Handling a rising response

[rick57]

I'm intending to do a speaker using PHL drivers, due to all the raves, and I'm attracted to the high efficiency. They don't publish FR graphs ("mostly sell to pro companies, who get test samples"), but I've heard that many of them have a rising response, that's "easily handled in the crossover".

I'd have to get some and test them to see by how much they rise, but I want to be confident that it's reasonably easily dealt with.

I was going to use them in an open baffle, driven actively.

Suggestions?

[rick57]

Open baffles cause a rise 6 db/ octave. Perhaps that's the phenomenon I've heard reference to.

How to calculate the parts/ values deal with it?

TIA

[SY]

Stick 'em in the open baffle, measure 'em, and then design the crossover to incorporate whatever EQ you think is appropriate. EQ will make ANYTHING flat on axis; the tough job is to control off axis response.

[rick57]

ok, so any suggestions to control off axis response?

[SY]

That's a little like asking, "How do I get to the grocery store from my house?" when I don't know where you live or where the grocery store is!

Decide what you want the off-axis response to look like (there are several schools of thought). Get the on axis response of your drivers. Get the responses at 10, 20, 30, 40, and 50 degrees off axis. Then you can make an intelligent choice about how and where to crossover. If you post your measurements, there won't be any shortage of opinions.

[rick57]

Sure it's a sweeping question. (I'm a newbie) - I thought that, leaving aside dipole vs bipole vs monopole, there were only basically only two schools of thought - make it as linear as possible for multiple listeners or positions, or assume a single sweet spot, and don't worry about off-axis response.

What other views are you thinking of, or are there threads or sites that you suggest?

Thanks again.

[SY]

No, it's (unfortunately) a LOT more complicated than that. For example, should the off-axis response be uniform (i.e. close to omnidirectional?) Or should it be uniform in the midrange and slowly roll off in the high end? Should the crossover response be all-pass or constant power? Or both? And then there's the vertical response...

The best way to start is Dickason's "Loudspeaker Design Cookbook." I've recommended it so many times, he oughta start sending me commission checks. There are also a wide range of speaker cad programs which will let you simulate the effect of the crossover on polar response.

For my own philosophy (and let me be clear that this isn't Received Divine Wisdom!), I think it's ok to let highs roll off at increasing horizontal angles, but you need to look carefully at the directivity of the drivers so that you pick a crossover point and slopes where there are no "horns" or "holes" off-axis. These can cause a lot of problems with tonal balance and imaging, at least in real-world semireverberant rooms.

[rick57]

Thanks, Some interesting food for thought.
(Only started getting into diy 6 months ago).
I'll read again my new copy of Dickason's.