Question for Dr. Geddes about his WG foam plug

I've read on your website a brief description and purpose of the open cell foam that you use in your wave guides. I'm looking for a little more information.

Are the foam cells random in size/distribution or are they ordered in such a way as to scatter (or perhaps absorb?) sound at certain angles to combat unwanted reflections and NOT scatter at other, more favorable, angles?

Is this discussed in any of your publications? I'm on the verge of buying your waveguide book. If this is discussed there, just might push me over the edge :)
 
There is no published information on the foam plugs and my book pre-dates its use.

Basically the pores in the foam are a random distribution of sizes arround a certain porosity usually defined as PPI or Pores Per Inch. I use 30 PPI, but it ranges from about 10 to about 100. The pores are indeed random as far as the wavefront is concerned so there is no "spectral" diffraction effect like you are talking about. There is some viscous loss in the material, but it is really pretty low, especially at LFs. If you hold a piece of it up to your ear it is very difficult to hear any effect. It is also difficult to measure its effect, but there are some results posted on my web site in one of the ALMA papers.

Its an interesting phenomina that something that has such a small measureable effect can have such a pronounced audible effect. This has interested me for years and I have some ideas, some of which we have published (also on my web site) and some which I have not talked about in public yet. But I am convinced that this fundamental lack of obvious correlation between the objective effects and the subjective ones is going to be a key issue for loudspeaker design in the future.
 
closed account
Joined 2007
There is no published information on the foam plugs and my book pre-dates its use.

Basically the pores in the foam are a random distribution of sizes arround a certain porosity usually defined as PPI or Pores Per Inch. I use 30 PPI, but it ranges from about 10 to about 100. The pores are indeed random as far as the wavefront is concerned so there is no "spectral" diffraction effect like you are talking about...

Since you have a patent one needs to license the concept from you for commercial use, but for DIY I guess we can have fun with the idea. Is this 10-100 PPI reticulated open cell PU foam? Like the one used in fish tanks, air filters, and also for some "spongy" looking speaker covers?
 
I've read on your website a brief description and purpose of the open cell foam that you use in your wave guides. I'm looking for a little more information.

Are the foam cells random in size/distribution or are they ordered in such a way as to scatter (or perhaps absorb?) sound at certain angles to combat unwanted reflections and NOT scatter at other, more favorable, angles?

Is this discussed in any of your publications? I'm on the verge of buying your waveguide book. If this is discussed there, just might push me over the edge :)

In layman's terms, here's what the foam plug does:

Horn_loudspeaker_animation.gif


In a horn or waveguide, there's a fraction of the sound that only goes through the device ONCE. Ideally, that fraction would be 100%. IE, all of the sound that's radiated by the speaker would radiate perfectly, with no diffraction or reflection.

But this isn't a perfect world, and there's a fraction of the radiation that's reflected or diffracted.

That radiation - the higher order modes - they must go through the foam plug a MINIMUM of three times:

1) first they radiate from the speaker

2) a fraction is reflected back, mostly from the mouth and throat

3) and that reflected energy ("HOMS") goes through the foam plug a third time.

I did a study on this a million years ago. In my experience, the foam plug makes a bigger difference on horns that aren't ideal. IE, the closer that your horn or waveguide is to perfection, the less you need the foam plug.

The HOMster! (or How I Learned How to Fix a Horn)

One way that you can visualize why HOMs are so obnoxious, is to make a phone call on a VOIP provider like Skype or Zoom. You'll notice that when sound from other people is delayed, repeated, or both, it's really obnoxious. It's jarring, it makes you uncomfortable, it makes the phone call unintelligible.

I flew out to audition the Gedlee Summas in 2005, and Geddes had the speakers playing all day long. We were hanging out and talking shop for hours. At the end of the day, I realized I'd basically lost my voice. What had happened, is that the Summas have such power handling, and such an absence of HOMs, that when they're playing at 100dB they sound like they're playing at 90dB. Their SPL levels are deceptive, because they have headroom to burn and very low HOMs.

It's a really neat trick, and very few speakers do it. I bought them as soon as I could afford them.
 
Last edited:
Since you have a patent one needs to license the concept from you for commercial use, but for DIY I guess we can have fun with the idea. Is this 10-100 PPI reticulated open cell PU foam? Like the one used in fish tanks, air filters, and also for some "spongy" looking speaker covers?

I let that patent expire, so it's public domain.

Yes, it has to be the porous stuff - people don't seem to understand that sometimes. With a PPI of 30 you can blow into it as if it's not even there, that's how porous it is. And yet, IMO, it does wonders to the sound. Don't understand why it never caught on.

I do agree with John. The best waveguides need it less, but in all cases that I have heard, it's an improvement. EQ needs to be done post application however, because there is a few dB loss at the high end - that can barely afford it. However, at home, with a high power CD, there is more than a few dB to spare for enhanced performance.
 
Still some interest

I'm glad to see there's still interest in Dr. Geddes' idea. I've experimented with a "plug" made from a stack of circles of increasing diameter (or decreasing, I guess, depends on your point of view!) of 1" foam on and off for years. Recently gave it another try and I think I hear an improvement. Of course, it might be the usual suspect, expectation bias.

As tweaks go this is a very economical one. and any horn owner should try it., Many people claim an improvement. Depending on which foam you buy, you probably don't need to spend more than $50-100 for it, at least for the stack of circles method, which was the best method I could find for myself. Production plugs were made from a single larger piece but that was beyond my price/ability to create.

Also it has the benefit of being zero impact to the waveguide, only requiring some type of adhesive or other method to hold the foam in place, maybe. As someone else probably said, there's probably some validity to the idea, since Geddes thought enough to pay to obtain a patent for it.

I won't quibble too much the techical terms, but I think the reflection at the mouth is a resonance (all tuned pipes have them, and a horn is one). The mouth resonance is easier to visualize, and will always be at a predictable time constant (throat to mouth = length of the "pipe"). The foam plug attenuates that, although the "traditional" solution is a towel ring or other way to soften the "sharp" or abrupt termination of the horn. .

I'll leave the official explanation of the HOM to Geddes. I think of it as a "ricochet" at a discontinuity BEFORE the mouth, that leads to part of the sound going off in another direction. I view these as much more complex, thus unpredictable. I'm sure higher math can model them but even thinking about it makes my brain hurt :)

Both the above are distortions that involve a time and/or spatial delay and (except the mouth resonance) probably very difficult to solve with electronic fix-its (e.g. Gunness). Geddes' foam plug is an elegant solution.

Dr. Geddes, thank you for all you've done for this hobby. Did you ever make any money off your patent? While it may not have been your intent on letting it lapse, thank you for "giving" it to us and, potentially to the commercial market.
 
Last edited:
Just a quick note that I've got a pair of 18Sound XT120 horns, B&C DE250s, and a measurement mic. Foam's on its way, and I'll be interested to see if the effects we're talking about can be measured.

My suspicion is there'll be a smoothing of the response into the VHF/ultrasonic range (fewer reflections = less destructive interference), but I'll try to put my expectations to one side and follow the science.

Chris
 
Foam's on its way

Where did you order the foam from?

I did a bunch of tests with different foams but never tried this type. All my tests showed a reduction in high frequencies, but no subjective improvement over reducing those same frequencies via EQ.

I think the improvement in some cases might simply be attenuating a bright speaker that could have been done equally as well with EQ.

Maybe this particular type of foam is the secret.
 
I think the improvement in some cases might simply be attenuating a bright speaker that could have been done equally as well with EQ.

Maybe this particular type of foam is the secret.

The frequency response, before and after, should be identical. That's why it requires EQ.

On a good waveguide, I've personally found that the effect is pretty darn subtle. I personally don't think it would be discernible in an A/B test, IF the original waveguide is very good.

The main thing that I noticed with my Summas, was that you can listen to them for a ridiculously long time, at high levels, without listener fatigue.

Another thing I noticed, is that once you get used to that sound, bad waveguides become really obvious. For instance, I can walk into the prosound demo room at Guitar Center and I can pick out which waveguides sound crummy, really fast. To me, it's especially obvious on pink noise. After listening and measuring to a few good waveguides and a lot of bad ones, I can generally get a sense of whether it's a good waveguide in a matter of seconds.

You can see this phenomenon in my latest project, (The Nightmare Before Labor Day) where I basically knew it was time to go back to the drawing board after I played pink noise over the waveguide for a grand total of about five minutes. I could just tell that something was "off." (I'm referring to the waveguide from two weeks ago, no the new one. The new one seems to work really nice.)

foam can be purchased here: McMaster-Carr

Note that the Gedlee speakers used one big solid block, which is a lot more expensive than building the "plug" up from layers.

If you REALLY want to see the foam work it's magic, put it in a waveguide or horn that's not ideal. When you do that, the difference is really obvious. In my "homster" thread, I was kinda shocked how it was able to take a horn that sounded like fingernails on a chalkboard, and make it actually sound quite tolerable. Still nowhere nearly as nice as a well designed waveguide, but much better.
 
Last edited:
Th effect is subtle and yes one has to EQ the response after installation.

I have measured the effect, but it is small. On the FR it is hard to even see, but on the impulse response it can be shown that later reflections in the waveguide are reduced. The ripples in the FR are reduced but only a small amount. But owing to the delayed aspect of reflections and HOMs the subjective effect is much greater than one would suspect from the FR curve.

Foam is "the icing on the cake" for a "good" waveguide, but worth the effort.

I have some of the foam and will sell it, but only in large bricks at $100 a brick plus shipping. The bricks are enough for a half dozen plugs or more.
 
A wish...

I hope the plug will see more use, since it's so easy to do. I propose verbing Earl's last name, in his honor. It's the least we could do, his idea being now public domain.

"Dude, your horn tops sound like two cats in a blender! Why don't you Geddes them?" :cool:

The idea occurs to me that one could easily "A/B" the mod if as is usual, you have two identical waveguides. Try one with foam, EQ it to be the same frequency response as without, and then compare side by side (in mono, I guess). While not a perfect A/B, it'd be an interesting comparison.
 
Last edited:
closed account
Joined 2007
I was wondering how much is the attenuation. I would assume that it increases with the frequency. I also vaguely recall reading something like "up to 3db over 1000Hz" by either Patrick Bateman or Earl Geddes (this may be all fruit of my imagination). In this situation I may be able to change by crossover by modifying two resistors and keep the response as before... experiences?

EDIT: Yes I indeed found this


  • I have treated oblate spheroidal waveguides with foam, and they experience a modest reduction in SPL. About 3db above 10khz, decaying to about 1db at 1khz. This horn, with a diffraction slot, is seeing a much greater reduction in SPL, with a foam plug which is smaller. My theory is that there are more HOMs in this diffraction horn than in an OS waveguide, and the foam is attenuating them.

and this

The most loss that I have seen from foam is about 3 dB and its usually more like 2 dB.

Since I am using the stock waveguide of the Beyma TP-150H, my results should be very similar to these. Keeping my fingers crossed.
 
Last edited: