Article
Whitepaper linked in article.

Interesting !

Also (for us cheapskates) it doesn't look like something overly difficult to reproduce / DIY for yourself.

No not expensive, they are just using plastic as a cover sheet. They used a 3D printer, but with enough personal DIY labor, could easily be made. I wonder if circles would change the performance? A router and casper jig would be much easier to cut various sized holes.

this kind of diffusor work on the high frequencies. Area where there is no issue.

I think the idea is that you can effect lower frequencies without having to build an incredibly deep diffusor.

So if - in the below example - you would need a 2m deep diffusor (!!) to effect something as low as 85Hz, with this approach you could achieve the same with a diffusor of just 20cm.

Using this approach to build incredibly thin diffusors that only effect mid/high frequencies might be useful in some situations, but I think the real advantage here is the ability to effect lower frequencies without the diffusors taking up half your room.

"For example, a typical man's voice can be as low as 85 hertz, with a wavelength of 4 m or 13.1 ft. If that's the lowest sound the Schroeder diffuser will have to deal with, the diffuser would need to be roughly 2 m -- or just over 6.5 ft -- thick.

But the new, ultra-thin diffuser design requires a thickness that is only 5 percent of the sound's wavelength. So, instead of being 2 m thick, it would only be 20 cm -- or less than 8 in. -- thick."

I wonder if you could use a stiff cardboard (not cheap corrugated packaging stuff), maybe polyboard or something like that - if so you could easily cut out the holes with a craft knife/scalpel.

this kind of statement is only binding on those who believe it

I guess you didn't read the white paper

The largest aperture is close to lambda/4 or width/lambda=0.25, with lambda being the design frequency's wavelength; therefore, w=lambda/4. This aperture equates to a a phase angle of 2*pi*6/7. Quite a large hole compared to the cavity underneath. The other, smaller apertures, are then going to be evenly ratioed of 1/N [N=7 in this case], down to 0, which is a completely covered cavity (i.e. no hole). [The seven (7) comes from the number of cells in the design, the stepping of aperture width (w) corresponds to N, 7 in this case].

Therefore, pick as low of wavelength (lambda) as you want. Depth is your only constraint, but the depth is now 10% that of a standard implementation. The paper also shows good agreement of performance between the two designs, at the tested frequency.

Also I would think a round hole is okay. The history of Helmholtz Resonators is clay pots in churches, or people have used Coke bottles mounted on a wall.

The covering should be relatively dense and rigid (per the white paper ). I do not think or know if cardboard/tag board would be heavy and rigid enough.

i read. It is a paper and only a paper. A paper is not the real life. The real life is in a room with measurements.

Someone needs to translate the maths into actual values for us more math challenged folk !

I was thinking the same, that the circular hole would just need to have the same area as the square hole it's replacing ?

You're probably right.

Excellent point, all the numerous people in this thread saying this paper is a giant predatory reptile living on the surface of a distant planet need to take note.

Nor is it fantasy, caught in a landslide, no escape from reality.

Agreed, you aren't really living unless you've got a tape measure in one hand and an SPL meter in the other.

Bump on an old thread.

I am a home recording hobbyist, and looking to do some DIY diffusion. Due to space constraints, I am very interested in these thin "MSD" panels. However I'm surprised that, after 10 years, there is not more information online about these MSD style panels, other than this white paper. And I haven't found anyone who has actually built one.

I'm willing to give it a shot at building my own. However, I am also not a professional audio engineer / acoustician, and while I understand some of the general concepts, I don't follow the formulas and math very well. If someone were to just give me a design that's already done I know I could build it. So my challenge is mostly in figuring out a design that will work for me. I'm pretty confident I can do the rest from there.

That being said, I'm having a hard time figuring out to calculate the design dimensions based on info in the white paper. The well depth is easy (wavelength / 20). They don't offer much explanation as to how they calculated the well width (denoted 'D' in their diagram) and the neck width (denoted 'w' in their diagram).

Based on Figure 1, it seems that they are implying that D is calculated using the same method on a typical Schroder diffuser. Is anyone else getting the same interpretation on that?

For the neck width (w), as far as I can see they don't really explain how they calculated those. I understand that they use several different neck widths, but I don't see anywhere where they explain what widths they used and how they calculated it. I only see some statements at the end of page 6 about the neck width being "the only tunable parameter" and comparing/contrasting the neck width to a traditional helmholtz resonator.

Maybe there's someone else out there that's smarter than me who can help me figure out how to calculate these variables? I'd be happy to post some pics and before & after sound samples after I get them built & mounted in my studio.

What I have heard is that these babies work, but they have seirously limited bandwidth. So they work but they are more or less useless because of limited bandwidth...

Sorry, I never got a notification on your reply, so I'm just now seeing it.

I read and re-read the research paper several times trying to understand it so that I could design & build my own. What I eventually noticed is that they are basically modifying the schroeder design so that the well depth sits parallel to the wall instead of perpendicular to it. So while they are decreasing the depth of the diffuser, they are increasing the width of it. Each well (or chamber) on the MSD has to be wavelength / 2. So if I wanted to treat something down to 100hz, the chamber would have to be 5.625ft wide. And that's just for 1 chamber. The design they built in their experiment was a 7x7 square (49 chambers). To build a 7x7 panel that is effective down to 100hz it would have to be more than 39ft square, which is absurd for my small 19ft x 18ft studio.

I'm not an acoustician, just a guy trying to pursue a hobby, so take this next bit with a grain of salt. With my limited knowledge of acoustics, I do understand that some diffusers are more effective over short distances than others. After I figured out what size these needed to be, I moved on, but I'd be curious to also understand how effective the diffussion is in varying proximities to the panel. If you have to be 20ft away from it before the diffusion is effective, then that also doesn't help me. That might be explained in the research paper, but I gave up on this option before I got to that level of understanding on it.

So these may be effective, but like anything else it depends on the application. For me specifically, these don't seem to be the right application, so I'm now looking into other options. When you say "serious limited bandwidth" I'm not sure if that's because they only work in a certain range, or if it is just impractical to build them to treat lower frequencies due to the required width. I wonder if it was a large concert hall / auditorium / stadium if these would be more appropriate and effective in that application. Of course in a large area like that, the need for diffusion is probably not as great since the larger waves can naturally dissipate more easily. So who knows. Maybe someone will find a good application for them in 20 years from now.

Seriously under recognized are polydiffusers. Use those.

Thanks for the tip, I'll check them out!

You are relatively new here so welcome! Please master my tagline. I have been told that the 90% is too low!

And of course, enjoy!

I believe it! For my day job I actually work as a construction project manager on large commercial construction projects. I know very well that a good build cannot make up for a poor design.

I've been reading Gearspace threads for a while, but only in the past 6 months decided to start actively participating. There are tons of knowledgeable people here, and I've learned a ton from you all just being a fly on the wall.

You are welcome! Obviously you know more than most people here. You know that in acoustics you know very little.

As a construction manager you know math which makes things so much easier to communicate.

Where are you? Nothing incredibly specific but at least a general geographic area. You are posting to the world.

And making it clear, enjoy!

I am in Phoenix, Arizona, USA.

I definitely know enough to know when I'm out of my depth! I studied audio engineering for about a year at a local community college here in AZ so I'm at least partially trained, but I'm mostly self-taught in audio production. After the first year in the program I decided to go down a different road professionally. In many ways I'm happy with my choice. I make decent money, I'm able to have a family and a home.
My job provides enough that I have some extra so I can afford (within reason) to buy better gear than I could back then. But on the other hand I have this sense of "what if I had stuck with music? What could I have accomplished?" It's kind of like the girl that got away for me. So ~15 years later I'm starting to get back into it, but not with the intent to make it a career.

Anyway, we're getting off topic for this thread, and since I'm sure nobody cares about my life story, I'll just leave it at that. Thanks again.

Wel, I enjoyed your life story.

Going more bluntly. put you location in your user profile.

Great history!

Saw a measurement of one of these, they worked at 1 octave. As it seems a diffusor well works by its volume not it's depth, so many ways to build one. It is also unclear from my memory if you can hear diffusion below 2-3-400Hz...