Help convert crossover 2.5 in 2 way (MTM)

[Sony]

Greetings forum.
I am still at a theoretical phase of transformation, I hope you can help me to know if I'm going in a bad direction since it involves some investment.

I have these stand-mount MTM loudspeakers composed of:

- Horn loaded tweeter (aluminum)
- Two 5" aluminum-ceramic woofers
- 20L enclosure rear vented (rear port tuned at 40Hz)

The current crossover is a 2.5 way one. My idea is to transform it to 2 way one and add 15" woofers in a separate enclosure.
The current XO filters the woofers in cascade and uses impedance correction on both woofers.
I am not happy with the result (midband integration, sound "glued" to each speaker, lacks scale), yet I feel its potential.

My idea was to make the XO purer and simpler: by going 2-way, 1st order, and getting rid of the Zobels. The losses in bass would be resolved by the new 15" woofer in a separated enclosure, thus making an overall 3-way system.

Check the current and the new XO I am considering. What do you think?

[dabbler]

You may run into breakup problems with those drivers and first order crossovers. Do you have the driver measurements?

[sreten]

Originally Posted by Sony

My idea was to make the XO purer and simpler:

Hi,

Your making it purely and simply more wrong, not any better.
Ist order electrical filters can work, but most of the time they don't.

If you want bigger speakers I'd suggest moving on what you have
and buying them (after audition). Do not build and hope it works.

rgds, sreten.

http://sites.google.com/site/undefinition/diy (see FAQs)
Zaph|Audio
FRD Consortium tools guide
RJB Audio Projects
Speaker Design Works
HTGuide Forum - A Guide to HTguide.com Completed Speaker Designs.
DIY Loudspeaker Projects Troels Gravesen
Humble Homemade Hifi
Quarter Wavelength Loudspeaker Design
The Frugal-Horns Site -- High Performance, Low Cost DIY Horn Designs
Linkwitz Lab - Loudspeaker Design
Music and Design

[AllenB]

The concept of first order electrical is appealing, but a first order network is not easy in practice. Drivers/enclosures picked at random for such a crossover are probably not likely to lend themselves to such simple treatment. There are a number of reasons why this could be the case.

That said though, and based on the limit of information supplied, your idea could work. Don't forget to highpass your mids.

[Sony]

Thanks for the answers.

Regarding the 1st order issue, the midrange woofer on present setup (see diagram) is already 1st order with L = 0.2mH (~6KHz), so the idea was to initially try to run the tweeter at 1st order (~6KHz) and see how it goes, and only then go to 2nd order on the tweeter if 1st order 6KHz doesn't work.

The most radical change is to go from 2.5 to 2 way on the top drivers, this way both 5" woofers will handle the midband and bass, while currently one of them handles midband and bass and the other just bass.

By removing impedance compensation I expect to add another woofer (15"), without the need of crossing the top woofers, again simplifying crossovers.
So, to be more accurate, this still is a 2.5 way. Before we had:

- MTW 2.5 way (with M = W = 5" driver, both going to 0Hz)

And now we shall have:

- MTMW 2.5 way (2M and W going to 0Hz)

Allen, the idea was to actually go for a 2.5 way (this is the accurate definition), and not 3 way.

Another note - if I use a 15" woofer it must be side-firing (not an option), but I assume there is no problem as I shall cut it very low (7mH 8ohm or lower).

Do I have 'green light' to proceed?

[AllenB]

Originally Posted by Sony

By removing impedance compensation I expect to add another woofer (15"), without the need of crossing the top woofers, again simplifying crossovers.

You hinted at this in the first post. Are you thinking that running the mids up will then make these speakers deficient in bass, so when you add the 15s, all will be well?

A couple of points, the tweeters will then need to be more loud. If you were to instead disconnect the second woofer so you only had one midwoofer and the tweeter, then the mid/treble would still be correct but the mid/bass would be deficient. If you run both mids up then you may face needing to cross the tweeter higher as well as find another 6dB of sensitivity for it. (all in theory of course).

The other issue is how well the mids perform at their top end. Even good full range drivers have cone issues higher up. There seems no sense in pushing this unless they are well behaved.

With regards to your choice of 15" crossover, it's hard to say whether it would work well. In theory, wouldn't you be trying to cross at the same "2.5 way" point?

First order is an admirable goal but not if the system can't handle it. With a design like this I'd be more interested in a "correct" crossover. If it were worth it I'd take measurements on all drivers and design a fresh crossover. I might also use subs for the room. There seems no reason (from your comments) that these wouldn't have potential down to the middle bass.

[Sony]

Allen, thanks for the tips.
Equalization will be less of a problem as I can play with the tweeter crossing point and crossing order if necessary. This can also handle the problems of having the tweeter working outside its ideal frequency range. Still, my first goal is to cross it higher and 1st order, with no attenuation.

Your observation on the woofers is also pertinent. I expect to keep on 1st order but if I detect peaks/resonance on the 2-6K range, I may have to use 2nd order on the woofers.

The main reason why I wanted to use both woofers (MtM) in the midband is because I feel the speaker currently has poor image. Maybe the problem is low dispersion from the tweeter because it's horn loaded. The horn places tweeter in the same axis as the woofers (that have a deep cone) but reduces dispersion against a dome without a horn. Still, by using both woofers on the midrange not only I increase the midrange radiating area; as I will have a closer to a concentric effect (pure MTM) and finally may eventually be able to cross the tweeter higher at 1st order reducing its low dispersion effect.

I was considering crossing the 15" around 100-200Hz while the 5" driver working as a bass-woofer is currently crossed at 500Hz. The reason for adding the 15" is to get a more tuneful and spacial bass. In large orchestras and other bass impressive tracks, I feel the 5" won't cut the mustard (although reaching 40Hz at -3dB). It's not only the frequency response curve that counts, but also the type of sonic wave a 15" will imprint. I imagine (have not tried) that a side firing 15" 97dB woofer could provide the result I am looking for.
Maybe the subwoofer is the best option but I thought subwoofers were more suitable for movies, and they also introduce more complexity since they have a specific amplifier (different sound from main amplifier as phase and gain problems) that must be integrated with the rest of the system. So, my hint here is that the purest solution would be to use a 15" 97dB 1st order crossed and linked to the same amplifier. The room is rather large and squared in shape.

My philosophy is to aim at the most simple and pure solution possible, therefore only having separate high and low pass filters (no band pass) and 1st order slope, even if this is a MTMW design.

Note that I am not an expert on this, all these thoughts may be wrong, that's why I welcome guidance from those who have experienced these type of solutions.

[AllenB]

Originally Posted by Sony

The main reason why I wanted to use both woofers (MtM) in the midband is because I feel the speaker currently has poor image.

You mentioned the sound 'sticking' to the speaker. Maybe you have diffraction issues, placement issues, or resonances to sort out?

Maybe the problem is low dispersion from the tweeter because it's horn loaded. The horn places tweeter in the same axis as the woofers (that have a deep cone) but reduces dispersion against a dome without a horn.

This 'low' dispersion shouldn't be a bad thing, if anything it may assist in reducing reflections, and present a better balacnced sound in a room. It is important that the mid 'meet' the tweeter with similar dispersion which pretty much mandates a certain crossover point. Unless I were redesigning the crossover I'd go lightly on the current one.

It's not only the frequency response curve that counts, but also the type of sonic wave a 15" will imprint.

A well implemented 15" woofer will not sound like anything at all. If a 10", for example, were capable of the same duty and set up correctly then it too should have no such sound covering the same frequencies. Perhaps you're looking for a particular response curve?

I thought subwoofers were more suitable for movies, and they also introduce more complexity since they have a specific amplifier (different sound from main amplifier as phase and gain problems) that must be integrated with the rest of the system.

Rooms behave differently at these frequencies, they respond to different treatment.

[Sony]

Originally Posted by AllenB

You mentioned the sound 'sticking' to the speaker. Maybe you have diffraction issues, placement issues, or resonances to sort out?

Diffractions: I assume not since a horn is used.
Resonances: zero. These are state of the art built. Ultra thick walls, multi internal brace.
Placements: Got little improvements.

Originally Posted by AllenB

This 'low' dispersion shouldn't be a bad thing, if anything it may assist in reducing reflections, and present a better balacnced sound in a room. It is important that the mid 'meet' the tweeter with similar dispersion which pretty much mandates a certain crossover point. Unless I were redesigning the crossover I'd go lightly on the current one.

This may be a key for the problem, but this means the current crossover will then be the cause of the problem (or a large part of it).

Originally Posted by AllenB

A well implemented 15" woofer will not sound like anything at all. If a 10", for example, were capable of the same duty and set up correctly then it too should have no such sound covering the same frequencies. Perhaps you're looking for a particular response curve?

Response curve does not say it all. I have heard one of those jbl large speakers (k2 s9900) that uses a 15" vented woofer with a poor curve response (48Hz) playing a large orchestra and it's like the real thing. On the opposite, speakers I've heard with -3dB at 20Hz or super-subwoofers that go to 6Hz, do not transmit the same portrait of life music (they shake the ground but it's a mechanical/artificial sound compared with the jbl). Also, my monitors have a much better curve response than the jbl's and don't convince me, so the answer is no - I am not seeking a response curve but the life sound I hear in the acoustic concert. Loudspeakers are musical instruments that must reproduce all the others, from pianos to drums, and maybe small drivers can't just reproduce the big authentic sound we hear in an acoustic concert.

Originally Posted by AllenB

Rooms behave differently at these frequencies, they respond to different treatment.

I totally agree. The problem is that I didn't verified this limitation with two other speakers who had traditional tweeters (Seas Excel and scan speak revelator). So maybe the horn directivity is the problem or/and the dispersion unmatched between drivers...

I may go for a partial solution before adding the 15" box, that is to simply convert the MTW 2.5 way in 2 way MTM with minor adjustments:
- woofers connected in parallel
- Use a single inductor as filter = 1/2 of the value of the inductor of the M driver, since the impedance is now half.
- merge the impedance correction circuits (or drop them).

[AllenB]

Originally Posted by Sony

Placements: Got little improvements.

What axis are you listening on and how are your speakers toed?

I have heard one of those jbl large speakers (k2 s9900) that uses a 15" vented woofer with a poor curve response (48Hz) playing a large orchestra and it's like the real thing.

Ahh. 15" drivers have narrow directivity in the midrange. Adding a horn top end will give good results. The 15's would need to face forward and go high. Your current speakers wouldn't help here. Secondly, a high sensitivity driver like you describe will sound very capable.