A Few Questions on Polycylindrical Diffusers

[Trdat]

I have been edging to add some diffusion to my small room which is set up landscape orientation 4.7 metres front to back and 5.2 metres from side to side. The back has openings on both sides and I really dont have much space to add diffusion as most of the space is taken up by absorbtion. RT60 for the room is between .2 and .5 so decay is decent and have posted a measurement of my room in another post. I know Polycylindrical diffusers are technically old technology and have read all the arguments on Gearspace about it even the concept on the fact that its not a diffuser but the consensus is that it is better than a bare wall. So I'm under impression it might provide some benefit and I know there are better options out there but from a DIY and lack of space perspective it seems like a decent option over a bare wall.

1. One set of instructions mentions that it should be left to vibrate and only be held by its vertical slats that hold it in place. Is this concept correct? What about resonances from the 3mm plywood. Or does that just go with the territory or does bracing it improve its ability to scatter sound?

2. The same set of instructions mentions a height of 15cm is a decent arc for scattering. Does anyone have any idea if a higher arc is better for scattering frequencies. Jen Eklund used to add the parameters in some program on Gearspace but his not around anymore. He was not totally against poly's just thinks there are better options.

3. And if the panel is say on the top front wall which will be fairly out of way of reflections is there any point? Or would a diffuser still serve a purpose in any place as reflections will sooner or later get to that point?

 

[fluid]

There is some information and testing from the book "Acoustics of small rooms" by Jiri Tichy and Mendel Kleiner.

"All diffusers are not created equal, however, the differences in acoustic properties between different surface irregularities can be heard. In an auralization study that used a 1:10 scale modeling five types of diffusers was compared using a listening test [30]. Linear arrays of 10 diffusers such as pyramids, polycylindrical, quadratic residue, wedges, and spheres were used. The arrays filled the same area. Sample drawings of single diffuser elements are as shown in Figure 9.19. Wedges and pyramids affected the sound quality in a very negative way due to their selective specular reflection of high-frequency sound, while a series of polycylinders resulted in coloration because of grating effects. Figure 9.20 shows partial results for the scale model frequency response of QRD diffusers versus a random arrangement of spheres used in the tests. The results showed clearly that the frequency response and its resulting timbre and coloration were decisive for the sound quality of the diffuse reflected sound. The spheres turned out to be the most pleasant sounding because of their smooth frequency response. Note that caps and inverted caps (cavities into the wall) will work similarly well for scattering. In old-style control rooms, it was common to see polycylindrical diffusers, on walls and in corners. Particularly, arrays of polycylindrical diffusers on walls may cause unpleasant repetitive grating effects. They should have different radii of curvature and be placed at different angles or at random. Any orderly array should be avoided".

 

[Count Arthur]

Looks at all the clutter in my home office/listening room - nailed it.

On a serious note, what should the spherical caps be made of? Could they be a lightweight foam, or do the need to be fairly dense, or have a hard surface?

 

[Frgirard]

I have been edging to add some diffusion to my small room which is set up landscape orientation 4.7 metres front to back and 5.2 metres from side to side. The back has openings on both sides and I really dont have much space to add diffusion as most of the space is taken up by absorbtion. RT60 for the room is between .2 and .5 so decay is decent and have posted a measurement of my room in another post. I know Polycylindrical diffusers are technically old technology and have read all the arguments on Gearspace about it even the concept on the fact that its not a diffuser but the consensus is that it is better than a bare wall. So I'm under impression it might provide some benefit and I know there are better options out there but from a DIY and lack of space perspective it seems like a decent option over a bare wall.

1. One set of instructions mentions that it should be left to vibrate and only be held by its vertical slats that hold it in place. Is this concept correct? What about resonances from the 3mm plywood. Or does that just go with the territory or does bracing it improve its ability to scatter sound?

2. The same set of instructions mentions a height of 15cm is a decent arc for scattering. Does anyone have any idea if a higher arc is better for scattering frequencies. Jen Eklund used to add the parameters in some program on Gearspace but his not around anymore. He was not totally against poly's just thinks there are better options.

3. And if the panel is say on the top front wall which will be fairly out of way of reflections is there any point? Or would a diffuser still serve a purpose in any place as reflections will sooner or later get to that point?

Use the search module on Gearslutz and the Gearslutz community to find your response like this

https://asa.scitation.org/doi/10.1121/1.1916170

Volkman polycilyndral diffuser in a room acoustic design.

 

[youngho]

1. One set of instructions mentions that it should be left to vibrate and only be held by its vertical slats that hold it in place. Is this concept correct? What about resonances from the 3mm plywood. Or does that just go with the territory or does bracing it improve its ability to scatter sound?

2. The same set of instructions mentions a height of 15cm is a decent arc for scattering. Does anyone have any idea if a higher arc is better for scattering frequencies. Jen Eklund used to add the parameters in some program on Gearspace but his not around anymore. He was not totally against poly's just thinks there are better options.

3. And if the panel is say on the top front wall which will be fairly out of way of reflections is there any point? Or would a diffuser still serve a purpose in any place as reflections will sooner or later get to that point?

Alton Everest's Master Handbook of Acoustics may have some potentially useful information to you. There are different sections for the absorption versus the diffusion aspect of polycylindrical elements:

1. In short, he suggests creating airtight cavities behind the surface of the poly, with bulkheads at each end and randomly spaced between to create cavities of different volumes. The bulkheads should have adhesive rubber weatherstripping along the edge to prevent rattling. The resonant nature of the plywood creates a membrane for absorbing bass, which is enhanced by filling the cavities with absorption. Bracing should not affect the poly's scattering.

2. "The dimensions of such diffusors are not critical, although to be effective their size must be comparable to the wavelength of the
sound being considered. The wavelength of sound at 1,000 Hz is a bit over 1 ft, at 100 Hz about 11 ft. A poly element 3 or 4 ft across would be
effective at 1000 Hz, much less so at 100 Hz. In general, poly base or chord length of 2 to 6 ft with depths of 6 to 18 inches meet most needs."

3. Probably not, but I was surprised to see how Blackbird Studio C used RPG BAD Arcs in unexpected (to me) areas. It might be useful to think of it as scattering, more than diffusion. If it's really the top of the front wall near the ceiling, that might be a place for absorption (like a corner or soffit trap bass trap).

Toole (2nd edition of Loudspeakers and Rooms): "Figure 21.11 shows the normalized diffusion coefficient for a single and then
for multiple hemicylinders with a 1 ft (0.3 m) diameter. Obviously, in isolation the device works very well, and in this thickness it is effective down to a usefully low frequency. However, when combined with others, it loses the ability to diffuse low frequencies. The geometric regularity is also visible in the cyclical pattern in the curve. The solution is obviously to space them, probably by random distances, and possibly to vary the depth. Changing angles is another option. As Gilford (1959) noted, other protruding, curved, or faceted convex shapes also work well. In home theaters such shapes can be incorporated into interior designs as vertical columns."

See attached for the image.

Hope that helps.

 

[Trdat]

Alton Everest's Master Handbook of Acoustics may have some potentially useful information to you. There are different sections for the absorption versus the diffusion aspect of polycylindrical elements:

I got that book I will have to read through it more carefully, thanks for pointing our it had info on poly's. But your first answer I need to get my head around, air tight cavities and bulkheads? What does that mean, does it explain in the book?

Your answer to number 2 was perfectly explained, my poly's will be smaller but thats okay I get how large they have to be for them to be effective down low. Does the book mention radius height? That is what I was after, it seems a elipse is more effective than a semi circle?

Well, I suppose sound waves are scattered throughout the room and sooner or later will hit a diffusor and as long as they hit the listening position after around 6-12 ms(or the supposed early early reflections time gap) those reflections will add envelopment. So technically having them in the not so ideal position might still yield results but ultimately you should know how many ms you want it to arrive to you and ensure that it does get to the listening position. I just need to know how to angle the diffusors or Poly's to get the scattered or diffused soundwaves to the listening position that is my next goal. Angle of incident I think it called(not sure), need to get onto that concept.

The last point about Toole's testing and the solution has come up many times in Gearspace forum and is understandable, they recommend the same with binary template scattering to alternate the pattern but poly's should have some space and gap between them.

 

[youngho]

The cavities are the airspaces behind the plywood poly surface. If they are airtight, they act as a small resonant chamber with some frictional losses, resulting in some absorption, which is enhanced at modal frequencies by adding absorptive material like fiberglass or mineral wall. The bulkheads are curved (semicircular or arc) pieces of wood or plywood that are placed at the ends of the poly and at some places in the middle. Imagine looking at the poly in cross-section--that's the shape of the bulkhead. There are pictures in the book. If you're concerned about bass absorption but only scattering, you don't have to make them airtight at all.

It does mention depth, but unfortunately, there is no specific data or formula given. It seems to me that width would be related to wavelengths that are re-reradiated/scattered and that depth would be related to scattering/reradiating angle, but I'm sure more complicated than that.

I don't understand the thought process in the next paragraph. I think the end of the section called "The Schroeder Diffuser" will be helpful.

 

[Flaesh]

in some program

AFMG Reflex https://www.afmg.eu/en/afmg-reflex 30 days trial version

 

[fluid]

Cox and D'Antonio's book "Acoustic Absorbers and Diffusers Theory, design and application" contains a lot of information on the design and background of all sorts of diffusors. The section on curved surfaces mentions geometric theory as being a good way to predict the effects with these references.

J. H. Rindel, ‘‘Attenuation of sound reflection from curved surfaces’’, 25th Conference on Acoustics (1985).
A. D. Pierce, Acoustics, an Introduction to its Physical Principles and Applications, McGraw-Hill Inc. (1981).

If you want to look to see if the information is useful rather than purchase Cox and D'Antonios book straight out then a quick google of the title with pdf could be helpful.