Why can't a MMT be 2-way?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello,
browsing the web and this forum for info about MMT quickly gives the answer that a MMT would be better off as a 2.5-way. An that's of course because of the distance from the lower woofer to the tweeter. But working a bit with a line array and the theories there I run into contradictions.
The result of stacking many drivers is that they act as one, taller driver within the range that the cc distance is less than 1 wavelenght. Now, how does this translate to a MMT? Why is the cc distance from respectively woofer to the tweeter still important when the 2 drivers act as one? Where is the center in this tall soundfield caused by the 2 drivers?

Also a second question: are the conditions for a MMT different with a rectangular tweeter (AMT, ribbon, planars) because of their more narrow vertical dispersion?

Regards,
Anders
 
frugal-phile™
Joined 2001
Paid Member
Sure an MMT can be a 2-way. But if you need to compensate for baffle-step a 2.5 way is a good way to use the same set of drivers.

Ideally centre-centre distance between midwoofers and tweeter should be less than a 1/4 wavelength for really good integration. This is next to impossible to do without a coax or a cross-over so low that it is a woofer to a midtweeter. So you compromise.

The art of speaker design is choosing the set of compromises that produce something that works for you -- all speakers have all sorts of compromises.

Here is an example of a 2-way speaker that works & takes the concept of multiple woofers even further than an MMT.

An externally hosted image should be here but it was not working when we last tested it.


Reiver Acoustic | Multiway | Jenna



dave
 
First, two ordinary drivers act not as a line source. Second, the main reason to design a MMT comes from the need to correct for baffle step behavior. That's acquired by applying a lowpass filter that counteracts the baffle step function of the enclosure. Actually I can't think of any other reason to design a MMT.
 
First, two ordinary drivers act not as a line source
How is that? What is the difference between 2 or say, 8 drivers in this respect?

My reason for this question about MMT 2-way is the use in an OB. Using 2x6.5" drivers could keep the baffle width narrow enough to control the dipole pattern from 100-200Hz up to almost 2kHz, get faster response, higher sensitivity and some bottom end gain. That would all be good things for an OB. With a sub, you can keep it 3-way instead of 4-way solution as Linkwitz LX521 with different kind of driver sizes.

I thought a basic reason for going 2.5 way in a MMT was the lobing issues?
 
I'm no OB-expert, but why would you pick a 2-way filter topology in a MMT-setup? As for line sources, a true line source doesn't have to cope with the line end artifacts (and therefore is more a theoretical model) which interfere with the cylindrical radiation pattern. Practical derivations from line sources (Keele's CBT included) may be beneficial in ordinary listening environments because of a higher ratio direct field to 'reverberant' field compared to point sources, but then again, you'd have to construct a real line, not two drivers stacked.
 
Ok, thanks for the feedback. In your opinon, at what number of drivers does a line source start? Would that number differ according to the size of the driver?
I've experimented with 3.5" FR drivers, and listened to both 16 and 4 in an array, and they act together well in both config.
 
Ok, thanks for the feedback. In your opinon, at what number of drivers does a line source start? Would that number differ according to the size of the driver?
There's a critical distance after which line source behavior stops. For a line array that's (too) short, the critical distance will be small. Then the listening distance is greater than the critical distance. The line source behavior has stopped at the listening position, so you end up with a point source. A point source that beams, with poor directivity vs frequency.

The formula for the critical distance is:
CD = L^2 * f / 700
with CD the critical distance [m]; L the length of the array [m]; f the frequency [Hz].

Soundindex/How far will the Line Array Speaker effect extend? - SounDoh
 
Last edited:
This low XO frequency requirement is what drives the interest in "Econowave" speakers.
That is using a pro type compression driver on a large horn body so you can crossover at 1,000 to 1,200 Hz.
All the C to C issues go away at low XO points.
In an MMT however, it still might make sense to go 2.5 way and XO the bottom woofer at 600 or something to do the baffle step compensation.

Dave
 
Great, thanks a lot for that post. Will look into this.
If I understand this right, if you have enough drivers stacked, the lobing issues between the lower woofer/s and the tweeter will disappear?
Interesting thought. I think it would be if the tweeter had the same power output per unit length (of the array) as the woofers. Then the woofers and tweeter cannot be distinguished acoustically at the crossover frequency. But that's not the case; the tweeter is shorter, so it has to put out more power per unit length.
 
Oh, dipole makes things even more difficult! Most modern dipole speakers are 4-ways (SL LX521, Nao Note RS, my Ainos) because with dipole action each driver has even much more narrow comfort zone.

A dipole speaker is always a bunch of compromises. DSP gives opportunity to use lots of eq in the low end, but this stresses drivers and amplifiers awfully. With 3-way it is impossible to achieve controlled directivity and/or decent distortion at high spl.

Your question about double Ms in dipoles (applies also to multiple woofers) - they are only used to get lower distortion at low end of their passband.

Please study the web pages of SL and JK
www.linkwitzlab.com
www.musicanddesign.com
 
Ok, thanks for the feedback. In your opinon, at what number of drivers does a line source start? Would that number differ according to the size of the driver?
I've experimented with 3.5" FR drivers, and listened to both 16 and 4 in an array, and they act together well in both config.

2 or more drivers in a row is an array. An array of 2 starts to pick up some line array charcteristics. It becomes more directional in the vertical plane. It starts to push out its region of nearfield behavior.

Now, when does it become a good array or line source? That depends on how close to the ideal you require. Besides length, you also need point density (enough elements close together) for reasonable performance. Perhaps when the array length is at least half the listener distance you can start to expect line like performance.

As to why 2.5 way, it generally give a much cleaner vertical response family. A pair of side by side drivers will have interference (off axis lobes) for the frequency range where they are more than a half wavelength apart. Rolling off one of the units can clean up the vertical response quite a bit and as others have pointed out it gives you some extra variables to flatten your LF response with.

You have to be careful of what rate you roll the lower woofer off, since this will create phase shift that gives response problems. This tends to force you to use woofers with the right amount of climb from low to mid frequencies. 6dB of climb from 100 to 2000Hz is a reasonable target.

The dual woofer 2 1/2 way tower speaker is a modern classic configuration with a lot going for it.

David
 
Lot of good input here, thanks;-)
Yes. I've studied both Linkwitz and Nao pages, also dipoleplus, but not all... My ideas was based on this knowledge and I thought maybe there is a way to keep a constant directivity dipole more simple. Mainly because I looked a lot at Mundorfs dipole AMT's. They seem to be among the most narrow dipole tweeters around, and if xo low enough, you just have the mids and a sub left to make it work.. Please have a look at this sketch:
D P project front side.jpg
It's an MMT with a decoupled ripole underneath. According to some simulations in Edge I think the dual woofers could work up to about 2kHz, but with this bigger AMT you could xo at 1.6kHz. Now that's an expensive driver with narrow vertical dispersion, som the the smaller AMT25d (half the lenght) would be better off but needs to be xo at 1.8kHz. The use of dual woofers is exactly based on what you mentioned Juhazi, to get the low end to work better, but at the same time get more sensitivity and quicker response from 2x6.5" instead of using a 8"-10" driver.
Maybe this could work 3.5-way (incl sub)? In a way it's not much different than Perfect8 The Piont.
Please fire your cannons;-))
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.