note that Chris and Joseph have some issues with how effectively mouth roundover will vary due to horn shape.

Horn mouth roundover, 135 or 180 degrees?

A question about optimum horn termination: What is the most optimum shape to terminate a horn? -The 180 degree-version has a complete rondover but the radius is smaller given a constant horn witdth. -The 135 degree-version has compromized the end of the roundover for a larger radius and a...

www.diyaudio.com

However, what horn for a two or three way system for my midwoofers and my room?

 

[Duke LeJeune]

But how might you explain Noah117’s issues with them?.

He's right. It's a tradeoff. I agree that 80 or 90 degrees coverage would be better for home audio, and I'd like a larger horn as well for pairing with a 15" woofer, in order to maintain good pattern control down lower.

But are you saying that because it has no round overed mouth-unlike the beamy JMLC AH425 horns, for example-where might the ME90’s response suffer due to consequent mouth edge diffraction?

There will be a reflection at the mouth because of the inadequate round-over, and that reflection energy will arrive at a slightly different time from the sound that travelled directly to the listener without reflecting at the mouth. I think this will tend to blur the imaging a little bit, and it might degrade the clarity a little bit, despite how good the polars look.

However, what horn for a two or three way system for my midwoofers and my room?

Good question.

One I haven't tried, and it does have a wide pattern, is the horn JBL uses in the M2. Apparently it can be purchased separately:

JBL M2 Horn Lens 5025594 - The Speaker Exchange

Brand New Genuine JBL M2 Horn Lens #5025594. It is normal for there so be some slight mold imperfections on the mounting flange.

reconingspeakers.com

My recollection is that, in the Model 19, Altec used the 416-8B up to about 1.2 kHz, where it crossed over to a 1" throat compression driver. So you MIGHT consider going with a 1" throat compression driver paired with a horn which works well down to that region, at least in the horizontal plane. The H290C sold by PiSpeakers works well with the B&C DE250, and it probably works well with the DE550 and DE550TN. The latter two have a 2" diaphragm, which is unusually large for a 1" throat driver, and so they probably have better low-end power handling than most 1" throat compression drivers.

 

[Duke LeJeune]

Duke, can you elaborate more on the recommendation to criss-cross in front of the central sweet spot?

Sorry I overlooked replying to your post!

Ime, one has to find the correct toe-in angle by trial-and-error. As a starting point, you might try 1/2 of whatever the horn's coverage pattern is. So if it's a 70-degree horn, you might try about 35 degrees of toe-in, but expect to make adjustments from there.

I cannot promise that your horns will work well in this configuration, but I THINK they can.

 

[ajant]

He's right. It's a tradeoff. I agree that 80 or 90 degrees coverage would be better for home audio, and I'd like a larger horn as well for pairing with a 15" woofer, in order to maintain good pattern control down lower.

But what I don't get is that assuming Cask05 is correct a 15" woofer will maintain its polar pattern up to ~ 900Hz, as shown here. https://www.diyaudio.com/community/...ver-that-can-reach-500-hz.321521/post-8022282 In which case, unless one has a much larger room than mine, what need for a larger horn than the ME90, one of the smaller Danley SM horns, et al?

 

[ajant]

Also Duke, please excuse all the text, but I’d appreciate your evaluation and comments on my journey and goals.

I've never heard the Altec 19 speakers, nor have I even heard the Altec 416 woofers except in VOTTs when in local cinemas long ago. I chose to have Jim Salk clone Gary Dahl’s sealed 416s when I first thought to clone his two-way speakers using the Radian 745neoBe/JMLC AH4254 driver/horn combo. Gary was thrilled with them for a year or so but ultimately found those Be drivers produced “hard” sounding highs and/or other unsatisfying performance-at least in that horn-and where repeated EQing attempts didn’t help. Luckily, I found this out before proceeding with that build.

And lucky for Gary, his hunch to use a friend’s pair of Yamaha JA6681B drivers with a new pair of 425 horns has delighted him ever since.

compression driver impedance question

Done here my blog about it Audio by Martin will have to listen to them, just measured so far. Been busy this weekend. Thank you. :) the spl gain of the waveguide at 3khz is about 9db!!! the gain is still about ~3db at 10khz loading fall like a rock after ~14khz and EQ must be used to...

www.diyaudio.com

compression driver impedance question

Done here my blog about it Audio by Martin will have to listen to them, just measured so far. Been busy this weekend. Thank you. :) the spl gain of the waveguide at 3khz is about 9db!!! the gain is still about ~3db at 10khz loading fall like a rock after ~14khz and EQ must be used to...

www.diyaudio.com

My VOT replaced by a "Super VOT", the Yamaha S6215

Dear all, After many years of satisfaction and enjoyment, I found an opportunity to upgrade my system. I started with the replacement of my 1505B driver, a 291-16A, with a Yamaha 6681B. Amazed by the result, I started to study a bit more about Yamaha loudspeakers and I found that the 15 inch...

www.diyaudio.com

Gary’s raves about that combo seem genuinely free of diyer/owner bias. But regarding speaker placement, here’s his system before and soon after he swapped in the Yamahas.

https://galibierdesign.com/wa-trip-01/

About speaker placement, wouldn’t the AH425 horn’s beaming directivity placed that far apart, vs. his apparent listening distance, cause poor imaging or other problems?

Btw, to minimize IM and third harmonic distortion, Gary said he designed those 3 cu ft. sealed cabinets to roll off the 416’s response at ~ 70Hz. Years later, he said that he only uses his pair of subs (Acoustic Elegance drivers with passive radiators underneath the Altec midwoofers) for HT use. For music, Gary claims that he gets adequate bass from room gain. Me, I don’t expect get that lucky, so I had Jim Salk also build me two pairs of Rythmik F12 subs.

Though the Yamahas’ HF response is limited https://www.azurahorn.com/JA6681B on 425 Horn (2).pdf , what’s below 10kHz must be sublime, as the driver had a strong enough following for Meyer Sound to have Yamaha build the JA6681B under their brand. Even today many prefer this driver though lament their failure to persuade OEMs to clone the long since discontinued diaphragm. https://www.diyaudio.com/community/...ompression-driver.163469/page-31#post-7680257

I too have a pair of them purchased from Troy Crowe.

But unlike Gary and/or those at that thread, I’d rather not live in dread of blown diaphragms or having an even more fragile ~2% beryllium suspension “finger” break on me. And Arez implied that the fingers will eventually deteriorate.

Beyond the Ariel

Unless I am missing something on the site I couldn't find a single measurement. Without measurements to back it up I would count the claims as marketing and be skeptical about those claims. If there are measurements I missed can someone post a link? Rob :) I've seen the proof, but won't break...

www.diyaudio.com

A shame for a driver that must otherwise sound wonderful.
https://audio-database.com/YAMAHA/unit/ja-6681.html

And the JA6681B can apparently sound great at frequencies low enough to cross low with the Altec 416 to thereby keep the latter from playing in that area of the midrange where it would otherwise distortion. Thus, for as long as the Yamahas diaphragms and suspension stay intact, Gary must be enjoying very clean sounding speakers. https://josephcrowe.com/blogs/news/altec-416-8b-in-100l-sealed

Unfortunately, as I have very limited DIY skills, and even less time and workspace, I can’t experiment with using my Yamaha drivers because, I don’t know if horns like Gary’s JMLC 425s would sound too “head in vise” beamy for me. Instead, I must decide on constant directivity horn/driver combo, ship them to Troy Crowe, who will design the crossover for them with the Altec 416s and finish building the system.

So, if not the ME90, is JBL M2 or what other constant directivity horn for the Altec 416 cabinets?

Comments about these M2 horn measurements and polar charts?

Measurements for speaker JBL M2

www.spinorama.org

Any recommended (beryllium?) driver for the JBL M2 for a two- or three-way system, though I doubt that I can hear beyond ~ 13kHz?

For the M2 horn, is it practical to expect to get flat response to ~ 13kHz and down to ~ 600Hz with what model driver and throat exit size? Would DSP be required or would passive EQ suffice for a specific (Be?) driver?

Also, any experiences or opinions with Danley horns?

SM Family of Loudspeakers | Danley Sound Labs

This technology is present in all our Synergy Horn™ Family. And Danley’s SM family of loudspeakers is simply a smaller, lighter version of our Synergy Horn family.

www.danleysoundlabs.com

 

[Robh3606]

For the M2 horn, is it practical to expect to get flat response to ~ 13kHz and down to ~ 600Hz with what model driver and throat exit size? Would DSP be required or would passive EQ suffice for a specific (Be?) driver?

In a word no. I use them with large format 4" drivers and measurements don't support use that low. I cross at 700Hz which is right at the knee ideal would be 800Hz where JBL does in the M2. I chose 700Hz as I got the best transition with my passive network. They will easily go as high up as the compression driver will balancing woofer sensitivity vs. compensation. Mine go out past 20K.

You could get 2435's to work barely in a passive network to about 15K depending on woofer sensitivity. Active no issues with DSP as they are used in the Vertec systems they were designed for.

Rob.

 

[Kingrex]

I would buy a small Sonos package of 3 for fronts and 2 for rear. Keep the main speakers out of it.

Cinema seems to play best with speakers built for the purpose. One of the best TV setup I had was quantity 5 very small, like 3x3x3 Bose cubes and a small sub. My PAP Trio 10 or 15 were by far the worst.

 

[Duke LeJeune]

But what I don't get is that assuming Cask05 is correct a 15" woofer will maintain its polar pattern up to ~ 900Hz, as shown here. https://www.diyaudio.com/community/...ver-that-can-reach-500-hz.321521/post-8022282 In which case, unless one has a much larger room than mine, what need for a larger horn than the ME90, one of the smaller Danley SM horns, et al?

You could use the ME90 with a 15" woofer, and cross over in that region. The ME90's pattern starts widening below about 1.5 kHz, so you might want to put a gentle dip in its on-axis response at 900 Hz to offset its increase in off-axis energy in that region.

Also Duke, please excuse all the text, but I’d appreciate your evaluation and comments on my journey and goals....

About speaker placement, wouldn’t the AH425 horn’s beaming directivity placed that far apart, vs. his apparent listening distance, cause poor imaging or other problems?

Not necessarily. The ear's vertical resolution is not all that good in the 900 Hz ballpark. I don't know what the minimum listening distance would be, but that's something one can trial-and-error to some extent. Ears at the midpoint height, between woofer and compression driver, might work better than ears at the compression driver's height.

That being said, I'd try to keep the vertical spacing as small as is practical.

Though the Yamahas’ HF response is limited https://www.azurahorn.com/JA6681B on 425 Horn (2).pdf , what’s below 10kHz must be sublime...

I would try adding an adjustable-loudness up-firing or rear-firing supertweeter to increase the in-room reflection energy in the top octave or so. This would improve the perceived spectral balance without imposing the juggling of tradeoffs that a front-firing supertweeter does.

The Azurahorn subscribes to a different school of thought than the constant-directivity one I subscribe to. The Azurahorn works well with a much simpler crossover, while a constant-directivity type generally requires EQ to compensate for its generally falling response, but once that has been done, the off-axis response has also been corrected.

Unfortunately, as I have very limited DIY skills, and even less time and workspace, I can’t experiment with using my Yamaha drivers because, I don’t know if horns like Gary’s JMLC 425s would sound too “head in vise” beamy for me. Instead, I must decide on constant directivity horn/driver combo, ship them to Troy Crowe, who will design the crossover for them with the Altec 416s and finish building the system.

So, if not the ME90, is JBL M2 or what other constant directivity horn for the Altec 416 cabinets?

On paper at least, the JBL M2 horn LOOKS to me like the better match.

Comments about these M2 horn measurements and polar charts?

Measurements for speaker JBL M2

www.spinorama.org

I tend to design for a somewhat off-axis primary listening axis, and tend to prefer a gently downward-sloping response along that primary listening axis. So I might design for 15 or 20 degrees off-axis.

There is one aspect of the M2 horn's design that I have some uncertainty about: If you look down in the throat, you'll see four "knuckles", two at the "north and south" positions, and two at the "east and west" positions. These knuckles look to me like relatively gentle diffraction-inducing features, which would presumably have the effect of widening the radiation pattern of the high frequencies in those dimensions. But in between the "knuckles" (northeast, southeast, northwest, southwest), we actually have valleys, thus presumably the radiation pattern shape is different in between the knuckles.

So I THINK the net off-axis response might not be fully characterized by the horizontal and vertical polar responses, as those directions have the benefit of the "knuckles" while the in-between off-axis responses do not. Whether or not this would actually matter in practice, I do not know.

Any recommended (beryllium?) driver for the JBL M2 for a two- or three-way system, though I doubt that I can hear beyond ~ 13kHz?

I would priorize matching the exit angle of the compression driver to the entry angle of the horn. So my first choice for the M2's horn would be the one JBL uses (D2430K I think).

I don't know what the future of Beryllium diaphragms will be. For instance Radian doesn't seem to have any in stock, and my understanding is that 18Sound's Beryllium diaphragm drivers are special order only.

For the M2 horn, is it practical to expect to get flat response to ~ 13kHz and down to ~ 600Hz with what model driver and throat exit size?

I don't know what the entry angle is for the M2 horn; presumably the compression driver they use in the M2 is a good match.

I would hesitate to use the M2 horn down to 600 Hz. It's a fairly shallow horn, and shallowness tends to work against good extension at the lower end of a horn's range.

Take @Robh3606's advice before you take mine, as apparently he has seen representative raw horn-and-driver measurements.

Would DSP be required or would passive EQ suffice for a specific (Be?) driver?

I don't know.

Also, any experiences or opinions with Danley horns?

Sorry, no. I'm prejudiced against horns that have what looks to me like an inadequate mouth round-over, but you'd probably be better off trusting experienced users over a prejudiced competitor.

 

[ajant]

Sorry, no. I'm prejudiced against horns that have what looks to me like an inadequate mouth round-over, but you'd probably be better off trusting experienced users over a prejudiced competitor.

I asked about Danley mostly because https://www.diyaudio.com/community/members/cask05.73439/ suggested them as they-like the Klipsch K510 with added "mumps"-are what he called "straight-sided" horns:

I'm not talking about "tractrix", "exponential", "hypex", or JMLC, IWATA, horn profiles, etc., which have curved throats (any of the horn's walls near the throat). All of these horns suffer from reduced high frequency polar coverage angles relative to straight-sided horns--which is undesirable and there isn't anything that can be done to correct it. Having wide HF coverage is one of the most difficult things to achieve, in terms of horns profiles. Only the straight-sided horns (like conical horns) have good HF polar coverage that doesn't shrink excessively as the frequency rises. If you can find a SEOS horn, this will be better than the other curved-wall horns, but still suffer a bit due to its slightly curved throat wall entrance.

About high in frequency would this degree of round over eliminate diffraction?

3D CAD plans for ES-290 Biradial Horn -- Horn No.1670

By purchasing these plans you'll automatically receive the 3D CAD files for the ES-290 Biradial Horn No.1670 via email. Zipped Folder Includes: .STL files for 3D Printing (Assembly and part files) .STEP, .iges, & .x_t files for CNC Machine CAM programming Does not include CNC programming...

josephcrowe.com

Chris and Joseph square off a bit here.

Horn mouth roundover, 135 or 180 degrees?

A question about optimum horn termination: What is the most optimum shape to terminate a horn? -The 180 degree-version has a complete rondover but the radius is smaller given a constant horn witdth. -The 135 degree-version has compromized the end of the roundover for a larger radius and a...

www.diyaudio.com

How Danley reduces diffraction may depend largely on their speaker model and they don't seem to discuss that at all here.

SM Family of Loudspeakers | Danley Sound Labs

This technology is present in all our Synergy Horn™ Family. And Danley’s SM family of loudspeakers is simply a smaller, lighter version of our Synergy Horn family.

www.danleysoundlabs.com

But there's discussion here. https://www.diyaudio.com/community/search/2297355/?q=danley+diffraction&o=relevance

I will be asking their northeast sales rep for horn solutions for my midwoofers, but if that doesn't work out suggested alternatives would be much appreciated.

 

[Duke LeJeune]

I'm not talking about "tractrix", "exponential", "hypex", or JMLC, IWATA, horn profiles, etc., which have curved throats (any of the horn's walls near the throat). All of these horns suffer from reduced high frequency polar coverage angles relative to straight-sided horns--which is undesirable and there isn't anything that can be done to correct it. Having wide HF coverage is one of the most difficult things to achieve, in terms of horns profiles. Only the straight-sided horns (like conical horns) have good HF polar coverage that doesn't shrink excessively as the frequency rises. If you can find a SEOS horn, this will be better than the other curved-wall horns, but still suffer a bit due to its slightly curved throat wall entrance.

It's always a juggling of tradeoffs.

Suppose we have a compression driver with an exit angle of 14 degrees and we want a horizontal coverage pattern of 90 degrees. If we go with a straight-sided 90 degree horn, the transition from 14 degrees to 90 degrees happens abruptly at the junction where the compression driver's exit meets the horn's throat. This rather severe discontinuity will cause diffraction, which is imo undesirable because diffraction in a horn can cause a type of distortion which cannot be fixed with equalization, and which may be undetectable at low SPLs but tends to become increasingly audible and objectionable at high SPLs. BUT, we do get our 90 degree pattern across pretty much the whole frequency range, possibly narrowing in the top octave as the compression driver itself starts to beam. The mouth termination presents another set of challenges, which different designers deal with in different ways.

For the most gradual transition between our 14-degree exit angle and our desired 90-degree coverage pattern, we would probably choose something with a very smooth and gentle continuous curve, like a JMLC horn, or like Joseph Crowe's horn that you linked to, and the horn's entry angle would ideally match the compression driver's exit angle (14 degrees in our example). But now our radiation pattern is not a consistent 90 degrees across the range, but rather is narrow at high frequencies and widens as we go down the spectrum. On the other hand, the mouth termination is usually fairly benign, as the mouth is essentially just an extension of the horn's profile, which even has a gentle roll-back on a JMLC horn:



Horns which trace back to Earl Geddes's work (like the SEOS horn) have a lot of curvature way down in the throat instead of spreading the curvature across the whole length of the horn, and the rate of curvature in the throat is calculated to introduce the least amount of diffraction while making the transition from the 14 degree entry angle (corresponding to our compression driver's exit angle) to the target 90-degree radiation pattern. In practice, despite the transition from throat-entry-angle-to coverage-pattern-angle way down in the throat, the radiation pattern will still narrow in the top octave or so as the driver's inherent beaming makes its pattern narrower than our horn's 90-degree walls. And once again, the mouth termination has to be dealt with.

Here is a graph somebody made which shows several different horn contours:

Note that the Conical, Oblate Spheroid, and Hughes (Peavey) curves do not have mouth round-overs because the mouth is not included in the math for those shapes; the designer has to figure out how to graft on (or blend in) a mouth. If you are familiar with Marcel Batik's horn design work as described on DIY Audio, his math can be used to get a smooth mouth transition on what would otherwise be a Conical or Oblate Spheroid or Hughes profile.

About high in frequency would this degree of round over eliminate diffraction?

3D CAD plans for ES-290 Biradial Horn -- Horn No.1670

By purchasing these plans you'll automatically receive the 3D CAD files for the ES-290 Biradial Horn No.1670 via email. Zipped Folder Includes: .STL files for 3D Printing (Assembly and part files) .STEP, .iges, & .x_t files for CNC Machine CAM programming Does not include CNC programming...

josephcrowe.com

Looks to me like that horn's profile (in the horizontal at least) is one long, continuous curve of constantly-but-gently-changing radius, with a roll-back somewhat reminiscent of the JMLC horn. I think Joseph Crowe's #1670 would have negligible diffraction in the horizontal plane down to as low as the horn has good pattern control.

It looks to me like its radiation pattern would become progressively narrower as we go up in frequency. I mentioned this phenomenon above but let me go into a bit more detail:

To a ballpark first approximation, for a non-diffraction horn, when you have controlled the radiation pattern for about 1/4 of a wavelength you have controlled the pattern's width at that wavelength. So for short wavelengths (high frequencies), the angle of the radiation pattern is set by the narrow angle of the walls way down in the throat region. For long wavelengths (near the bottom end of the horn's range), the angle of the radiation pattern is much wider because the effective wall angle is much wider.

(This same principle applies to conventional speakers, and you're probably familiar with it as the "baffle step". You can think of a conventional speaker's front baffle as a 180-degree horn. The radiation pattern width will be no wider than 180 degrees at frequencies where the distance from the center of the driver to the cabinet edge is more than 1/4 wavelength. Below the frequency where the distance to the baffle edge is 1/4 wavelength, the "180 degree horn" is too short to control the radiation pattern's shape, and it begins to widen until it's essentially omnidirectional in the bass region.)

Anyway, as you can see, it's always a juggling of tradeoffs.

 

[ajant]

Thanks for this valuable feedback, though which raises very important questions regarding horn choices, directivity pattern vs. preferences for get some ratio of direct (focused) and indirect (spacious) sound, speaker placement vs. room size and dimensions and preferred listening distance.

While I haven't done measured comparisons of my 2 ft midwoofer cabinets versus the usable space between the southeast corner, if equilateral triangle listening with the speakers is the optimal placement method, won't that pretty much defeat the purpose of having constant directivity horns of which vertical and horizontal coverage angles?

And while using such horns (JBL M2 certain Danley models, Troy Crowe's big ES290) might beam like the JMLC 425 won't corner placement eliminate aural spaciousness-and perhaps even more so because with equilateral triangle speaker placement I would therefore have to listen a lot closer than the 11 ft I had preferred.

 

[Duke LeJeune]

if equilateral triangle listening with the speakers is the optimal placement method, won't that pretty much defeat the purpose of having constant directivity horns of which vertical and horizontal coverage angles?

I'm not sure what you're asking. Are you asking whether the radiation pattern's coverage angles matterif the speakers + listener are set up in an equilateral triangle?

... won't corner placement eliminate aural spaciousness...

Again, not sure I understand... are you asking about whether corner placement would make the radiation pattern's coverage angles unimportant when it comes to spaciousness?

Thanks!