MEH / Synergy Horns vs. direct radiators - deep dive w/ measurements

[Tell]

I wouldn't use a smaller horn aperture than ~32" horizontally and ~19" vertically...because you're giving up the real advantages of doing MEHs by using a smaller horn mouth.

And building or buying a larger MEH mouth is bird feed in terms of added costs. Why not just find a way to integrate into your room? If your room REALLY IS too small for larger horn mouths--first consider using headphones instead. Your listening room is too small.

(I don't like to use any listening room less than 13'-14' in depth/horizontal dimensions because of problems of first bounce reflections from the back of the room that come too soon to the listeners.)

Right now, I can go out and buy a Celestion Axi2050 and put it on a K-402 horn, and not do any crossing to lower frequency drivers--and I get down to 200-250 Hz at the lower breakpoint frequency using only one amplifier channel and amplitude (SPL) EQ only--and it will perform just as well or better in-room for home hi-fi audio as an MEH built to go down to that frequency (or higher).

You don't need to have all the "MEH rigamarole" to get down to 200 Hz. The real reason why I think MEHs are used is because their horn apertures support at least horizontal polar control down to ~200 Hz or lower.

Before doing your "small MEH", first go listen to a full-range MEH that have been dialed-in--like the ones that Danley makes in its Synergy series...in a room that's treated for early reflections (less than 4 ms). Listen to the neutrality of sound quality that comes with flattened amplitude AND phase response...AND full-range directivity down to the room's calculated Schroeder frequency.

You'll never want to go back after hearing it.

Chris

So what is the "real advantages" of MEHs? Controller directivity from a single point is what I gather, and also that there is advantages in lower distortion from the mid drivers according to some (but not according to OP it seems). And getting that down to 500Hzish is still way better than what most normal speakers do today isn't it? Or is it something inherently important with getting that very last bit controlled as well?

And yeah sure I can buy either of those compression drivers you guys suggested and put it in a Klipsch horn, I'm sure it would sound awesome! ...or maybe I can't because I don't have that kind of money to spend and from a quick search I'm not sure I can even buy them "locally". Instead I'm aiming for all the drivers for around maybe a third of the cost of a single one of those Celestion drivers and then 3D print the horn it self, so it will be a fun DIY project for not too much money
About the size, yeah sure I can fit quite a large one in my 4x5meterish living room, but I have to consider WAF and also I personally wouldn't like a too big of a thing cluttering the space, that 55" TV is enough. But I'm aiming for maybe 40cm or maaybe 45cm at the widest, will still give nice directivity for most of the range.

I would LOVE to have a listen to a proper full range MEH yeah, but I have no idea where I would do that. You know anyone near Oslo with one?

 

[Chris A]

As far as I've found, the MEH is pretty much a perfect loudspeaker configuration in terms of its hi-fi performance. But not many people know/understand this.

Advantages:

1) full horn loading providing inaudible modulation distortion (AM and FM distortion side bands 20-24 dB below direct radiating), can handle any music transients realistically
2) full-range directivity down to the room's Schroeder frequency--meaning the loudspeaker sounds the same off-axis as well as on-axis
3) automatic time alignment if the crossover networks are designed without all-pass phase growth
4) point source presentation
5) for a fully horn-loaded loudspeaker--extremely compact size (with full-range directivity control without resorting to diffraction to achieve it)
6) except for the horn itself--very easy to fabricate the loudspeakers

Disadvantages:

none...if considering full range horn-loading/directivity control as part of what you get with the deal

There are actually more advantages, but the enumeration above is enough to put it into a class by itself.

The problem with doing one as a "fun project" is that you'll quickly find that it will outperform all other loudspeakers with similar quality drivers, especially if you spend a just a little effort to dial them in. The prototype K-402-MEH I built almost 10 years ago is one example of "dialing in". I actually would like it if I had four more of them in the setup to replace the other loudspeakers. You'll have to start making excuses for the other loudspeakers you might have around as to why they don't sound the same or as good.

I'm not that familiar with the prevalence of MEHs in your part of the world. Even here, the majority of Danley Synergies and SPL Unities are usually found in fixed installations at places of worship and stadiums, etc. Perhaps there are DIY versions in or around your country.

Most home hi-fi versions seem to be DIY--surprisingly, although there are a few that have Danley SH-50s, SH-60s, and perhaps SH-96s. The HRE1 (formerly called the Hyperion--reportedly now $46K USD per pair) is nothing but an SH-60 on top of a custom dual 15" direct radiating bass bin, with multi-amped DSP flattening of amplitude and phase, and plenty of on-board class D amplifier power. I think it would be much less expensive DIY--probably much less than 1/5th the price for everything including DSP, amplifiers, and sub-20 Hz performance.

The K-402-MEH prototype outperforms my two custom 1st-gen Jubilees in terms of polar directivity. It's actually the best loudspeaker in my setup--and is serving as the center in a 5.1 array (with two DIY TH subs).

Chris

 

[Tell]

Yeah this is what I've read up on MEH as well, it really seems to be the perfect point source. Well unless OP can convince us otherwise

But I know I've heard them before, I have taken pictures of them on a festival 11 years ago on a festival on the furthest tip of Lofoten where they where playing on a beach out into the Atlantic ocean. Looks like it might be some tapped horns underneath as well. I remember it sounding great but didn't really think too much about the speakers in particular back then. And I mean standing in the sweet spot with no reflections whatsoever to talk about almost any type of speakers can sound great if done right

But anyways, I understand acoustics quite well and I do get why those probably sound so very good, which is also why I want to build a pair. And perhaps my future small MEHs will be expanded into something even bigger when I realise I want more of that good stuff

While I have you on the line, what dispersion angle do you recommend for home usage? And do you recommend corner placement or is it fine to place them more like normal speakers (just making sure to take care of any early reflections of course)?

 

[voodooless]

Most certainly one can create a low distortion Meh, even for not that much money:

A Bookshelf Multi-Way Point-Source Horn

This originally started in this thread which asks the question "why aren't we building a bookshelf synergy monitor" and answers it with "we are" or at least "let's try and see what we get." I got inspired back in the Trynergy thread to build a semi bookshelf point source horn The results with...

www.diyaudio.com

 

[Chris A]

While I have you on the line, what dispersion angle do you recommend for home usage?

How wide and deep is your listening room? Generally, if you've got 4.5m wide x 5m+ deep x 2.7 metres high, I'd recommend a 90 x 60 degree horn. Otherwise, if your room is narrower or your ceiling is quite low (like a basement), then the coverage angles need to be smaller. That's why a lot of 50 and 60 degree MEHs are being used in EU-sized rooms (and I'm familiar with room sizes in the UK and Germany--but not so much in Scandinavia).

And do you recommend corner placement or is it fine to place them more like normal speakers (just making sure to take care of any early reflections of course)?

Well, believe it or not, my thinking on this has changed recently. Paul Klipsch said to always put them into corners, but Dave Griesinger (formerly of Lexicon and the inventor of Logic 7) says pull them in a bit--just until a stereo pair forms a strong phantom center image. With MEHs you can spread them out ~6 m before the phantom center begins to fade away due to in-room reflections beginning to take hold. You need to be within the "critical distance" of each of your loudspeakers. That's a function of directivity, time alignment, and point source capability.

Griesinger makes a very convincing case for stereo bass--all the way down to infrasonic frequencies. The only way to do that is to pull them slightly out of the corners--just enough to decouple side-to-side modes, but not enough to rob the front left/right loudspeakers of boundary gain below ~40 Hz (about 7 ft/2m). You can read more on that recent thread link that I participated in.

I get stereo bass in-room (but not stereo infrasonic bass below ~30 Hz). My left/right loudspeakers are within 0.5m of the side walls. I have directivity control down to ~100 Hz. Below that--it's all about room modes, etc.

Chris

 

[mwmkravchenko]

When you watch the video, there are legitimate questions raised, and issues raised with the driver's sensitivity. The distortion peak could be from two issues. The driver, the port velocity when in use, or a combination of the two. Might be a winter project for me to design and manufacture a MEH. I generally turn to the Le Cleach horn profile ( spherical wavefront launch ) when I am doing horn designs. But that kind of pushes this concept in the wrong direction. A conic is infamous for have resonances. That may be the source of the 1 kilohertz peak. Roughly a 34cm or 12 inch wavelength. You begin to have issues when you are generally a half wavelength.

Mark

 

[Chris A]

A conic is infamous for have resonances. That may be the source of the 1 kilohertz peak. Roughly a 34cm or 12 inch wavelength. You begin to have issues when you are generally a half wavelength.

I would question where the NFS microphone was placed during the SH-50 measurement--as Erin I believe never really detailed that portion of the measurement. The microphone simply being in a null with the nearfield horn wall and midrange/woofer port placement in such nearfield measurements could lead to some of the problems seen.

What bothers me is that the OP didn't consider this (even after I pointed out that this could be the source of the problems), and also didn't consider that he's basing most of his comments on one SH-50 measurement (NFS). (That certainly looks like decision bias to me in interpretation.)

The other MEH design presented--I'm not familiar with, and the visual presentation of it leads me to have more questions than answers. Basically two data points and a lot of questions.

Chris

 

[mwmkravchenko]

I would question where the NFS microphone was placed during the SH-50 measurement--as Erin I believe never really detailed that portion of the measurement. The microphone simply being in a null with the nearfield horn wall and midrange/woofer port placement in such nearfield measurements could lead to some of the problems seen.

What bothers me is that the OP didn't consider this (even after I pointed out that this could be the source of the problems), and also didn't consider that he's basing most of his comments on one SH-50 measurement (NFS). (That certainly looks like decision bias to me in interpretation.)

The other MEH design presented--I'm not familiar with, and the visual presentation of it leads me to have more questions than answers. Basically two data points and a lot of questions.

Chris

Hi Chris. Our thread starter has some valid measurements. All we have are conjecture. I have not tried to model the midrange section of that horn. But my guess was already typed.

Mark

 

[Chris A]

Hi Chris. Our thread starter has some valid measurements. All we have are conjecture.

Okay...let's take them one at a time (note that this took a bit of time to compile--so the next time, you can do it):

1. You may have only conjecture. The SH-50 that I acquired as a test article seven years ago has a lot of data. The one NFS measurement sequence done by Erin (not the OP) has problems. There are no other NFS measurements referenced in the YT video
   
   The "Solano" (?) isn't an NFS measurement. I see only ARTA being used.
   
   This isn't the first time that NFS has produced odd results that are not seen in anechoic or other valid test types.
   
   
2. I also see that the types and "problem areas" of measurements the OP concentrated on are bogus. The conclusions the OP professed are wrong or misleading.
   
   Here is a screen grab from 1':33":
   
   
   
   
   
   Here is an outside (quasi-anechoic) measurement of the K-402-MEH at 96 dB(1m) with a very cheap P.Audio BM-D750 compression driver that chatters above 10 kHz--which can be ignored:
   
   
   
   
   See any burst decay issues? I don't even using a cheap 2" titanium dome compression driver.
   
   Here is a screen grab from 1':55":
   
   
   
   
   
   Here is a HD measurement taken at 95 dB@1m on-axis of the K-402-MEH:
   
   
   
   
   See any local rises in HD, especially around the crossover frequency (~550 Hz)? You simply cannot generalize MEH behavior based on one or even two designs. I suspect that the local rise in second harmonic distortion (which is also NOT accompanied by audible IMD in a horn-loaded loudspeaker) was either an artifact of the measurement or perhaps something was wrong with that particular SH-50. I rather suspect the former is the culprit.
   
   And no--one cannot hear second harmonic (only) distortion at that level. But IMD is easy to hear--because it's inharmonic. That wasn't presented in the video, however.
   
   Here's a screen grab from 2'33":
   
   
   
   Do you still think this is a characteristic of an MEH--which is rated at 123 dB (continuous music)? I certainly don't. I suspect the NFS measurement. Also, this is 0.5 dB in a single very narrow band @102 dB. While I don't have an NFS to double check the measurement, this measurement would not be difficult to replicate for either the K-402-MEH or the SH-50 using REW and trace arithmetic. This is an issue? Not in my book. It's a fluke in a measurement--and certainly not audible even if it was the case for that SH-50.
   
   Here is a screen grab from 3':15":
   
   
   
   
   Doesn't anyone see the problems here with the OP's ("drive by") interpretation?
   
   a) First, all drivers on the horn aperture outside of the apex-mounted compression driver will see a first, second, and third notch reflection from the horn throat entrance. That's what you're seeing here. That is the natural low pass effect of the MEH design on the lower frequency drivers. It's supposed to do that. So what's the problem here?
   
   b) What you're not seeing is the horn gain--16+ dB gain over the direct radiating condition that the OP forgot to mention. Disingenuous? (You bet-no one could make that mistake honestly.)
   
   [I don't care about the "Solano" MEH, since it looks to have some real issues, so I'll spare you from going over that portion of the video--which uses ARTA--and not NFS.]
   
   
3. The OP also clearly doesn't understand compression drivers and horns.
   
   No one that I know uses the term "bottlenecks" to describe compression driver ports, and he clearly wouldn't have said what he did if he understood it. For me, the OP lost just his credibility when talking about anything with compression drivers and horns from my witnessing this video.
   
   No one that I know uses THD or HD to make their arguments when measuring horns--because the distortion that is really audible is IMD, not HD, and in a well-implemented horn, IMD is 20-24 dB lower in horns than using the same drivers in direct radiating mode at the same on-axis SPL. (Almost everyone misses this point.)

Chris

 

[gnarly]

Our thread starter has some valid measurements. All we have are conjecture.

I'm fine with considering the measurements as valid.
I'm not fine when deep pixel peeping conclusions are made from a single measurement set of a single speaker model, even if it's just for the purpose of drawing solid conclusions on that particular model. It pays not to be too certain about measurement ime.

For me, when I measure an issue with a speaker, it's time to try to isolate the issue's source, before drawing any conclusions on whether or not other units of the same model will likely have the issue too.
For instance, I would not just assume the distortion bump of the SH-50 tested, applies to all SH-50s. Maybe it does, but Erin's review said the unit was off-the-shelf which probably means a demo/rental. 900Hz is the bottom end of the CD's range, is the CD OK?
Or is the speaker totally fine for what it is, and the 900Hz bump is real, and simply part of the passive's electrical and acoustic design?
Who knows without further testing....and hopefully with the testing of another same unit to compare.


I'm totally not fine with the idea, that the measurements of only a couple of models/units of a particular type of speaker
......can be used to make universal assessments about that particular type of speaker. Talk about going out on a limb....

Anyway, here's another synergy sample....it simply is what it is, and is not meant to imply any universal properties of synergies
It uses a compression driver, four 4" mid-range, and two 12" lows. So 6 ports in a 90x60 conical horn, to interact/interfere with CD and each other.
Designed for >100Hz and use with sub. 36"x22" size. It's not my most refined synergy measurement wise, as it doesn't have secondary horn flares to reduce mouth diffractions. But it's smaller than ones that do have the flares, and I can fit three in my room for LCR.....which is more important to me than perfect measurements.

Just grabbed some quick 1m indoor non-gated, no FDW measurements, to look at SPL levels, and distortion measurements

Here's 86dB distortion. Arbitrarily picked 1.3kHz

And 116dB, same 1.3kHz reference

Anyway again....Just another MEH sample....and there is nothing about it I'd be willing to extrapolate onto other MEHs in general.
Other than maybe I love how it sounds

edit: if any are wondering about the very flat acoustic frequency responses in all the above, it is because i deliberately did a quick flat response FIR tuning at 86dB/1m to a reference axis. Because I think that makes it's easy to see if compression occurs as SPL is raised.
Here's the individual responses of the 4-ways at 86dB, with their summation in black. 1/24th

 

[Ciobi69]

I would like to make a pair to ‘’slim’ my system I got the kef blade 2 meta and they are awesome , but expensive I would like to monetize and do the Scott jmod synergy , I already have subs , I think they should sound similar, the meh horn have a good vertical and horizontal directivity as my speakers, if somebody could help me in choosing

 

[Chris A]

You've got a couple of DSP guys here (Mark/Gnarly and myself). If I remember, the MEH plans you mention use a passive crossover.

Perhaps another with direct experience building those will chime in. You can also check over on the Multiway forum at diyAudio.com for plan builders of that MEH.

Chris

 

[Tell]

I would like to make a pair to ‘’slim’ my system I got the kef blade 2 meta and they are awesome , but expensive I would like to monetize and do the Scott jmod synergy , I already have subs , I think they should sound similar, the meh horn have a good vertical and horizontal directivity as my speakers, if somebody could help me in choosing

That would be an interesting swap, Kef Blade 2 Meta to a pair of diy MEHs, keep us posted!

 

[HiMu]

If I remember, the MEH plans you mention use a passive crossover.

Perhaps another with direct experience building those will chime in. You can also check over on the Multiway forum at diyAudio.com for plan builders of that MEH.

Both the Scott Hinson's MEH and JWSounds JMOD are designed for 3ch amplifier with dsp, but can be used with the B&C FB464 crossover if only 2ch amp with dsp is available. https://www.bcspeakers.com/en/products/crossovers/0/16/FB464

I'm building the Hinson's MEH, all the parts are ready but still waiting for the speaker elements to arrive, hopefully at the end of next week.

 

[mwmkravchenko]

I think that it's time I take a crack at a truly high-end one. I started, but got distracted. No compression drivers for that. Either a high efficiency planar or a dome tweeter manifold. THe compression drivers have way too much high frequency distortion for me.

Mark

 

[stoneeh]

I would question where the NFS microphone was placed during the SH-50 measurement--as Erin I believe never really detailed that portion of the measurement. The microphone simply being in a null with the nearfield horn wall and midrange/woofer port placement in such nearfield measurements could lead to some of the problems seen.

From Erin's review:

The reference plane in this test is directly in front of the tweeter. I also tested with the reference set at the “baffle”. I took the speaker outdoors to perform a ground plane test (at 2 and 4 meters) to verify where the acoustic center really was and it turns out that it is - indeed - at the tweeter. Therefore, the results you see here are with this as the reference point. I also determined that the nearfield-to-farfield transition for this speaker occurs at approximately 1.50 meters. Meaning that at distance closer than 1.50 meters the speakers do not all sum as a coherent wavefront; you must be at least 1.50 meters distance away for this to occur.

I know some folks who believe this wouldn't even matter though, since "it's a single point source from zero distance out anyway":

The outcome literally appears to be, measures like, and sounds like a single crossover-less driver.

An example of its effect is that you can walk up to the speaker, move around, literally even put your head deep into the horn mouth of a three-way Synergy horn, and never hear any clue there is more than one driver.

Source: https://audioxpress.com/index.php/article/q-a-tom-danley

 

[Geert]

If you want to test another one, test the design that Scott Hinson made, that is wel documented, tested and proven by many diy builders

Do you know of a good measurement of the low end performance of this design (below 200Hz, the lower design frequency of the horn)?

 

[Chris A]

I think that it's time I take a crack at a truly high-end one. I started, but got distracted. No compression drivers for that. Either a high efficiency planar or a dome tweeter manifold. THe compression drivers have way too much high frequency distortion for me.

Mark

Be careful of the nonlinear effects in flexible planar drivers due to higher dynamic loading in a horn throat environment. This is why planar electrostatic and ribbon drivers, as well as flexible planar drivers (like Magnapan) are not used in horns.

An AMT will probably work best, or perhaps a dome driver of more robust construction (i.e., rigid dome), but be careful of the maximum SPL that it sees in operation.

Chris

 

[mwmkravchenko]

Be careful of the nonlinear effects in flexible planar drivers due to higher dynamic loading in a horn throat environment. This is why planar electrostatic and ribbon drivers, as well as flexible planar drivers (like Magnapan) are not used in horns.

An AMT will probably work best, or perhaps a dome driver of more robust construction (i.e., rigid dome), but be careful of the maximum SPL that it sees in operation.

Good advice. I have designed everything you mentioned but a compression driver. That is not easy to to do well. As for non-linearities, they are there as always and there are ways to do it well. An AMT is not really a good candidate for this application. As for planars not being used on horns I can think of quite a few that are. Alcons does, there are horns for a few not well made Chinese knock offs. Akiplab used to make reasonably decent planars, now they are hit and miss. No QC. When done well a planar is one of the cleaner ways to reproduce high frequency sound. I have sat in front of community Beryllium compression drivers and had nothing negative to say. But they are not being made anymore as far as I know. And crazy expensive.

Mark

 

[gnarly]

You've got a couple of DSP guys here (Mark/Gnarly and myself). If I remember, the MEH plans you mention use a passive crossover.

Perhaps another with direct experience building those will chime in. You can also check over on the Multiway forum at diyAudio.com for plan builders of that MEH.

Chris

Hi Chris, the JMOD documentation shows tri-amp active filtering presets for each of it 3 sections, and describes the use of B&C's passive crossover for the DCX464 as optional, for a bi-amp setup. I think John does a really nice, and generous, job of documenting his build. https://www.jwsound.live/designs/jmod

He uses a more convention processing technique than either of us do. I'd call it a more traditional use of asymmetric high-pass, low-pass, and PEQs & shelves.
Many ways to get there, huh?
And yes, like you have helped many with your PEQ strategy, I'll be happy to help if anyone wants to go down the complementary lin-phase crossover path.