Max's speaker and a few words...

DaveC said:

It'll work and I like the AT drivers, but 1500 Hz xo is right in the middle of a very sensitive frequency range. I'd aim for an octave lower and bring the tweeter in much higher. But that's just me...

Click to expand...

DaveC said:

Also you should simulate dispersion patterns and try to match at xo, with the goal of a nice, smooth polar plot.

Click to expand...

DaveC said:

That's also going to be a very wide cabinet, the cabinet shape is important and that might be wider than ideal, but again, some like this.

Click to expand...

DaveC said:

You could also make the woofer section and xo separate and this would allow you to have multiple mids/highs of different types.

Click to expand...

DaveC said:

And, needs more port area. 2 ports that size would be better.

Click to expand...

DaveC said:

My advice with driver selection is to avoid xo in the most sensitive ranges of our hearing. This will allow you to bypass many potential issues and will make the speaker much more likely to sound good. If you transition in a sensitive region and the timbral match of the drivers is poor you won't be able to ever get excellent results. This is one of the most common issues with speakers in general. It can be technically perfect yet the drivers don't sound the same.

DSP is not going to change the directivity of the driver. An even polar response is the most important objective measurement of a speaker's performance and should be top priority.

A wide baffle will have lower baffle step frequency and will effect diffraction... this should be taken into account.

Larger port area, among other things, reduces port air velocity and increases contribution by the port. Ideally port area = Sd of driver. Port resonance should normally be just under Fs of the woofer, but port tuning also determines at which point the woofer will start rolling off. Also, if the woofer tries to reproduce signal with lower frequency vs the tuning frequency the box will no longer support the woofer cone and you will get excessive excursion. One huge advantage of DSP on a woofer is the ability to tune the cabinet lower, compensate for the earlier roll-off and implement a rumble (high pass) filter to eliminate signal below tuning frequency.

All these things can be fairly easily simulated on a computer these days, which I would highly recommend. Good luck!

Click to expand...

DaveC said:

Start out with something easy like BassBox Pro and go from there. Read the entire manual and info, plus Toole's, Dickason's and D'Apollito's books.

Click to expand...

Folsom said:

If the tweeter has a waveguide it can crossover at maybe as low as 700hz. The issue with crossing lower than generally around 1700hz is that the tweeters run out of excursion prematurely. But the waveguide will give them a boost in the lower range so it won't be any different.

Here's an idea... Use the 10 A 77 25 10 as the midrange and make a waveguide to cross the tweeter real low. The 10in has a natural roll off at 1khz. The MMS is also much closer to the you 15 want to use. It also is more stable closer to the tweeter (phase looks like it could be more consistent). I'd look at the BEYMA TPL-150H tweeter to do that, crossing over lower. (This tweeter is found in speakers at ANY price, its a nice tweeter)

The combination of a waveguide and larger midrange will give it better directivity, with a improved polar plot. And then it would be somewhat unique as well.

Were it me, I'd probably mount the 15 on the rear and the port fire down, to make it sort of like a super monitor. Id cross it over at 200hz. The wavelengths are so long that mounting it on the rear is more an aesthetics thing, as the frequencies will spread literally about the same as if it were mounted on the front. Otherwise I'd probably just make it a floor stander. In that case I'd move the 15 to 1/4 of the way to the bottom so that it isn't pounding against the midrange as hard (moving it to the back would do this too). And when I'm talking about interaction, I do mean on the outside, not inside where you plan to enclose the midrange.

Click to expand...

DaveC said:

Another possibility if you don't care about xo frequency so much is a SEAS coax, you can probably get a xo and box design from SEAS that would eliminate a lot of design work. These are similar to the coax drivers used by TAD, in fact they used to have a Mg cone model that is nearly the same as the TAD Evolution series coax driver if you can find it... This is another way to go that would almost guarantee success.

http://www.seas.no/index.php?option...&catid=52:prestige-coaxial-drivers&Itemid=464

Click to expand...

DaveC said:

It's a good place to start that can also serve as a benchmark for later versions.

For your DSP system, I would use that to low pass to the woofer with the SEAS recommended passive on the mid band pass and tweeter high pass. You can experiment, but with a ~200 Hz xo frequency you'll find time alignment won't matter, and it's impossible if you dsp the woofer and not the rest.

As for the mid and tweeter... alignment, crossover and an optimized polar pattern are all done for you, including tweeter and midrange diaphragms made of the same material, looking at this and other SEAS designs I think they are careful about this.

For the woofer, AT is a good choice, I'd get the largest unit with the largest motor you can afford. They are impressive. In my testing Acoustic Elegance TD15H+ is a standout as well. You have to ask for the "+" version in a 15, but their 18" might work out well for you too. There is no replacement for displacement, as they say for cars, but it's surface area for woofers.

Click to expand...