To d’appolito or not to d’appolito: the case for an MTM 2.5 way

[MikePM]

I have been following various discussions and have been unable to fathom the benefit of the d’appolito configuration. Phil Bamberg has a discussion of his 2.5 way MTM design that makes far more sense to me. (http://www.bamberglab.com/s25lcr_xo.shtml)

Generally speaking a TM images better, presents a wider soundstage and has better dispersion characteristics than the MTM done in a d’appolito configuration. The only benefit I can see to the MTM is its greater dynamics due in large part to the woofer surface area. If the MTM is implemented as a 2.5 way, the second cone reinforces the first as the lower frequencies start running off and improves dynamics, and by not using the tandem configuration in the mid frequency range you retain the better imaging and sound-staging. This also gives the added benefit of being able to place the speaker on its side as a center channel without the dispersion problems of the typical d’appolito configuration.

From what I’ve seen from the generally fairly low x-over points between the mid-woofer and the tweeter in the MTM designs shown on this site, going to an MTM 2.5 would not lower the SPL attainable by the speaker and would largely be more versatile.

Am I missing something or is there some benefit to the d’appolito configuration that I don't get, and is anyone aware of a good 2.5 way MTM design using drivers similar to the Modula and not as expensive as the Seas excel drivers in the BESL design. Thanks for you help and comments.

[JonMarsh]

There's a lot of representations you've made about differences between MT's and MTM's which I think you'd have to do a carefully controlled comparison of properly designed versions of each to substantiate.

Otherwise, they're just your opinions about, perhaps, the one's you've heard.

Conventional MT's do usually have vertrical dispersion problems, so I'm not convinced that laying one on it's side in an TMM but built like an MTM would be any panacea- from the designs I've worked on, it probably wouldn't be symmetrical in dispersion, though with a low enough crossover frequency, I think you could get it to work fairly well.

BESL offers MTM designs, with good components and careful implementation. His series 5 is an MTM, not a TMM.

There are many with designs published on the web by DIY folks that feel more as you do, regarding MTM's versus TMM, and I'd be surprised if you couldn't find some that were working on designs using similar drivers with that format.

OTOH, having done the TMM thing (first time about 22 years ago) as well as MTM's, I'll say that with conventional crossovers and crossover frequencies it's easier to get smooth power response with a TMM and not be fighting as many lobing issues in the crossover region. However, that does come at the cost of a more complex crossover- more parts overall, as there are two independent low pass networks, and one highpass.

You appear to have strong opinions about one and the other, so why don't you share some of the designs you've developed which highlight these differences, so we can learn from your experience, and perhaps compare them with other designs that may not show up such a stark difference between the two approaches?

As John Steward Mills would might paraphrase were he around these days, the more we investigate and discuss together, the closer we'll all be to some objective notion or knowledge of reality.

~Jon

[MikePM]

Jon: Thanks for your reply. My present occupation as an attorney probably caused my question/opinion to come on stronger than it was intended. (in my prior life I was a geophysicist hence some passing knowledge of acoustic wave theory and signal processing - albeit at much lower frequencies)

What my post was really driving at was what benefits are there to the d'appolito configuration that using a 2.5 way crossover wouldn't have? From you response, I assume the crossover to roll-off the low frequency only woofer is the major addition to x-over complexity. Is this not compensated for by decreasing the complexity of the other cross-over, and does the configuration "sound" better? or just different?

I have heard Phil's series 5 and series 2 designs, Danny Richies designs at GR-Research the new Snell series 7 MT and MTM designs as well as other commercial speakers. The only MTM configuration I have heard that created an exceptionally wide sound state was was the Audes Blue which is actually an MTM with a side woofer (3-way).

With respect to Phil's designs - his newest is the 2.5 series which is a 2.5 way in an MTM (not TMM) configuration. This design struck me as having merit and Phil himself described them to me as having the dynamics of his MTM d'appolito with the better soundstage of his MT. He also felt they worked better in as a center channel than his other designs.

I started following the DIY when I began rebuilding a pair of Allison Acousitc CD-8 speakers that I bought new 20 years ago. The Allison convex dome tweeter is great, IMHO, but is not supported in the DIY field and I have been unable to get frequency responce curves inorder to update the x-overs.

[Evil Twin]

What my post was really driving at was what benefits are there to the d'appolito configuration that using a 2.5 way crossover wouldn't have? From you response, I assume the crossover to roll-off the low frequency only woofer is the major addition to x-over complexity. Is this not compensated for by decreasing the complexity of the other cross-over, and does the configuration "sound" better? or just different?

I have heard Phil's series 5 and series 2 designs, Danny Richies designs at GR-Research the new Snell series 7 MT and MTM designs as well as other commercial speakers. The only MTM configuration I have heard that created an exceptionally wide sound state was was the Audes Blue which is actually an MTM with a side woofer (3-way).

Something to keep in mind is that MTM or TMM is just a concept for the driver/crossover arrangements, and without doing a controlled experiment (an optimized design for each with identical drivers) you'll be hard pressed to isolate the performance varibles inherent to a specific set of desgin choices from the merits of the configuration.

With regards to the crossover, our experience suggests that you will NOT have a compensation in reduced complexity to the crossover in the main crossover point by going to a TMM from an MTM- most of the reasons for crossover points have to do with the behavior of the drivers- i.e., energy storage and dispersion, and rising midband distortion- these are factors to consider carefully in the overall design.

An optimally executed MT will be just as complex as an MTM for a given set of drivers, if the overall performance goals with regards to crossover frequency and slope are required. Compare the M8ta and original M8 crossover designs against the M8a MTM or the Modula MTM. Now, there are simplified designs working more towards the cost/complexity end of the performance tradeoffs with the Modula MT and the Natalie P, but that was a deliberate choice, especially for the MT, to develop a crossover with substantially reduced cost, albeit at the tradeoff of more driver stress at higher playback levels and somewhat less ideal off axis response.

MOST MTM systems I've seen are not well designed, or rather, IMO, they have unacceptable trade-offs- typically much too high a crossover frequency (2.5 to 3 kHz) for both the driver size and spacing, as well as the energy storage issues of the cone.

The RS180 is pretty clean up to 1400 Hz- above that there's significant energy storage, so in the Modula MTM it's run with a 1400 Hz crossover.

In the 8" two way systems, the crossover is usually ~1250 Hz, due to energy storage issues as well as dispersion (I like good off axis repsonse through the crossover region up to 45 degrees+).

Most of the designs you're talking about are what we call "proprietary" DIY- the only DIY part is for someone to assemble a kit which is sold by the author, as the design isn't published.

Here at HT Guide, most of our discussion is focused around what you might call "Open Source" DIY, i.e., freely published and discussed designs, which you're free to use or not use, upgrade or use as is, as you please, no charge.

Build a TMM design if that's your preference- we're just offering some information based on experience and analysis, i.e., pointing out where the stumps and roots are in the darker parts of the woods where you might stub your toes while wandering around through the design process.

If you like what you've heard at BESL or GR research, and if you're not setup for measurement and design yourself, then buying one of their kits might be the safest way to get where you want to go. The contributors at HT Guide (myself included) publish a lot of detailed info on these designs (more than many of the commerical DIY designs), but we have limited time to support this hobby, and have our own notions of what directions we want to use for our next project.

Because the Modula's were designed with limited budgets, they didn't use W18's and Millenium Excels, as that would have been out of the budget range for the interested parties. The results are surprisingly close, though- the value in the RS drivers is very real.

[dsrviola]

I've wondered about this off and on. Just about everything I'm finding on the subject refers to 2 way MTM vs 2.5 way TMM. What I'm curious about are 3 ways using TMM instead of MTM, like the http://www.vonschweikert.com/vrline/VR-6M.HTM (and it's larger sibling) or the http://www.stereophile.com/content/b...speaker-page-2. The B&W had complimentary midrange crossovers. I think the VR6 and VR8 had parallel mids.

Comments? Experiences?