Waveguide profile?

[Deward Hastings]

Maybe it's "hiding in plain view" somewhere, but I'm coming up short finding useful information on chosing or optimizing a waveguide profile (contour) that could be applied to a dome tweeter mod/conversion. I've established (for a possible project) a desired depth (AC offset), a maximim diameter (baffle width), and a throat size (determined by the tweeter), all "practical" rather than "design optimization" considerations, but can't seem to get from that to radiation angle, lf augmentation, or "best curve" . . .

Any suggestions or pointers?

[JonMarsh]

you could go back to all the discussions, even back in 2007 about waveguide profiles for compression drivers for oblate spheriod waveguides such as the DDS ENG-1. Dennis H participated in those a lot, put up some Excel graphs and profiles.

I've worked with the DDS and others, enough that I'm disenchanted for now with even the widest band compression drivers available under $200, so I went back to playing with domes and the good ole MCM waveguide to see how far quick pragmatic effort could be taken... i.e., somewhat low effort (compared with projects like the Ardent, which due to changes in work load for my day job just aren't possible this year).

Isiris

BMS Tweeters / DDS Waveguides

New Waveguide Study

I'm sure there's lots more, but this will give you a good starting point for reading and further searches.

[Dennis H]

Conical (straight-walled) horns are good for constant directivity. Earl's oblate spheroid is a variation on the conical theme with a specific throat profile. The thing is the OS theoretical advantage only works if the source puts out a plane wave and that won't happen with a dome. So, you just do the best you can at the throat to get the dome to have a smooth transition to the cone. Out at the mouth, you want as big a round-over as you can handle to help eliminate the horn honk (diffraction that reflects back down the horn.)

And finally, stuffing the horn with fish foam seems to help tame a lot of nasty reflections that may be going on inside. Patrick Bateman over at DIY was experimenting with horns/waveguides for his car and suggested that the foam might be more important than the throat shape. Earl uncharacteristically admitted that that was probably true although it's something he usually keeps quiet about.

[Dennis H]

I found the Robert White (RCW on the boards) article I was looking for. On the 2nd and 3rd pages, he does the math and builds waveguides for a dome tweeter.

http://sound.westhost.com/articles/waveguides1.htm

[Paul Ebert]

...I've worked with the DDS and others, enough that I'm disenchanted for now with even the widest band compression drivers available under $200, so I went back to playing with domes and the good ole MCM waveguide to see how far quick pragmatic effort could be taken... i.e., somewhat low effort (compared with projects like the Ardent, which due to changes in work load for my day job just aren't possible this year).

Am I understanding this correctly? That is, that you will not be able to complete the Ardent this year?

[JonMarsh]

I've seen that before, too- wasn't really all that thrilled with his results.

Wish I could get the POM waveguide Jantzen/Troells used, but barring that, still keep coming back to the MCM, especially considering the results we're getting with that. His process for building a waveguide makes mine for milling and fitting look like a walk in the park in comparison! :W

[JonMarsh]

Am I understanding this correctly? That is, that you will not be able to complete the Ardent this year?

No no no, it means I couldn't do another project like the Ardent from scratch this year- it's far too much work on a cabinet level! The Ardent's should be wrapped up by the end of March if not sooner. I'll be back to working on them this weekend, even.

First some test measurements for the Modula Xtreme, though.

The heck of it is that I have an idea for a modified Ardent- simplified, actually- two ideas, even, and could use some of the parts I cut for a second set last year. Probably won't consider starting on that until about June. It depends on how these other things go.

[Deward Hastings]

Well clearly it wasn't a "new" question (oldguy got there first ), but still not much in the way of *answers*. Basically I read "conical section, with the throat and terminus rounded to taste". And not much in the way of polar measurements to guide that taste (like "how much does a 5-7 inch "waveguide" influence the 60-90 degree polar response"). That's assuming, of course, that the intent of the waveguide is to match midwoofer beaming.

Could similar pattern control be obtained simply by surrouding the tweeter with a donut of rigid fiberglass or rockwool insulation to absorb the side radiation?
Without any danger of adding "horn resonances"?

[JonMarsh]

Some have "felt" that's a better/easier path, such as the Joe Rasmussen Elsinore project, which uses a stepped felt rectangular absorbing system, which reportedly/allegedly even improves the off axis behavior of the venerable XT-25 (the original tweeter in that project- however, it doesn't improve the distortion behavior of that tweeter, unlike some waveguide loaded experiments done by AJ and others)

Certainly it could be something to investigate, depending on what your personal goals are- it might be adaptable to a wider range of tweeters than the true waveguide path is, which seems highly dependent on managing the immediate tweeter/waveguide interface for good performance in the top octave. (to date, I've only seen two documented solid examples of this, the MCM waveguide with the 27TDFC, as done by John Krutke, and the MCM waveguide with the D2608/9130, as done by moi- I've tried several other tweeters with much poorer results, have one or two more left to investigate. I think AJ successfully used it with the XT-25, but I haven't seen measurements). DBX did something similar in commercial speakers back in the 80's, I think.

[Paul W]

Image not avilable

That's probably what Lipinski is trying to accomplish in this speaker. Doubt rock wool or rigid fiberglass would be dense enough...maybe wool felt or heavy foam. There is a patent floating around for "soft" horns, felt or a similar material.

For conventional conical axi-symmetric WGs, the included angle sets the pattern down to around 1 wavelength in outer circumference. A large roundover (~1/4 WL radius) beyond the 1WL circumference not only reduces diffraction, but tends to extend LF control. You can cheat sizes a bit but, eventually, physics wins.

[Deward Hastings]

Image not avilable

That's probably what Lipinski is trying to accomplish in this speaker. Doubt rock wool or rigid fiberglass would be dense enough...

That's indeed what I was contemplating (as an alternative to a waveguide), although perhaps with a conical rather than stepped front surface. I've seen rockwool at 6 and 8 lb/cubic foot . . .

For conventional conical axi-symmetric WGs, the included angle sets the pattern down to around 1 wavelength in outer circumference. A large roundover (~1/4 WL radius) beyond the 1WL circumference not only reduces diffraction, but tends to extend LF control. You can cheat sizes a bit but, eventually, physics wins

That sets an unfortunate lower limit, since my contemplated baffle width(s) restricts the waveguide diameter to no larger than the midwoofer, but the desired crossover is below 1500Hz.

[Paul W]

Yup, unfortunate, but a very common limitation. That's where the cheating starts. When pressed, the constant directivity crowd seems to favor cheating the transition to the baffle more than the conical section.

[Paul Spencer]

Why use a dome at all? Compression drivers are designed for horns or waveguides as they have a suitable phase plug.