Determine the distance between the centers of the two midrange drivers and determine the wavelength of that.
If we use two 5” mids and a Mundorf AMT in an MTM as an example, we have a center to center spacing minima of 9 inches …….thats a max crossover frequency of 1500hz………much too far for 5k regardless of the slope.
Why does Tekton use so many tweeters?
Many drivers start distorting at about half xmax so doubling shares the burden and gives more clean headroom. If you ever heard a single driver vs a line array of the same driver you will know what I mean. With the array you can crank the volume and you can go insanely loud without ear discomfort.
Added bonus is the looks....there is something about symmetry that is very pleasing.
The Pluvia is a 4" and the Mundorf 6cm so the center to center is more like 17cm.
I am slowly coming around to the fact that MTM may not be the best, especially with 1st order filters. Thanks everyone for your nudges and constructive criticism.
This post over on AVS forum lays it out well:
18Hurts
4983 posts · Joined 2015
#2 · May 25, 2021
You asked!
A D'Appolito Array--or MTM if you like has a set of rules that Joe published but is generally not followed by manufacturers typically because of (wait for it) cost and complexity.
Generally speaking that design was only meant to be done vertically and follows some rules as far as center-to-center distance between the drivers at the crossover point. It works very well when the rules are followed and not so well when they are ignored. Ever see one laid on it's side? If so, those things are not following the rules and when done wrong, make a mess of things.
The "rules" if you will demand that the midranges "match" very closely and if any way possible, the midrange to tweeter crossover point center-to-center distance should be within 1/3rd of the crossover wavelength....or as close as possible.
For example, say you have a pair of 4 inch midranges and a 1" dome tweeter to cobble together. Stack up the pair of fours and the dome then measure the center-to-center distance to get the proper crossover frequency. Say, for example you can have a center-to-center distance of 7 inches. Multiply that by 3 and whatever a 21 inch wavelength is...done! Roughly, that is 635Hz or so.
Uhhh.... there are no 1" dome tweeters that crossover that low so how far can you push it until lobing becomes obvious? Well, about one full octave works decently enough or around 2KHz which can be done with rather expensive and stout dome tweeters. Thanks to some recent dome tweeters, you can cross them over at 1,500 Hz so a pair of 5" mids can be used and crossed over at less than an octave center-to-center distance without too much lobing becoming noticeable.
Cost? Well, those tweeters that crossover that low are not cheap, they can struggle with those low frequencies at higher SPL since two mids are naturally 3dB more efficient and considering two mids can handle twice the power--the tweeter should be able to output 6dB more output to "keep up" with the dual mids. Not only does the dome have to be more efficient to "keep up", it has to operate at the low frequency limits at a higher peak SPL...that costs money.
Money? Yeah, I built a D'Appolito array with 5" mid-woofers, selected a 1850Hz crossover point and used a steeper crossover filter (18dB/Oct) to protect the 30mm stout dome tweeter. I paid three times the cost for that low frequency output of the tweeter (compared to 3,000Hz) and the passive crossover parts were twice as much being 18dB/Oct 1,850Hz VS 12dB/Oct 3,000 Hz. Of course, you are spending twice as much on the mids because they are doubled up so. Tweeter cost tripled, crossover component cost doubled and mid-woofer cost doubled so quite expensive.
Could you get better results, about the same efficiency/power handling with a 6.5" mid-woofer and crossing the tweeter higher at 2.5 KHz with a much lower cost passive crossover? Could you get the same or better efficiency, peak SPL, power handling with an 8 inch mid-woofer crossing over at 2 KHz and use a waveguide to get the dispersion to match at 2KHz at around the same price?
In audio, there are certain styles and designs manufacturers push because they sell! They don't work for free, want to stay in business so will produce things that the public wants. The D'Appolito MTM array was originally done back in the 1980's as a way to mimic a point source full range speaker--like a coax but without the problem of IMD mucking up the treble response as the soundwaves of the tweeter bounce off the moving midrange cone. The two mids would couple at a specific frequency which cut the vertical dispersion to prevent floor/ceiling bounce roughly how a multi-driver vertical line array works. It works well when done correctly but for the most part, is the price worth the benefit?
The D'Appolito Array worked well when done right but Joseph D'Appolito himself stated it was time to move forward as technology, driver improvements and measuring accuracy have really taken off since the early 80's. Basically, it was a good design for the time but you can do better at the same or lower cost. Since Joe designed it, him stating it is time to abandon his own design is rather interesting...I get it.
I've owned a D'Appolito Array speaker for 17 years, purchased it in the late 90's. Worked very well for the most part and I built one of them 10 years ago. Five years ago, I completely redid my multi-channel system and did not go with MTMs--nothing personal, used waveguides to control dispersion and more efficient, higher power handling drivers to get the SPL I required without requiring arrays.
I'd say the reason you are seeing a decline in the amount of D'Appolito arrays is driver and crossover filter costs, more build complexity at a performance level that allows better options. As always, manufacturers have to be different from each other so the styles change as with most things. Don't forget, Klippel released a robot that will take thousands of measurements to determine speaker accuracy so manufacturers are responding by improving the off-axis and lobing/beaming performance of their speakers in response. The days of talking heads and a simple measurement are going away as speaker designs are now put under the robotic machine overlords producing massive amounts of accurate data and the audio companies have to focus on performance more these days. The markets are shifting east and coupled with actual accurate measurements, the days of using your ears is eroding away. An MTM done wrong is obvious with measurements so why risk it?
The farther you get from a point source, the worse things get.......for a valid reason......and that's phasing
As an engineer, working with multiple microphones in proximity on the same source, the effects of phase shifts and cancellations becomes easily discernable.......and once you've heard it, you can't un hear it.........and this has spoiled my listening enjoyment as nearly every speaker i've ever heard has some phase issues.
These midranges I found myself :
Faital Pro 5PR120
Faital Pro 5PR160
PHL 950ND
BMS 5N160
B&C 5MDN38
Faital Pro 5PR120
Faital Pro 5PR160
PHL 950ND
BMS 5N160
B&C 5MDN38
I regrettably cannot share details as they are customs from a commercial speaker (Daudio S1). They are rigid, have a very low Fs and a very high linear xmax. So they can move a lot of air. Thick enough baffle recommended given this. Four of these woofers probably would cost 4000-4500 euro today. Significant money, but for open baffle usage I am not aware of better woofers. Can bring you in contact with the seller if you want...
I did half a day of freestyle combining drivers I have.....freestyle just hanging from wires in the air.
To my surprise I found that with a little eq to flatten the rising response I preferred the naked nd3st over the ESS Heil. I tried 4 different horns as well but naked sounded best. It improved even more when I made a small waveguide out of porous packing foam. It is as airy as the Heil but has better dispersion and more body.
Instead of MTM I am now looking at doing:
+2000Hz Semi naked 1.4" compression driver 18sound nd3ts
400-2000Hz 6.5" or 7" midrange
80-400Hz 15"
<80Hz 2x 21" or 4x 18"
So now I am on the lookout for a approximately 7" midrange. I tried some full range drivers I had at hand (5-6.5") but they all start to sound shouty at higher volumes. An old peerless 7" SDS woofer did a lot better here but I am sure this can be improved upon.
Any suggestions for a 6"-7" low distortion and robust dedicated midrange/wideband driver with high sensitivity to cover 400-2000 with 1st order slopes? (so with a very wide usable frequency range of let's say 100-8000Hz) . That is almost full range territory. Something in the 100-200 usd range?
I still am interested in the Alpair 12P but I never heard a Mark Audio driver. So I don't know if they get shouty when pushed hard.
The Satori MR16P looks very nice but I would like a bit more than 88db sensitivity.
To my surprise I found that with a little eq to flatten the rising response I preferred the naked nd3st over the ESS Heil. I tried 4 different horns as well but naked sounded best. It improved even more when I made a small waveguide out of porous packing foam. It is as airy as the Heil but has better dispersion and more body.
Instead of MTM I am now looking at doing:
+2000Hz Semi naked 1.4" compression driver 18sound nd3ts
400-2000Hz 6.5" or 7" midrange
80-400Hz 15"
<80Hz 2x 21" or 4x 18"
So now I am on the lookout for a approximately 7" midrange. I tried some full range drivers I had at hand (5-6.5") but they all start to sound shouty at higher volumes. An old peerless 7" SDS woofer did a lot better here but I am sure this can be improved upon.
Any suggestions for a 6"-7" low distortion and robust dedicated midrange/wideband driver with high sensitivity to cover 400-2000 with 1st order slopes? (so with a very wide usable frequency range of let's say 100-8000Hz) . That is almost full range territory. Something in the 100-200 usd range?
I still am interested in the Alpair 12P but I never heard a Mark Audio driver. So I don't know if they get shouty when pushed hard.
The Satori MR16P looks very nice but I would like a bit more than 88db sensitivity.
I have experience with the Alpair 10 and it can be used half power when used fullrange before it start to get shouty.
Maybe use the Beyma 6G40ND in stead.
Maybe use the Beyma 6G40ND in stead.
I was looking at the Beyma. Looks very good indeed. Perfect frequency response & sensitivity.
If only the Purify PTT6.5M08-NFA-01A were half price...... Except for their real world sensitivity of 89 db they look perfect.
You could also consider 8" for mid? Probably better in case high volume is required. Easier to cross with 15" and could still mate with Ess.
I am very interested in that driver but for this application it is too big. At 4khz it is already down 20db @30 degrees off axis. It would Crete a big hole in my directivity plot.
The Beyma 6G40ND still looks the best for 400-2000k with 1st order slopes. The dotted line is 30deg off axis.
On paper the Beyma looks perfect. Now the only question is if it sounds good. If I believed measurements say everything this is the one.
I don't think it would run out of steam either crossing at 400hz. It can still produce 113db @ 200hz
Are there more driver suggestions? This is arguably a crucial driver covering most of the telephone band frequencies so it can make or break the speakers.
Last edited:
Also I will experiment with using the midrange driver open baffle, no baffle, aperiodic and sealed......combined with either a single compression driver or back to back out of phase compression drivers. So I will a/b test different arrangements rangeing from fully dipole tobhalf dipole to only dipole bass.
I am mainly interested in open baffle bass below the Schroeder frequency....let's say 200 Hz in a large room.
I see many people use open baffle with a sealed/reflex bass. I would do the opposite. I could even try the 15" aperiodic and the 6" sealed.
I am mainly interested in open baffle bass below the Schroeder frequency....let's say 200 Hz in a large room.
I see many people use open baffle with a sealed/reflex bass. I would do the opposite. I could even try the 15" aperiodic and the 6" sealed.
Sawasdee @ErikNils r u in Bkk? I lived there in my teenage years, considered home for a long time. Still occasionally miss some aspects of Thailand. Or course, I miss the food, tho I cook it myself often. 😉
If you're looking to use these speakers for something like PA or DJ work (play loud in a large room, you said in an earlier post, right?), examine Xmax, power handling & max SPL. Loudspeakerdatabase.com gives you a graphs for max SPL through the frequency range.
The SB Acoustics SATORI MW19P might be worth considering for that mid driver. Ditto WO24TX -- this has a robust motor & excellent extension on both freq ends. The latter is my choice for an end-game speaker for a friend.
The SB Audience drivers are also worth studying. They appear to be under-rated & not well used. Since you have the DSP power in the Flex 8, you don't need to stick to old T/S "rules". Some of the SBA 15" & larger pro woofers are quite impressive, and at least here in Canada, not outrageously priced. I recently used a Nero-15SW800 in a sealed sub that I'm still tuning with DSP.
If you're looking to use these speakers for something like PA or DJ work (play loud in a large room, you said in an earlier post, right?), examine Xmax, power handling & max SPL. Loudspeakerdatabase.com gives you a graphs for max SPL through the frequency range.
The SB Acoustics SATORI MW19P might be worth considering for that mid driver. Ditto WO24TX -- this has a robust motor & excellent extension on both freq ends. The latter is my choice for an end-game speaker for a friend.
This is a new project for my friend Mark who has an enormous "great room"
30x52' with vaulted peak ceiling >20' high. The goal is not to fill the entire space with sound but the LR portion, which is about a quarter of the total floor space. Mark seeks a game-ending system, with some financial limits. He also happens to be a master metals fabricator with a lot of skills & tools, which will be brought to play.
I post this as a reality-check / peer-review of my plans.
This pic shows a fraction of the big room.
Initially, Mark wanted scaled-up versions of my LX521...
30x52' with vaulted peak ceiling >20' high. The goal is not to fill the entire space with sound but the LR portion, which is about a quarter of the total floor space. Mark seeks a game-ending system, with some financial limits. He also happens to be a master metals fabricator with a lot of skills & tools, which will be brought to play.
I post this as a reality-check / peer-review of my plans.
This pic shows a fraction of the big room.
Initially, Mark wanted scaled-up versions of my LX521...
The SB Audience drivers are also worth studying. They appear to be under-rated & not well used. Since you have the DSP power in the Flex 8, you don't need to stick to old T/S "rules". Some of the SBA 15" & larger pro woofers are quite impressive, and at least here in Canada, not outrageously priced. I recently used a Nero-15SW800 in a sealed sub that I'm still tuning with DSP.
Sawasdee krab! I am up north in Lamphun
Those are gorgeous drivers but I feel they are too large for a 2000hz 1st order crossover. The 6.5" satori looks very nice too but gives up too much max spl. I try to get a driver that at least does >110db
Ah Lamphun. I passed through there around 2008; my sis organized a 2+ week tour with old friends. all around the west & north.
Re the 8" drivers & 1st order: you limit yourself a lot with that. Even LR2 would help. I use mostly LR4 routinely & if it means using the more appropriate driver, that's usually a better compromise than the wrong driver with the crossover you prefer.
Re the 8" drivers & 1st order: you limit yourself a lot with that. Even LR2 would help. I use mostly LR4 routinely & if it means using the more appropriate driver, that's usually a better compromise than the wrong driver with the crossover you prefer.