AI Driven Design Ideas

[PGansz]

I have been playing with ideas for a four way design using all SB textreme drivers from top to bottom: 5in 8ohm textreme mid, 8ohm aluminum tweeter, 9 1/2in 4ohm textreme woofer, flanked by twin 9 1/2in 8ohm textreme woofers with a big front slotted port.

I was using AI for visual guides on materials and design ideas. What do you think? Has anyone tried this?

 

[diyhobby]

I have been playing with ideas for a four way design using all SB textreme drivers from top to bottom: 5in 8ohm textreme mid, 8ohm aluminum tweeter, 9 1/2in 4ohm textreme woofer, flanked by twin 9 1/2in 8ohm textreme woofers with a big front slotted port.

I was using AI for visual guides on materials and design ideas. What do you think? Has anyone tried this?

Wish AI can give the DSP crossover instead of me using REW. Should ask AI to give a Dutch & dutch 8C cardio type setup using SB drivers and miniDSP

 

[NotInYourHead]

I'd worry less about the look and more with the DSP/crossover necessary to tame such a configuration. What are your design goals?

 

[PGansz]

Yeah there's no question this will be a complex situation and decent investment in time and dollars.

It all seems pretty straight forward on paper: cross the tweeter to the mid at 3k, cross mid to front woofer at 220-280, cross front woofer to four side firing woofers at about 60hz. The sensitivity of each configuration brings them all exactly on level and should offer constant directivity.

Originally I thought it would be three way and use the four 9 1/2's as a sub, but with these many drivers (increase in sensitivity) and surface area, I effectively should not need any subs.

The last real constraints other than designing the crossover network itself is actual box volume, and port configuration. I'm not too concerned especially with modern tools available.

The design is really important to me. I am going for something really unique and tasteful.

 

[Heinrich]

Yeah there's no question this will be a complex situation and decent investment in time and dollars.

It all seems pretty straight forward on paper: cross the tweeter to the mid at 3k, ...

.. stop. Too high for halfway decent integration. Better aim at 1.2kHz.

 

[PGansz]

.. stop. Too high for halfway decent integration. Better aim at 1.2kHz.

I appreciate your concern, care to explain?

Thr tweeter can be crossed safely between 1khz and close to 3khz. The 5in mid plays very well there...

SB26ADC-C000-4 / Aluminum - Sbacoustics

Phase optimized diaphragm design for coherent high frequency radiation, copper cap for reduced voice coil inductance and minimum phase shift.

sbacoustics.com

I am relatively new to this and have only played with one set of diy speakers. I intend for this project to take quite some time to get right so I am in no hurry.

 

[Heinrich]

I appreciate your concern, care to explain?

Sure, you may find some comments on the so called 'lobing' issue with two+way speakers. What you decidedly want, is an as wide as possible central lobe, best directing slightly upwards. With digi x/o that is done easily. Don't worry, the looks won't be affected, if you plan for a small center to center distance of the drivers.

Thr tweeter can be crossed safely between 1khz and close to 3khz. The 5in mid plays very well there...

As said, it's not each driver individually, but how they (the waves actually) interfere. Coaxes solve the problem, or a standard looking new designs that got a ton of praise especially for mitigating the problem a lot. "Outstanding, indeed!" they said ...
But, 1kHz is a little bit low, especially as no waveguide is to be used. At higher volumes the tweeter might sound stressed. Who knows can't overhear that.

So, as nearly always, get yourself a mike for measurements and teach yourself what to ask the data for ;-) Then you'll see, looks is a nice (sometimes) add on, not the starting point for things, the ol' here guy say.

Have fun!