Dispersion control with foam

[dirkwright]

I found this thread:
Speaker Directivity - The Gear Page
regarding controlling dispersion in larger diameter woofers, like 12" and 15".

The idea comes from a Jay Mitchell, whom I do not know personally.

Potentially infringing material removed by moderation.

[a_tewinkel]

Isn't this simply a way of converting an open U-frame speaker to a rear-ported resistance box, i.e. a cardioid-like speaker?

Could be done with open cell PU foam, however I have my doubts about performance with high volume displacements. Area of the "port" should be large enough to ensure linear behaviour of the foam. Other materials may be better suited. Keep in mind guitar cabinets have different requirements than hi-fi speakers.

[dirkwright]

Originally Posted by a_tewinkel

Isn't this simply a way of converting an open U-frame speaker to a rear-ported resistance box, i.e. a cardioid-like speaker?

Could be done with open cell PU foam, however I have my doubts about performance with high volume displacements. Area of the "port" should be large enough to ensure linear behaviour of the foam. Other materials may be better suited. Keep in mind guitar cabinets have different requirements than hi-fi speakers.

The foam goes on front of the driver, so I'm not sure how it would become a "rear-ported resistance box". I don't know what that is anyway.

The NRC for the foam I ordered is not known, so it would be probably better to specify a particular NRC curve to obtain optimum results. The problem I found is that the NRC rated foams are not flat sheet, or at least I couldn't find any.

It's true that guitar speakers are open baffle types, but the principles of beaming at high frequencies is no different as far as I know. I'm just learning about this technique so bear with me here.

[a_tewinkel]

Okay, get the idea. I misinterpreted it at first and associated it with this: DIY archive of Kimmo Saunisto (see "Cardioid bass").

If what he says is true (about measured results), it may be interesting for hi-fi indeed. The ring probably attenuates and low-passes the sound so the radiating area at high frequencies is reduced. It reminds me of the woofer mountings of newer Emerald Physics speakers and also of a compression driver. Large radiating area, small aperture, in this case realized by the "doughnut" ring.

[dirkwright]

Originally Posted by a_tewinkel

If what he says is true (about measured results), it may be interesting for hi-fi indeed. The ring probably attenuates and low-passes the sound so the radiating area at high frequencies is reduced. It reminds me of the woofer mountings of newer Emerald Physics speakers and also of a compression driver. Large radiating area, small aperture, in this case realized by the "doughnut" ring.

He states that forcing the driver to have a smaller effective radiating area at higher frequencies makes the dispersion angle greater at those higher frequencies. Of course, this is not a light switch, where it's "on" or "off", and I'm sure there could be some very cool refinements to this technique if I had the right measuring gear (and knew how to use it!). For example, we could specify a particular NRC curve, which typically for acoustic foam goes up with frequency, but the shape of the curve changes based on the thickness and other properties of the foam. Also, the ideal NRC curve could be matched to the driver's response. Even further, the 3" hole in the middle is just his application, but what if a different size hole is better? What if the hole size should be designed to match the tweeter or midrange size? Should the thickness of the foam vary over the diameter as well, effectively changing the NRC curve? So many questions...

[dirkwright]

Of course, what I'm talking about is matching the dispersion of the lower frequency driver to that of the higher frequency driver at the crossover frequency. That's the sticky wicket, I think. If this trick works, then it's ok to use a 12" (or 10" or 15") driver in a simple two way loudspeaker.

[dirkwright]

Another cool feature of this modification is that the virtual 3" driver that is formed by the hole will be surrounded by acoustically absorptive foam so there should be very little if any diffraction at the higher frequencies.

[dirkwright]

Someone should try this with the larger full range drivers.

[weltersys]

Originally Posted by dirkwright

Of course, what I'm talking about is matching the dispersion of the lower frequency driver to that of the higher frequency driver at the crossover frequency. That's the sticky wicket, I think. If this trick works, then it's ok to use a 12" (or 10" or 15") driver in a simple two way loudspeaker.

The foam donut is a good idea for a guitar speaker which is used full-range, as it gives wider dispersion in the upper "ice pick" 4K range.

A single front loaded 12 inch cone has an approximate -6 dB 60 degree beamwidth at 2.5K, and -6 dB 30 degree beamwidth at 4K.

Good two way designs using 12" or 15" cross over at a frequency that the HF horn dispersion matches the LF driver dispersion, around 800 to 1200 Hz for a 90 degree 6 dB down point. Some PA speakers cross as high as 2K to protect the HF driver, but large woofers generally have bad breakup modes above 2K, so dispersion that high is a lesser problem than sound quality.

At any rate, the foam donut is a good idea for those that are using large speakers full range or crossed high enough where HF beaming is a concern.

Art Welter

[weltersys]

Originally Posted by dirkwright

Someone should try this with the larger full range drivers.

Larger full range drivers rely on the HF beaming to keep the on axis frequency response somewhat flat. Widening the dispersion would result in a severe on axis drop in HF, the speaker would sound "dull".