• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

paper: "Ultrathin Acoustic Metasurface-Based Schroeder Diffuser"

pozz

Machine
Forum Donor
Editor
Joined
May 21, 2019
Messages
1,485
Likes
2,116
#2
Amazing. The lobing behaviour matches incredibly well. If they take it further it could easily change studio design as we know it.
 

stevenswall

Active Member
Joined
Dec 10, 2019
Messages
181
Likes
93
#6
I wish I knew the scale... Learn blender and see how much it costs to print. Looks like most acoustic treatment companies have their heads in the sand and are stuck with decades old diffusion and absorption methods. You'd think they'd have a half decent employee who would look at recent developments, and a product manager would would get them out of the stone age.
 
Joined
Dec 26, 2019
Messages
3
Likes
2
#7
Interesting stuff! I have found another attempt of slimming down diffusors using Helmholtz resonators in this scientific article: Metadiffusers: Deep-subwavelength sound diffusers. Seems like a problem might be that there also seems to be a lot of absorption in very specific frequencies. Or is that a problem? Also, there seem to be a lot of flat areas on these diffusors, doesn't these flat areas just reflect the sound as if it had been a wall?
 
Joined
Apr 15, 2018
Messages
88
Likes
65
#8
Interesting stuff! I have found another attempt of slimming down diffusors using Helmholtz resonators in this scientific article: Metadiffusers: Deep-subwavelength sound diffusers. Seems like a problem might be that there also seems to be a lot of absorption in very specific frequencies. Or is that a problem? Also, there seem to be a lot of flat areas on these diffusors, doesn't these flat areas just reflect the sound as if it had been a wall?
I guess is the same principle used in the article.
 

pozz

Machine
Forum Donor
Editor
Joined
May 21, 2019
Messages
1,485
Likes
2,116
#9
Interesting stuff! I have found another attempt of slimming down diffusors using Helmholtz resonators in this scientific article: Metadiffusers: Deep-subwavelength sound diffusers. Seems like a problem might be that there also seems to be a lot of absorption in very specific frequencies. Or is that a problem? Also, there seem to be a lot of flat areas on these diffusors, doesn't these flat areas just reflect the sound as if it had been a wall?
I'm pretty sure the use cases they have in mind are commercial and industrial and not to do with studio or residential construction, at least at this point.

These diffusors are very flat but their internal structure scatters sound as if they were much deeper (as in conventional wells in Schroeder diffusors).

Edit: The OP article posted by @headshake mentions studio designs as well.
 
Last edited:
Joined
Dec 26, 2019
Messages
3
Likes
2
#10
Yea, i get how the basics of resonators work. And that the slits (or holes in the first article) changes the phase in a way that scatters the sound. But are the flat areas just an effect of the fact that the resonator need an ”enclosed” space or does the flat areas help in the diffusion, being a part of the PR or QR pattern so to speak? My intuition says that the slits (or holes in tve first article) are so small in comparision with the entire surface area of the diffusor panels. Does what I’m saying make sense? It seems from the graphs, however, that the diffusion is as effective.

So to clarify: I get how the depth can be reduced, but the area of where sound can enter the resonators seems so small compared to the overall size..

Thankful for all input into this intersting topic :)
 

pozz

Machine
Forum Donor
Editor
Joined
May 21, 2019
Messages
1,485
Likes
2,116
#11
If I understand correctly, the MSD they created and tested is frequency-specific. To increase bandwidth the flat areas may be filled in as they figure out how to effectively space the cavities. The B(broadband)MSD shown has wells sized for four center frequencies between 5.7kHz and 11.5kHz. And they say that they can achieve the same performance as a standard SD, so likely more research will show different designs.

The most broadband diffusor developed was the RPG Diffractal, by comparison.

They didn't mention this, although I'm sure they're aware of the lobing effects when you have many of the same diffusor side-by-side along a wall. I'm curious how they'll deal with those.

In terms of the actual physics I think they reason they can be so thin is that they are manipulating the increased pressure/standing waves of the air at the mouth of cavity. So the air is serving as an extension of the device. IIRC, from Cox's book, with a regular SD this effect was known but treated as largely problematic given that it introduced absorptive effects.
 
Last edited:

Old Listener

Senior Member
Joined
Apr 28, 2016
Messages
303
Likes
213
Location
SF Bay Area, California
#12
I wish I knew the scale... Learn blender and see how much it costs to print. Looks like most acoustic treatment companies have their heads in the sand and are stuck with decades old diffusion and absorption methods. You'd think they'd have a half decent employee who would look at recent developments, and a product manager would would get them out of the stone age.
A couple of research reports do not indicate that useful products based on that research can be produced and sold NOW.

I find this sneering attitude toward companies that have put in the time and effort to produce useful products to be very offensive.
 

stevenswall

Active Member
Joined
Dec 10, 2019
Messages
181
Likes
93
#13
A couple of research reports do not indicate that useful products based on that research can be produced and sold NOW.

I find this sneering attitude toward companies that have put in the time and effort to produce useful products to be very offensive.
Yes, products never built, tested, or attempted will never move the market forward.

Useful to make light: A candle. Something useful that is more efficient than every other kind of thing in its product category: an LED bulb. Useful is far too low a standard. I'd respect a company trying to figure out the figurative LED bulb more than a company just making incremental improvements to or selling candles, which is what I think is happening with acoustic panels since zero metasurfaces/materials seem to be on the market.

Could you point to a company that is developing or creating absorption panels that aren't based on decades old designs/Corning 703/Rockwool, etc? Just seems like the typical "wait for a company like Tesla/Apple, then play catch up." I'd love to give my money to a company pushing the boundaries when I move towards room treatment.

I sneer because I've been shopping for a year for speakers, and many of them have issues that $100 computer speakers don't have (JBL LSR308 hiss, Tekton Electra/DI beamyness and lobing, both vs JBL 104,) or the manufacturer talks about a "drawback" to something like a coaxial driver even though there was a whitepaper on a speaker that solved the issue a decade ago (Gelenec 8260), or they company could have bent, elongated, and enlarged a port to get rid of chuffing on a $2,300 set that chuffs way more than an $800 set that goes even louder (Elac Navis vs Kali IN-8.)

To avoid these issues with sound panels, looks like making some will at least be cheap and as good as what's on the market for absorption. Before I bought my house I was shopping for a $10k 3D printer to try making the grid shaped material in 2' x 4' panels, and I've emailed one of the guys who made the ring shaped dampener for air flow through tubes... Maybe at some point I'll go down that path again as some schmuck with a hobby until billion dollar companies have a first mover peer that bankrupts them or makes them change.
 
Joined
Feb 14, 2019
Messages
24
Likes
23
Location
Central NZ
#14
Could you point to a company that is developing or creating absorption panels that aren't based on decades old designs/Corning 703/Rockwool, etc? Just seems like the typical "wait for a company like Tesla/Apple, then play catch up." I'd love to give my money to a company pushing the boundaries when I move towards room treatment.
Not sure if this is what you are looking for, but it seems different to the examples you mention. https://www.revolutionfibres.com/products/phonix/
 

pozz

Machine
Forum Donor
Editor
Joined
May 21, 2019
Messages
1,485
Likes
2,116
#15
Joined
Feb 14, 2019
Messages
24
Likes
23
Location
Central NZ
#16
Interesting. That's a commercial distributor though. No datasheets posted either, that I can see.
Agreed. There's no useful technical information on how this product works and it seems they are only focusing on office furniture as the main application.
 

pozz

Machine
Forum Donor
Editor
Joined
May 21, 2019
Messages
1,485
Likes
2,116
#17
Yun Jing (Last author of this paper) was my Materials 2 professor 5-6 years ago. Not my favorite teacher but very interesting guy. In one class he brought in a sample that came out of his lab. It was a small honeycomb looking sample probably like 5'' x 5'' that was "designed" to be potentially be integrated into the skin of an aircraft. It had the absorption material within the structural honeycomb elements if I remember correctly.

I spoke to him about for a bit but honestly remember very little. I don't think he like me due to my chronic absences. I later tried to get on his research team but "they were full (my chronic absences)".

If anyone cares enough I can email him any questions/ask for additional figures etc.
Could you please ask him for an update on the applications side of the research? A few other questions:
  • Could he provide a description of what happens to the air at the cavity mouths to achieve these effects?
  • How did he and his team think of this approach? Why did they decide to pursue this work?
  • Is <1kHz practical with this technology? What the lower frequency limit would be?
  • Is true broadband diffusion possible? How linear would it be across the spectrum? Are they able to control linearity?
  • Has he considered how to deal with lobing/periodicity effects when placing many of the same diffuser side-by-side?
  • Does he know of any companies working to make this technology available for retail or custom orders (i.e., if it's actually been put to use outside of laboratories)?
  • For absorption only, what is the effectiveness of this technology? Can it be used for linear absorption <500Hz? <100Hz?
Maybe your chronic absence for the past 5-6 years has made him miss you.:)
 

Old Listener

Senior Member
Joined
Apr 28, 2016
Messages
303
Likes
213
Location
SF Bay Area, California
#19
My comments in bold below.

Yes, products never built, tested, or attempted will never move the market forward.

You continue to be glib about the amount of time, effort and money need for developing a new product. Have you participated in any R&D effort? What was the product and what was new about it? How long did the effort take? How much clock time and how many man-hours were required to get to a market ready product? How much did the effort cost?

Useful to make light: A candle. Something useful that is more efficient than every other kind of thing in its product category: an LED bulb. Useful is far too low a standard. I'd respect a company trying to figure out the figurative LED bulb more than a company just making incremental improvements to or selling candles, which is what I think is happening with acoustic panels since zero metasurfaces/materials seem to be on the market.

If I need candles, I appreciate the effort of a company producing candles.

You mentioned candles and then LED bulbs. There is a very long chain of R&D leading from candles top LED bulbs.


Could you point to a company that is developing or creating absorption panels that aren't based on decades old designs/Corning 703/Rockwool, etc? Just seems like the typical "wait for a company like Tesla/Apple, then play catch up." I'd love to give my money to a company pushing the boundaries when I move towards room treatment.

The big advances in many fields in record years have come from the application of computation and communication. Absorption panels may not have changed much but the ability to measure and analyze room acoustics has been a huge step forward.

You seem focused on absorption products and have not mentioned reflection or diffusion. References like Tools and Everest discuss the effect of all three on sound. It makes sense to understand what you need to fix problems for a specific room before choosing products.

I sneer because I've been shopping for a year for speakers, and many of them have issues that $100 computer speakers don't have (JBL LSR308 hiss, Tekton Electra/DI beamyness and lobing, both vs JBL 104,) or the manufacturer talks about a "drawback" to something like a coaxial driver even though there was a whitepaper on a speaker that solved the issue a decade ago (Gelenec 8260), or they company could have bent, elongated, and enlarged a port to get rid of chuffing on a $2,300 set that chuffs way more than an $800 set that goes even louder (Elac Navis vs Kali IN-8.)

It isn't clear what your problem is. You haven't said what your budget is or what your needs are. Can't you find even one speaker that meets your needs at a price that you can afford? If you can find such a prouct, just buy it and be happy with your choice.

A number of people on this forum have very positive things to say about the LBL 3 series monitors. I think that it is great that they offer such good sound at such a low price. If my application was for listening at 18" from the speakers, I'd be concerned about hiss. If I expected to listen from 9', I'd be less concerned.

Coaxial drivers are without their issues. One of the advances common to the JBL 3 series, 7 series and the M2 was the waveguide. In a speaker that uses coaxial mid and tweeter drivers, the waveguide seen by the tweeter is changing shape as the mid driver produces sound.



To avoid these issues with sound panels, looks like making some will at least be cheap and as good as what's on the market for absorption. Before I bought my house I was shopping for a $10k 3D printer to try making the grid shaped material in 2' x 4' panels, and I've emailed one of the guys who made the ring shaped dampener for air flow through tubes... Maybe at some point I'll go down that path again as some schmuck with a hobby until billion dollar companies have a first mover peer that bankrupts them or makes them change.

You might first put some effort into measuring sound quality in your room and then seeing what is actually needed.
 
OP
H
Joined
Apr 20, 2019
Messages
67
Likes
46
Thread Starter #20
I sneer because I've been shopping for a year for speakers, and many of them have issues that $100 computer speakers don't have (JBL LSR308 hiss, Tekton Electra/DI beamyness and lobing, both vs JBL 104,) or the manufacturer talks about a "drawback" to something like a coaxial driver even though there was a whitepaper on a speaker that solved the issue a decade ago (Gelenec 8260), or they company could have bent, elongated, and enlarged a port to get rid of chuffing on a $2,300 set that chuffs way more than an $800 set that goes even louder (Elac Navis vs Kali IN-8.)

To avoid these issues with sound panels, looks like making some will at least be cheap and as good as what's on the market for absorption. Before I bought my house I was shopping for a $10k 3D printer to try making the grid shaped material in 2' x 4' panels, and I've emailed one of the guys who made the ring shaped dampener for air flow through tubes... Maybe at some point I'll go down that path again as some schmuck with a hobby until billion dollar companies have a first mover peer that bankrupts them or makes them change.
I am thinking of making a panel. If I make any progress I will be sure to post here. Right now I am trying to be lazy and use code to generate the panel.

I too have been looking for a speaker and don't like the idea of spending $1 towards something that hisses. Let me know what you find.
Could you please ask him for an update on the applications side of the research? A few other questions:
  • Could he provide a description of what happens to the air at the cavity mouths to achieve these effects?
  • How did he and his team think of this approach? Why did they decide to pursue this work?
  • Is <1kHz practical with this technology? What the lower frequency limit would be?
  • Is true broadband diffusion possible? How linear would it be across the spectrum? Are they able to control linearity?
  • Has he considered how to deal with lobing/periodicity effects when placing many of the same diffuser side-by-side?
  • Does he know of any companies working to make this technology available for retail or custom orders (i.e., if it's actually been put to use outside of laboratories)?
  • For absorption only, what is the effectiveness of this technology? Can it be used for linear absorption <500Hz? <100Hz?
Maybe your chronic absence for the past 5-6 years has made him miss you.:)
A couple of those can be answered in the supporting doc:

"Supplementary Note 3. Scalability of the MSD The simulated phase responses of unit cells (5cm and 0.25cm thick) at 343Hz and 6860Hz are shown in Fig. S3(a)."

"The maximum energy absorption coefficient is 0.16 for 6860Hz and 0.04 for 343Hz, respectively. "

Here is a 2d calculator that will get you close for values, you just need to change the depth and add the holes for the phase.

http://www.oliverprime.com/prd/

more reading:
http://www.sea-acustica.es/fileadmin/publicaciones/Guimaraes04_ID156.pdf

A key part of the docs for me:

"Supplementary Note 4. The influence of period number of QRS. Figure S4 shows the simulated normalized diffusion coefficients of MSD with different number of periods: 1×1, 2×2, 4×4, and 6×6 with normal incidence. The results show that the diffusion coefficient slightly deteriorates as the period number increases for both SD and MSD, which agrees with the conclusion in Refs. 2 and 3. In addition, once the period number exceeds 1×1, the trend of the diffusion coefficient curve becomes stable, although with different amplitudes at different unit-cell numbers. This result, therefore, validates the comparability between SD and MSD at a period number of 2×2."
So you don't want a bunch of these panels across the wall and if you do no more than 2x2. This is what makes me what to generate wall-sized sequences rather than small panels that get less effective as you repeat them.

The one part of all of this that I don't understand is how to determine the wall thickness of the upper cavity of the cell with the hole in it.

My next step is to dump some values into a treemap and see what it looks like.

For my ideal panel I'd like to define the range of frequencies and have some weight to ranges to compensate for preference and the differences of absorption between frequencies.

https://journals.aps.org/prb/abstract/10.1103/PhysRevB.99.024302
Would be neat to try this out for each chamber and combine tech into one panel.

would a round cell perform better or worse? round speakers are bad, is a round diffuser good?
 

Attachments

Last edited:
Top Bottom