I'm somewhat looking for confirmation and an expanded explanation of what I already intuitively know.
Horizontal MTM, as in Center speakers, are not a good design, we simply use them because it is necessary not because they are good.
But specifically what are the problems that a Horizontal MTM encounters that a Vertical MTM does not.
Through the grapevine, I've been lead to believe that one solution to the problems of a Horizontal MTM, is to not run the bass drivers is parallel up to the Tweeter Crossover. That, either configuring them in a 3-way design or into a 2.5-way design improves the inherent flaws in this design.
But if that is true, then I must ask, why don't more manufacturers of Center Speakers use these techniques?
While all aspects are open to discussion and my following enlightenment, I'm most interest in the specific problems that arise in an Horizontal MTM design where both Mid-Bass drivers run in parallel up to the Tweeter Crossover.
Thanks for any help you can give.
Steve/bluewizard
Horizontal MTM, as in Center speakers, are not a good design, we simply use them because it is necessary not because they are good.
But specifically what are the problems that a Horizontal MTM encounters that a Vertical MTM does not.
Through the grapevine, I've been lead to believe that one solution to the problems of a Horizontal MTM, is to not run the bass drivers is parallel up to the Tweeter Crossover. That, either configuring them in a 3-way design or into a 2.5-way design improves the inherent flaws in this design.
But if that is true, then I must ask, why don't more manufacturers of Center Speakers use these techniques?
While all aspects are open to discussion and my following enlightenment, I'm most interest in the specific problems that arise in an Horizontal MTM design where both Mid-Bass drivers run in parallel up to the Tweeter Crossover.
Thanks for any help you can give.
Steve/bluewizard
Somewhere I've seen simulations, where horizontal MTM produce nulls. In reality, in home environment, I don't think that lobbing problem is relevant. And If you need to cover large venue you would not be using typical hifi MTM anyway.
Basically the issue is that an MTM speaker has a fairly narrow lobe in the vertical plain with nulls above and below the tweeter axis. This will vary depending on the spacing of the woofers, the closer together they are the less the effect.
Now this is not a major issue when in the vertical orientation if the speakers are positioned with the tweeter at seated ear height. However when you turn it on its side, you get nulls to the left and right of the tweeter axis which will not be desirable in a HT situation.
One of the things you can do to minimize this is to use a very small face plate tweeter and mount it above the two tweeters rather than between them. Lower crossover frequency also will help.
See attached for roughly what I mean by not having the tweeter between the two M's Note I haven't tried this myself, just going off what I have read 🙂
Tony.
Now this is not a major issue when in the vertical orientation if the speakers are positioned with the tweeter at seated ear height. However when you turn it on its side, you get nulls to the left and right of the tweeter axis which will not be desirable in a HT situation.
One of the things you can do to minimize this is to use a very small face plate tweeter and mount it above the two tweeters rather than between them. Lower crossover frequency also will help.
See attached for roughly what I mean by not having the tweeter between the two M's Note I haven't tried this myself, just going off what I have read 🙂
Tony.
Attachments
When placed like this the orientation causes a drop in the level critical to speech intelligibility due to the lobes nulling this range off axis.
Is why you may see a TM between the woofers or a single 3-4" fullrange between them.
If you want a nasty surprise try auditioning some of your favorite pieces of music strictly through a MTM-layout Center speaker, with the Left and Right speakers silenced.
I listened to a number of Center speakers from some pretty expensive vendors, at prices up to $5,000, and the sound quality was pretty bad from most of them. The Left and Right speakers, by comparison, were more in the hi-fi league, and most were listenable.
My takeaway? Not only is the MTM layout flawed, but most Center speakers aimed at the home-theater market, even those at very high price points, do not have flat response, and don't even match the L & R speakers from the same vendor.
I listened to a number of Center speakers from some pretty expensive vendors, at prices up to $5,000, and the sound quality was pretty bad from most of them. The Left and Right speakers, by comparison, were more in the hi-fi league, and most were listenable.
My takeaway? Not only is the MTM layout flawed, but most Center speakers aimed at the home-theater market, even those at very high price points, do not have flat response, and don't even match the L & R speakers from the same vendor.
Thanks for the responses.
WINTERMUTE -
"One of the things you can do to minimize this is to use a very small face plate tweeter and mount it above the two tweeters rather than between them."
I think you meant to say - '...mount it above the two WOOFERS...' or Mid-Bass.
So, you say a low crossover helps, but what do you consider low?
The odd part, which LYNN OLSON eluded to is, if the standard Horizontal MTM is so bad, they way do virtually every maker of Center speakers use it?
This started in another discussion about the KEF Q800DS which appears to be a Di-Pole Surround speakers. People were concerned that it doesn't come in matching pairs. Rather all the speaker are the same -
KEF Q-800DS AV-LAND KEF Q800DS Dipole Speakers
It uses two Uni-Q compound drivers, one somewhat facing forward, and one somewhat facing backwards. Since they are not in symmetrical pairs, meaning there is no specifically Left only/Right only speaker, some people were concerned that there would be phase problems. One could look at it as, they are all RIGHT speakers.
As it turns out, it is not a 2-way Di-Pole, but rather at 3-way. The tweeters run in parallel, and one assumes Di-Pole phasing. However, one of the bass drivers covers 300hz and below, the other cover at least 300hz to 2.5khz.
But it being a 3-way, does not eliminate the Phase problem. Since there are no left/right speakers. If we assume they are on the Side, then one low-bass faces forward, and on the other side of the room, one faces rearward. Can that work without being a noticeable difference in the sound?
Or if we think of the speaker in the rear of the room. On one speaker the low bass is going to be pointed at the center of the wall, and the other low bass is going to be pointed at the side wall. That is, both low bass pointing to the left, or both low bass pointing to the right.
But, discussing this raised the whole question of Horizontal MTM. In this case, we have true Di-Pole on the tweeter (I assume) and Low-Bass/Mid-Bass on the larger drivers. I can see how a low-bass/mid-bass probably has an advantage, but it would seem the phase would certainly be in error relative to the placement of the speakers. The only solution would be to place the speakers on one side of the room Upside Down to make sure the low-bass was facing the same direction on both speakers.
But, this only highlights the flaws and complications of the design in general. Rather than the very very very common 2-way MTM, wouldn't it be better to make Center Speakers 3-way? With out a real overlap in frequencies, wouldn't that eliminate the lobing? Wouldn't the spacing of the Low-Bass/Mid-Bass then become irrelevant. Assuming an adequate Mid-Range wouldn't that actually clear up the quality of voice?
Even here at DIY, we still see countless 2-way MTM, or on occasion, a 3-way horizontal WMTW. But the twin low bass drivers are the problem in either case.
So, with this problem universally known, why don't people come up with a good working alternative? There must be some better possibilities. Yet, the standard 2-way horizontal MTM still dominates the industry and DIY?
My initial question which has been answered (more or less) was what are the problems. My second question is, if these problems are universally known, why doesn't somebody come up with something better?
I suspect a few have tried alternatives. Yet a design we know doesn't work still dominates both commercial and DIY designs.
It just seem curious.
Steve/bluewizard
WINTERMUTE -
"One of the things you can do to minimize this is to use a very small face plate tweeter and mount it above the two tweeters rather than between them."
I think you meant to say - '...mount it above the two WOOFERS...' or Mid-Bass.
So, you say a low crossover helps, but what do you consider low?
The odd part, which LYNN OLSON eluded to is, if the standard Horizontal MTM is so bad, they way do virtually every maker of Center speakers use it?
This started in another discussion about the KEF Q800DS which appears to be a Di-Pole Surround speakers. People were concerned that it doesn't come in matching pairs. Rather all the speaker are the same -
KEF Q-800DS AV-LAND KEF Q800DS Dipole Speakers
It uses two Uni-Q compound drivers, one somewhat facing forward, and one somewhat facing backwards. Since they are not in symmetrical pairs, meaning there is no specifically Left only/Right only speaker, some people were concerned that there would be phase problems. One could look at it as, they are all RIGHT speakers.
As it turns out, it is not a 2-way Di-Pole, but rather at 3-way. The tweeters run in parallel, and one assumes Di-Pole phasing. However, one of the bass drivers covers 300hz and below, the other cover at least 300hz to 2.5khz.
But it being a 3-way, does not eliminate the Phase problem. Since there are no left/right speakers. If we assume they are on the Side, then one low-bass faces forward, and on the other side of the room, one faces rearward. Can that work without being a noticeable difference in the sound?
Or if we think of the speaker in the rear of the room. On one speaker the low bass is going to be pointed at the center of the wall, and the other low bass is going to be pointed at the side wall. That is, both low bass pointing to the left, or both low bass pointing to the right.
But, discussing this raised the whole question of Horizontal MTM. In this case, we have true Di-Pole on the tweeter (I assume) and Low-Bass/Mid-Bass on the larger drivers. I can see how a low-bass/mid-bass probably has an advantage, but it would seem the phase would certainly be in error relative to the placement of the speakers. The only solution would be to place the speakers on one side of the room Upside Down to make sure the low-bass was facing the same direction on both speakers.
But, this only highlights the flaws and complications of the design in general. Rather than the very very very common 2-way MTM, wouldn't it be better to make Center Speakers 3-way? With out a real overlap in frequencies, wouldn't that eliminate the lobing? Wouldn't the spacing of the Low-Bass/Mid-Bass then become irrelevant. Assuming an adequate Mid-Range wouldn't that actually clear up the quality of voice?
Even here at DIY, we still see countless 2-way MTM, or on occasion, a 3-way horizontal WMTW. But the twin low bass drivers are the problem in either case.
So, with this problem universally known, why don't people come up with a good working alternative? There must be some better possibilities. Yet, the standard 2-way horizontal MTM still dominates the industry and DIY?
My initial question which has been answered (more or less) was what are the problems. My second question is, if these problems are universally known, why doesn't somebody come up with something better?
I suspect a few have tried alternatives. Yet a design we know doesn't work still dominates both commercial and DIY designs.
It just seem curious.
Steve/bluewizard
Last edited:
hmmm yes I did mean above the two mid bass' 🙄 Low depends on the spacing. but figures I see mentioned are lower than I would ever try to cross any conventional tweeter. That is under 2000Hz.
Zaph has done a horizontal MTM which he says has good response up to 50deg off axis. The crossover has been optimised to do this. It may be worth having a look at what he has done Zaph|Audio - ZA5 Speaker Designs with ZA14W08 woofer and Vifa DQ25SC16-04 tweeter crossover of that model is around 2000Hz.
edit: as to why they do it (ie horizontal MTM) I think you will find it is more marketing driven than anything. Having something that matches the L/R speakers is good aesthetically. It also lets the marketers say that as the centre speaker is the same as the L/R front speakers you will have a seamless tonal transition (even though the horizontal MTM may actually guarantee this isn't the case).
I personally don't see the need for a center channel speaker. I only run 2 channel stereo and the imaging is good enough that I never feel that I need a centre channel 🙂
On the question though in the link I posted above, Zaph says the following:
So that may be the answer to your question 🙂
Tony.
Zaph has done a horizontal MTM which he says has good response up to 50deg off axis. The crossover has been optimised to do this. It may be worth having a look at what he has done Zaph|Audio - ZA5 Speaker Designs with ZA14W08 woofer and Vifa DQ25SC16-04 tweeter crossover of that model is around 2000Hz.
edit: as to why they do it (ie horizontal MTM) I think you will find it is more marketing driven than anything. Having something that matches the L/R speakers is good aesthetically. It also lets the marketers say that as the centre speaker is the same as the L/R front speakers you will have a seamless tonal transition (even though the horizontal MTM may actually guarantee this isn't the case).
I personally don't see the need for a center channel speaker. I only run 2 channel stereo and the imaging is good enough that I never feel that I need a centre channel 🙂
On the question though in the link I posted above, Zaph says the following:
So that may be the answer to your question 🙂
Tony.
Hi all,
Did a search on my own contributions regarding this MTM subject that's closely related to arraying drivers.
Maybe there are something in these threads that can shed some light...
http://www.diyaudio.com/forums/multi-way/100135-diy-centre-speaker-advice-3.html#post1189081
http://www.diyaudio.com/forums/planars-exotics/102084-line-array-question.html
b🙂
Did a search on my own contributions regarding this MTM subject that's closely related to arraying drivers.
Maybe there are something in these threads that can shed some light...
http://www.diyaudio.com/forums/multi-way/100135-diy-centre-speaker-advice-3.html#post1189081
http://www.diyaudio.com/forums/planars-exotics/102084-line-array-question.html
b🙂
Thank you for all the responses. I'm busy reading the links provided.
Off on a tangent, if I were to use two bookshelf side-by-side, or two small floorstanding side-by-side, I would get the bass drivers very close. However, I then create a gap between the tweeters.
Using my Tower speakers (2x 8") as an example, I could probably get the bass drivers as close as 8.5" to 9", but that would place the tweeters at the same distance apart. Though in practice, I probably wouldn't need to use 8" woofers. The frequency associated with 8.5" wavelength would be 1590hz. In these particular speakers, the crossovers are at 150hz, 1khz, 6khz.
I'm guessing if I used boxes with 6.5" drivers and 7" spacing, that still only pushes the wavelength spacing up to about 1930hz.
It would seem the solution is a strong mid-range driver with a modest bass plus tweeter in a 3-way would solve a lot of problems.
I'm running out to time now, but I will read with interest the link on Line Arrays. Curious what characteristics makes a long line of drivers appear as a point source. I suspect one of the characteristics is that they are vertical rather than horizontal.
Thanks again.
Steve/bluewizard
Off on a tangent, if I were to use two bookshelf side-by-side, or two small floorstanding side-by-side, I would get the bass drivers very close. However, I then create a gap between the tweeters.
Using my Tower speakers (2x 8") as an example, I could probably get the bass drivers as close as 8.5" to 9", but that would place the tweeters at the same distance apart. Though in practice, I probably wouldn't need to use 8" woofers. The frequency associated with 8.5" wavelength would be 1590hz. In these particular speakers, the crossovers are at 150hz, 1khz, 6khz.
I'm guessing if I used boxes with 6.5" drivers and 7" spacing, that still only pushes the wavelength spacing up to about 1930hz.
It would seem the solution is a strong mid-range driver with a modest bass plus tweeter in a 3-way would solve a lot of problems.
I'm running out to time now, but I will read with interest the link on Line Arrays. Curious what characteristics makes a long line of drivers appear as a point source. I suspect one of the characteristics is that they are vertical rather than horizontal.
Thanks again.
Steve/bluewizard
I think MTM designs work pretty well. But obviously they are designed to work in a vertical alignment primarily, and the home cinema horizontal arrangement is a bit borked. 😀
An MTM swaps vertical lobing for combing. Choose your poison! 😎
I have these lying about, and I wouldn't mind having a second to do stereo.
These are the better series wired polycone bass units, which makes for a better impedance and less resonance issues in a shared enclosure. FWIW, the crossover is a simple bass coil, and a CL network on the metal tweeter. I could do better of course.
It is one of those nice features of the series wired MTM that the conversion from a good two way design is almost trivial. So find any good two way design, double the box size and the job is almost done. 🙂
An MTM swaps vertical lobing for combing. Choose your poison! 😎
I have these lying about, and I wouldn't mind having a second to do stereo.
An externally hosted image should be here but it was not working when we last tested it.
These are the better series wired polycone bass units, which makes for a better impedance and less resonance issues in a shared enclosure. FWIW, the crossover is a simple bass coil, and a CL network on the metal tweeter. I could do better of course.
It is one of those nice features of the series wired MTM that the conversion from a good two way design is almost trivial. So find any good two way design, double the box size and the job is almost done. 🙂
- Status
- Not open for further replies.