My first attempt at passive crossover design. Comments welcomed (VituixCAD)

[Mark HT]

This is my very first attempt at a passive crossover. Drivers are Bliesma T25B and M74B with data taken from Bliesma datasheets.
I have these drivers on order and intend to cross them actively to a pair of woofers.

Any comments? I have a few months until the drivers arrive and just want to get some practise in before I can take measurements and
start work designing the final crossovers. XO is 2.8k but I also tried going up as far as 4k. I plan to experiment with active XO before
finalising the design to see what audible effect there is by changing the XO point.

I have attached the VituixCAD project and would be interested to see how some of the seasoned pros would attack it.

Any advice is much appreciated.



T25B-M74B_2800.zip

[tktran]

Looks like a solid attempt. Good work.
Some questions - I take it you’ve simulated it on the baffle you intend to use?
Vituixcad also has a Diffraction simulator and you could simulate off axis responses
And/or you could use bliesma’s supplied diagrams.
You could also incorporate the 30 and 60 degrees off axis; which is important.

[Mark HT]

Thanks for the reply. I haven't simulated on baffle but it will be mounted in wall so minimal baffle effects.

I will definitely incorporate off axis but at this stage I was just after a little reassurance I'm on the right track.

FYI Here is a pic of the response with the mid (un)inverted:

[tktran]

Yes that’s good; you can tell that you’ll get a deep null when you invert the polarity because you can see that the drivers have very close phase matching at the crossover point (top right window of the Vituixcad sixpack)

One of the strengths of Vituixcad is to measure in a full 360 degrees, and view all the responses in all manner of ways including vertically and horizontally, out to +/- 180 degrees, as well as CTA-2034 standard. When you optimize the passive crossover for the on axis response, you get to see the effects on all other angles.

DSP allows you to take shortcuts but sometimes those shortcuts mean not enough time was spent optimizing the acoustic design. Optimize the baffle shape/dimensions and cabinet dimensions, placement of the drivers and thus acoustic solution before you optimize the crossover.

Invest time in taking real measurements in dual channel mode. Measuring itself is an art and important to get correct before you move on. Simulation is not a replacement for measuring. The old computer science saying- Garbage in garbage out.

Many 4 inch, and some 5 inch drivers can make it from 500Hz to 2.8 K, so there should be a really good reason why you need to use a 3” mid-dome.

For instance, if you look at some excellent designs like Jim Holtz/Curt Campbell's Statement II, they use a very affordable 4.5" full range cone driver (Vifa NE123W) in it's optimal pass band, which is even beyond 3Khz. This is fantastic engineering because you're not wasting resources to solve something using a simpler more elegant method.