I see a few designs doing this like Wilson, the Salon 2 etc.... what's the practicality of this approach?
Seems like much more complexity that is not worth it because, take the Scan speak mids for example, looks like even the smaller midrange drivers do not perform much better above 3k.
Seems like much more complexity that is not worth it because, take the Scan speak mids for example, looks like even the smaller midrange drivers do not perform much better above 3k.
Well the big key issue or rule to observe is drivers C2C spacing according to crossover frequency wavelength. You want to be under half the WL C2C of the xover. The other issue is the gain of having 2 mid drivers and how / where in frequency they start to develope array gain. Most of the time its not worth the hassle, but if you need alot of lower midrange dynamic capability, its a good approach. A 2.5 way design may be better in alot of cases where more output is needed.
In speaker building there is not allways a better or worse. If we look at commercial speaker, marketing is more important than objectivity.
More chassis in a speaker usualy make it more expensive in the line up. Doubling up on midrange drivers is relatively simple, if you already have a product with a single one.
With DIYS the same may be the cause, but also the lower load and lower distortion for a given SPL and the higher dynamics of doubled cone area. Crossing over a tweeter 1000Hz lower may make it sound better, too. The dispersion of two midrange driver in M-T-M may be desired as well.
You can find all kinds of pro's and con's if you want to. Pick your own!
More chassis in a speaker usualy make it more expensive in the line up. Doubling up on midrange drivers is relatively simple, if you already have a product with a single one.
With DIYS the same may be the cause, but also the lower load and lower distortion for a given SPL and the higher dynamics of doubled cone area. Crossing over a tweeter 1000Hz lower may make it sound better, too. The dispersion of two midrange driver in M-T-M may be desired as well.
You can find all kinds of pro's and con's if you want to. Pick your own!
When you look at the filter of the Kef 104/2 ref mid-80s design, the soundstage was pretty good & full. Filter was complex though.
But having two low diameter indeed in spite of a bigger gives more headroom : the lower diameter allow them to have a higher cut-off in the treble range and at the same time crossing lower (2 cones move twice less at iso spl than one). It can gives also the needed spl if a single have not sensivity enough, to manage better the baffle step frequency. two mid in // gives + 6 dB sensivity which can be the loss in the low mid for most of the baffle widths because of todays narrow collums loudspeakers.
Nowadays people prefer bigger standalone mid and huge wave-guide for the tweeter to match better the two size in the pass-band. that can make too larger baffle size width, a 6" is bigger than two 4" at iso surface. but you can do that with MTM & MMT as well : https://heissmann-acoustics.de/dxt-wave/
I was wondering though if littlier MTM were not more confortable to listen to in farfield than MT.
In cm, WL is : 346/frequency wich you want to know the length. 346 m/s is the speed of sound around 22°/C air.
for instance 346/2000 (hz) : 0.173 m , so 17.3 cm WL . Ideally you want the center of the mid and the tweeter to be at 1/4 WL... often impossible. 1 to 1.2 WL can give better result than all what is between 1/2 and 1. There are soft to sim that : VituixCad, Basta, ...
Now it migth be a little different I surmise with a MTM that has a tulipe patern dispersion shape...
https://suesskindaudio.de/en/ara/
But having two low diameter indeed in spite of a bigger gives more headroom : the lower diameter allow them to have a higher cut-off in the treble range and at the same time crossing lower (2 cones move twice less at iso spl than one). It can gives also the needed spl if a single have not sensivity enough, to manage better the baffle step frequency. two mid in // gives + 6 dB sensivity which can be the loss in the low mid for most of the baffle widths because of todays narrow collums loudspeakers.
Nowadays people prefer bigger standalone mid and huge wave-guide for the tweeter to match better the two size in the pass-band. that can make too larger baffle size width, a 6" is bigger than two 4" at iso surface. but you can do that with MTM & MMT as well : https://heissmann-acoustics.de/dxt-wave/
I was wondering though if littlier MTM were not more confortable to listen to in farfield than MT.
In cm, WL is : 346/frequency wich you want to know the length. 346 m/s is the speed of sound around 22°/C air.
for instance 346/2000 (hz) : 0.173 m , so 17.3 cm WL . Ideally you want the center of the mid and the tweeter to be at 1/4 WL... often impossible. 1 to 1.2 WL can give better result than all what is between 1/2 and 1. There are soft to sim that : VituixCad, Basta, ...
Now it migth be a little different I surmise with a MTM that has a tulipe patern dispersion shape...
https://suesskindaudio.de/en/ara/
Last edited:
Let's say you have 2x 5" drivers and a typical 100 - 110mm (4 - 4.5") tweeters.
Closest you can get the mids to each other with tweeter in the middle is 2x (1/2 of mid diameter) + tweeter diameter in between.
Take 13560 and devide that C2C number (roughly 10"), which gives you a max xover of roughly 1400 hz. This frequency will prevent off vertical axis cancelation.
A smaller faceplate tweeter around 3 inches diameter can help raise the possible xover frequency up a little to 1500 hz. Either way you need a capable tweeter that can handle that low of an xover. The outcome of the response curve will interact with diffraction off the cabinet edges as well. Sometimes increasing xover point to about 2/3 to 1-1/3 WL can help, but it depends on your tweeter dispersion and other factors. A crossover simulation would be best to perform to get it right.
An MMT arrangement may be better and can help raise the xover higher up if the mid driver can support it. That way only one mid driver is doing the upper mids and is closer to the tweeter.
The only main reason to do an MTM is for limited vertical dispersion in low ceiling rooms with excessive early reflections but the off axis response will make the speaker sound scooped in the mids and shouty. Not good IMO
Closest you can get the mids to each other with tweeter in the middle is 2x (1/2 of mid diameter) + tweeter diameter in between.
Take 13560 and devide that C2C number (roughly 10"), which gives you a max xover of roughly 1400 hz. This frequency will prevent off vertical axis cancelation.
A smaller faceplate tweeter around 3 inches diameter can help raise the possible xover frequency up a little to 1500 hz. Either way you need a capable tweeter that can handle that low of an xover. The outcome of the response curve will interact with diffraction off the cabinet edges as well. Sometimes increasing xover point to about 2/3 to 1-1/3 WL can help, but it depends on your tweeter dispersion and other factors. A crossover simulation would be best to perform to get it right.
An MMT arrangement may be better and can help raise the xover higher up if the mid driver can support it. That way only one mid driver is doing the upper mids and is closer to the tweeter.
The only main reason to do an MTM is for limited vertical dispersion in low ceiling rooms with excessive early reflections but the off axis response will make the speaker sound scooped in the mids and shouty. Not good IMO
Last edited:
Ahh awesome ....
Does half the WL C2C of the xover need to be observed with one mid ?
How do I calculate for one ?
Im currently aiming for the Scan-Speak 18M/4631T Revelator 7
Does half the WL C2C of the xover need to be observed with one mid ?
How do I calculate for one ?
Im currently aiming for the Scan-Speak 18M/4631T Revelator 7
The larger Wilson speakers have what looks like an MTM arrangement, but it is a non-typical MTM because (as the original poster said) the mid drivers are different. This is a picture of the Alexx V upper MTM section, not shown is the 10 + 12 woofer section. The upper is a 15 cm driver, the lower is an 18 cm driver. Each driver has its own enclosure and its own baffle step / diffraction signature. All three drivers are positioned independently in the z-direction to manage the delay of each driver. Wilson says this speaker is a 3-way, and the two mids have broad overlap. We can assume that the upper and lower mids each have their own filter network, so there is potentially a lot of tailoring and optimization. Afterall, at this price level there is no real cost constraint on the design. I am not aware of any other company that uses an MTM arrangement with two different mid drivers.
The Salon 2 Ultima is different. It is a 4 way speaker, 3x8 + 6 + 4 + 1. This would seem to indicate that the 6" and 4" driver have a crossover frequency between them and there is not a broad overlap between them.
The Salon 2 Ultima is different. It is a 4 way speaker, 3x8 + 6 + 4 + 1. This would seem to indicate that the 6" and 4" driver have a crossover frequency between them and there is not a broad overlap between them.
"The Salon 2 Ultima is different. It is a 4 way speaker, 3x8 + 6 + 4 + 1. This would seem to indicate that the 6" and 4" driver have a crossover frequency between them and there is not a broad overlap between them."
Right this would be the design I would entertain....
Right this would be the design I would entertain....
So for starters, a 4" would allow you a higher crossover frequency than a 6". Maybe they wanted or needed the tweeter to cross higher. Some would argue that a 4" should have no problem covering all the midrange and handing off to 8" woofers. They may have been trying to limit the band that each driver covered, or maybe something else?
I would only assume the 8" drivers are much closer to being subwoofer
or use very low crossover point.
So the 6" midrange is used to cover more bass content.
Then a second smaller midrange is used, which helps further reduce tweeter
distortion. By allowing a much higher crossover for tweeter.
The real advantage I see is it allows smaller Tweeter .
Since 1" tweeter still limited to directivity from large cone surface.
you could use 3/4 or smaller tweeter. Which demands higher crossover.
But provides less directivity than common 1" driver
Another way to think of it if your familiar with 2 way systems.
Which use a 8" for bass and a 3" fullrange for highs.
Sound is more impressive than most assume.
But for " Ultra" Hifi most would assume that such a 2 way
would need a sub added and also a tweeter.
So bingo bango.
A system which easily supports a very large sub, and very small tweeter.
with very good midrange detail and low distortion.
Maybe for us " strange " Rebels who listen to Classical or Jazz music
at absurd Heavy Metal sound pressure levels...LOL
I find 3" mid to not be suitable for Cellos and Drums at high sound pressure
levels.
Also find 6.5" mid not suitable for violins and horns at high pressure levels.
So dual midrange seems feasible.
or use very low crossover point.
So the 6" midrange is used to cover more bass content.
Then a second smaller midrange is used, which helps further reduce tweeter
distortion. By allowing a much higher crossover for tweeter.
The real advantage I see is it allows smaller Tweeter .
Since 1" tweeter still limited to directivity from large cone surface.
you could use 3/4 or smaller tweeter. Which demands higher crossover.
But provides less directivity than common 1" driver
Another way to think of it if your familiar with 2 way systems.
Which use a 8" for bass and a 3" fullrange for highs.
Sound is more impressive than most assume.
But for " Ultra" Hifi most would assume that such a 2 way
would need a sub added and also a tweeter.
So bingo bango.
A system which easily supports a very large sub, and very small tweeter.
with very good midrange detail and low distortion.
Maybe for us " strange " Rebels who listen to Classical or Jazz music
at absurd Heavy Metal sound pressure levels...LOL
I find 3" mid to not be suitable for Cellos and Drums at high sound pressure
levels.
Also find 6.5" mid not suitable for violins and horns at high pressure levels.
So dual midrange seems feasible.
Your question seems too vague to me...two midranges in WHAT approaches?
The DIY Statement II speakers use two midranges in a WMTMW arrangement because they needed two Vifa NE123w's to get the SPL needed to match the two Dayton Audio RS225s they use for bass. When they (Jim Holtz and Curt Cambell) upgraded from the Statement IIs to the Bourdeaux WWMT they only needed one Accuton to reach the necessary SPL.
For a two (or 2.5) way speaker the MTM or TMM is to get higher SPL levels and double the displacement. An 8" two way is difficult to design because of beaming and challenging center-to-center spacing, so a MTM or TMM with dual 5" or 6" drivers can get you the same SPL and move roughly the same amount of air.
Whether you do an MTM, TMM or TMM 2.5 way depends on the design goals that you set and the tradeoffs that you are willing to make.
The DIY Statement II speakers use two midranges in a WMTMW arrangement because they needed two Vifa NE123w's to get the SPL needed to match the two Dayton Audio RS225s they use for bass. When they (Jim Holtz and Curt Cambell) upgraded from the Statement IIs to the Bourdeaux WWMT they only needed one Accuton to reach the necessary SPL.
For a two (or 2.5) way speaker the MTM or TMM is to get higher SPL levels and double the displacement. An 8" two way is difficult to design because of beaming and challenging center-to-center spacing, so a MTM or TMM with dual 5" or 6" drivers can get you the same SPL and move roughly the same amount of air.
Whether you do an MTM, TMM or TMM 2.5 way depends on the design goals that you set and the tradeoffs that you are willing to make.
maybe a 6" ins spite of a 8" will have less low end that pollute the bass driver ? ligther cone as well and also a higher break-up than a 8" for crossing to a 4".
I had that case in a 3 ways Harbeth I made 8+3+0.75 sligthy horned. The trade off because of the 8" driver and its early break-up was not to cross it over 500 hz to avoid the upper resonance polutte the 4" and stilll stay passive with not a too big electrical slope tat are also more Z in front of the bass driver. Many also want than the bandwidth of a mid to be large enough between its high pass and low pass and have no cut-off in the mid higher than 500 hz. So I assume a 4" is the perfect size for a 500 hz to 2500/3000 hz without suffering of the first close break-ups of the woofer often seen with modern hard cones as soon as 4000 hz !
It is possible that WMTm are indeed true 4 ways !
You have to factor in off axis response from your mids to determine a good xover mid LP and tweeter HP.
Again, here the 1/2 WL rule applies. In practice, a 6 inch driver will easily stay pistonic to a bit over 1k, but it will start beaming around that point. Alot of times it depends on the type of mid driver used including the cone geometry, depth, etc. A 4 inch driver in MTM will be good with a 2 to 2.5k xover depending on slope used. D'appolito's research determined a 3rd order LP slope was desirable for a good compromise in lobing, but some of those findings have been updated over the years. In theory a third to half WL xover is best but highly impractical if you want to use the most of the mid driver bandwidth and not stress the tweeter.
As others said, a 3/4" tweeter will work better for reduced spacing if you use dual 4" drivers and those should easily cope with 500 hz HP 2nd order. A really good 10" or 2x 8" should work well as a 3 way for that. You will end up needing some sort of midrange notch filter to flatten out the mid FR bump in most cases and use a lower LP on the lower of the dual LF drivers. The thing you'll have to watch out for then using a dual LF driver is too much lower midrange and standing waves in an enclosure of that height causing FR ripple.
Again, I would highly advise using simulation software first to frame out a practical design approach or build an already proven MTM design like the Seas Bragi kit (which is excellent BTW).
Again, here the 1/2 WL rule applies. In practice, a 6 inch driver will easily stay pistonic to a bit over 1k, but it will start beaming around that point. Alot of times it depends on the type of mid driver used including the cone geometry, depth, etc. A 4 inch driver in MTM will be good with a 2 to 2.5k xover depending on slope used. D'appolito's research determined a 3rd order LP slope was desirable for a good compromise in lobing, but some of those findings have been updated over the years. In theory a third to half WL xover is best but highly impractical if you want to use the most of the mid driver bandwidth and not stress the tweeter.
As others said, a 3/4" tweeter will work better for reduced spacing if you use dual 4" drivers and those should easily cope with 500 hz HP 2nd order. A really good 10" or 2x 8" should work well as a 3 way for that. You will end up needing some sort of midrange notch filter to flatten out the mid FR bump in most cases and use a lower LP on the lower of the dual LF drivers. The thing you'll have to watch out for then using a dual LF driver is too much lower midrange and standing waves in an enclosure of that height causing FR ripple.
Again, I would highly advise using simulation software first to frame out a practical design approach or build an already proven MTM design like the Seas Bragi kit (which is excellent BTW).
Last edited: