MTM sound characteristics

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
celef, it's funny that you described the sound of MTM "enormous headphones". I was thinking about it while eating late breakfast. I grew up with headphones, and even after I got really good speakers, I kept using headphones along with them until MTM was introduced to me. I found MTM does everything that headphones do, and better, and I gradually stopped using headphones since then...
 
Not sure what you are using, but assuming you digital xo in active system...do you happen to use Acourate to generate filters? I ask because Acourate includes Horbach-Keele type filters that seem like a perfect match for MTM. I think their approach tends to add more active channels (4 or 5-way), but they achieve very consistent vertical dispersion. The size and location of the drivers, together with their xo are the key.

Here's a link to their papers. I posted on a thread of mine. I'm also exploring MTMs, and H-K is one approach I'd like to try. Attached to post #12.
https://www.diyaudio.com/forums/multi-way/355704-lobing-mtm-tpl-150h-8pe21-2.html#post6240525

Hope it helps. Interesting read to me.

Great! Acourate is very interesting. I always wanted to try Keele's MTM. Thank you for the information. :)
 
d'Appolito initially went with 3rd order Butterworth; generally speaking even order is now favoured including by Joe for most 'real world' conditions.

No single answer to the question I'm afraid, since it depends on the drivers -size, position / spacing, on baffle on + off axis responses & also what you're willing to put up with. ;) Some aren't as bothered by a degree of polar shift as others. For MTMs, I won't use more than 1 wavelength C2C separation between the midbass drivers at the XO corner frequency, and when possible lower. You mention 8in drivers & a 'large tweeter', so assuming regular cone & dome units, that presumably means you've got at least 12in C2C spacing. In that case, I wouldn't want a corner frequency > 1.13KHz, LR4. Lower would be better from the POV of the polars, but increasingly impractical with regular tweeters. Steeper would also improve this side of things, but puts more demands on the tweeter as the transition band is narrower (i.e. its forced to stay flatter for longer). So ultimately it's a question of balance & where your priorities lie -power-handling & distortion performance or a more even polar response. YMMV as ever.

Thank you!
Large tweeter: Beyma TPL-150H. So midranges are 450mm (17.7") c-t-c which is the closes I can get them without cutting into the horn. That's a wavelength of about 750Hz which is too low for the TPL. I'll take the even order LR 4th order or higher for experimentation.

What's you view regarding time alignment vs phase in MTM? Playing with Xdir I see I could get very nice polars with the above separations, xo at 1600Hz, by introducing phase lag of 90 degrees at xo point and reducing woofers by 6dB vs tweeter. Polars look good, but time alignment would be lost.
 
Just another Moderator
Joined 2003
Paid Member
Hmmm I'm not following why an MTM is not good for multiple seat positions. we watch movies with mine and the three main seats in front of the TV are all good. yes if you want the absolute best stereo imaging it is best dead centre, (but I don't think that is something that is limited to MTM's) but it's still good a seat position either side... Move to two seat positions either side (with my setup) and you will be sitting almost directly in front of one of the speakers, so there is no chance IMO no matter what the speakers are, that you will get proper stereo imaging with that much of an offset from one speaker to the other.

Have I missed something? My understanding has always been that a vertical arrangement MTM has very good horizontal dispersion, and that any "issues" are with the vertical polar response. I guess if you have people of widely differing torso height and a very narrow lobe then they would not be great in any position for some listeners.

Tony.
 
Well, the different phase angles is what causes said, so it amounts to more or less the same thing. The issue becomes worse with higher crossover frequencies for a given spacing of the midbass drivers, which is why, if the speaker is to be used by more than one person, they really need to be crossed as low as possible, as that reduces the issue. If they're only likely to be listened to by one person, and you can set them up accordingly to a given listening position, you've more freedom in this regard.

I was curious about Amphion's crossover frequency and I found this video. They cross a bit lower than usual for different reasons, but this decision also made it possible to make a good MTM for sure.

YouTube
 
Thank you!
Large tweeter: Beyma TPL-150H. So midranges are 450mm (17.7") c-t-c which is the closes I can get them without cutting into the horn. That's a wavelength of about 750Hz which is too low for the TPL. I'll take the even order LR 4th order or higher for experimentation.

What's you view regarding time alignment vs phase in MTM? Playing with Xdir I see I could get very nice polars with the above separations, xo at 1600Hz, by introducing phase lag of 90 degrees at xo point and reducing woofers by 6dB vs tweeter. Polars look good, but time alignment would be lost.

forgot to add the attachment!
 

Attachments

  • MTM xdir 225mm phase and level.jpg
    MTM xdir 225mm phase and level.jpg
    134 KB · Views: 407
Have I missed something?

Nothing AFAIK Tony. Setting the other critiques aside, it's just that the limited vertical dispersion that is either its raison d'etre or Achilles heel (depending on your POV) tends to restrict matters more than some other configurations, when multiple heights and sometimes listening distances may become involved. This may or may not be a problem depending on circumstance.
 
diyAudio Moderator
Joined 2008
Paid Member
Showing the 90 degree phase style. Based on the images in GMs link I mirrored one image to show the effect.

a) Linkwitz-Riley, full output in the forward direction, this is symmetrical so mirroring it achieves nothing in particular.

b) A single MT with 90 degree phase difference.

c) An MTM based on 'b', more even dispersion. Done this way, the peaks and nulls line up.
 

Attachments

  • lobing.png
    lobing.png
    120.4 KB · Views: 966
I've dug a little on a 6db crossover time align / phase aligned mtm.
Specifically dunlavy sm1 (6.5" 2-way).
However, I read someone on audioasylum recommended 10' listening distance, like most of his speakers.

I do know a tight dispersion speaker can get annoying, you get mad at it when you stand up.
But that is (was) an issue when i was sitting 7' away.
 
Perhaps the comb filtering is smoothed the farther away you are, between the ears and reflections, similar to how stereo comb filtering isn't much of an issue and how we psychoacoustically deal with it. I don't think it's synonymous with "phasy sound" which to me implies movement and/or time.
 
Well, a 'comb filter' effect (i.e. cancellation nulls in the frequency / amplitude response) is the result of the differing phase angles between radiation sources at different physical distances from a given point in space, so you can't actually have one without the other. Phase certainly does cover other matters, but since in your post you said you presumed celef / plasnu were referring to combing and not 'phasy sound' as somebody moves, I was simply reminding you that cancellation nulls don't exist without differences in phase.

Re crossover order / vertical polars, if fallible memory recalls, Joe shifted to LR4 with the idea that the vertical lobe reduced issues with floor / ceiling reflections compared to the B3 he originally favoured. Which it probably can in many cases, but as usual what works out in practice will ultimately depend on the specific system & individual requirements.
 
Last edited:
diyAudio Moderator
Joined 2008
Paid Member
I see. Some say that aiming a null at the ceiling isn't bad, or is better than an alternative. Many factors involved.

I don't like ceiling reflections, including when its tweeter contribution is unsupported by the crossover band. Controlling tweeter directivity right down to the cross is a challenge, but ceiling absorption is a viable alternative.

Another factor is that the ceiling reflection is critical in a small spot, and can be fixed by a relatively small absorber, yet the power distribution and balance is still widely affected.
 
One thing i really like about a MTM, apart from the extra vertical control,
is how imaging can be better than a TM.

When listening in the vertical sweet-spot, where it is equidistant to the M-M drivers, an apparent acoustic image can form in the center from the M-M (or M-M and W-W, if a WMTMW)

It's like how/when two identical speakers run in mono create a very solid center image in the horizontal plane (when listening equidistant)

The MTM adds an analogous vertical plane centering i think..

I've played with this idea where the T had it's own box, along with each M having it's own box, and I could stack them as wished.
After tuning was applied to TM, MTM, and TMM configs, the thing that stood out the most was the tighter MTM mono image. This was both indoors and out, so pattern control issues were diminished. fwiw
 
In the prosound world, "Phasy sound" is often used to describe the swishy sound that's heard from line arrays when the wind blows, and all the multiple HF units arrive to ear at slightly different timings. Same thing walking around indoors, especially with pink noise.

A great test for any speaker i think, it to play pink noise and move around....head, body, however large an area you want good coverage. Can be very revealing in terms of lobing, combing, etc.
 
I am a bit wary of stepping into MTM speaker theory these days. Very controversial. Very heated. Can't even agree about series wiring or parallel wiring. Never mind front or rear ports. :eek:

But AllenB just nailed it about BW3 filters.

858394d1594004005-mtm-sound-characteristics-lobing-png


An MTM has symmetrical vertical lobing even with third-order slopes. Which incidentally have flat-power response at crossover. Which is a good thing, IMO.

635014d1505220144-home-theater-speakers-vifa-pl14wj-dappolito-mtm-jpg


Better, IMO is the MTTM idea here by Roy Allison:

628054d1501378412-classic-monitor-designs-allison-ic20-speaker-jpg


Or using some sort of ribbon:

532765d1456057805-classic-monitor-designs-mtm-scanspeak-raal-ribbon-selah-audio-jpg


This is by my friend Rick Craig of Selah Audio in Canada. I have heard this sort of thing and it works brilliantly. Falls off inverse distance rather than inverse square, so doesn't shred your ears close up. The folks at the back hear it almost as well as the folks at the front.

Cylindrical dispersion which gives that lovely PA sound and projection.

Everything we know about good drivers and crossovers come into play. But MTTM is just a brilliant idea. Everything works better. I've tried it. But even MTM is an improvement.

519648d1450294626-noob-woodworking-experience-start-mordaunt_short_ms821c-jpg


Lower distortion. Because each driver works at 1/4 power. Less interference with poor room acoustics. But more expensive to do. How it is. :)
 
Last edited:
Thanks Mark, I'll try that, I'll only be able to compare one driver to two drivers though since my speakers are single full range drivers. I can easily move them around though and it could be interesting to hear if single speaker is better than two when listening to mono recordings.
 
I see. Some say that aiming a null at the ceiling isn't bad, or is better than an alternative. Many factors involved.

Agreed. For clarity, I wasn't expressing an opinion by the way, just stating Joe's reason for changing his preference to LR4. Digging out his remarks from the Thor:

'I favor in-phase, i.e., even-order, crossovers for most applications because
they are the least sensitive to inter-driver phase differences and timing errors. In the case of the MTM configuration they also limit off-axis response in the vertical which greatly reduces floor and ceiling reflections.'


I don't have any fixed view on the subject myself beyond 'do whatever works in a given situation / for a given set of requirements'. As you say, many factors involved.
 
Thanks Mark, I'll try that, I'll only be able to compare one driver to two drivers though since my speakers are single full range drivers. I can easily move them around though and it could be interesting to hear if single speaker is better than two when listening to mono recordings.

Cool Scott,
I'm very interested how a full range driver sounds under pink,....on - vs off -axis.....pls do fill in...


and in my strong opinion........
SINGLE SPEAKER MONO IS THE ONLY WAY TO EVEN HAVE A FAIR CHANCE AT EVALUATING SPEAKERS...
(i said it quietly lol)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.