Too many diffusion questions!

[asmonica]

hello everyone, ive been reading everest's master handbook of acoustics, and doing research like crazy.
1: i know the quadratic residue sequences, but i need to know the hard rules for relationships of well depths to widths and heights. also, when you get a depth sequence of say, 0, 1, 3, 2, 0= flush with the front surface.. if 1= 2" deep, does that mean 2= 4" deep?

2: in terms of density, if i was to use ceiling tile for the squares of a 2-d QRD(like the rpg hemiffusor), would it be dense enough as long as i sealed the surface with wood glue or something, and stuffed the cavities with fluffy fiberglass batts?(i know 703 is better, but i found 3 garbage bags full of r-19 in the dumpster behind my job)my reasoning is that any frequency that would resonate inside the cavity would not have sufficient energy to just pass through the material like it was nothing and would be mostly reflected. instead of trying to diffuse low end, i plan on taking ethan's advice and absorbing as much of is as i can.

my goal right now is the utilize the things i can get for free; various building materials that people have thrown away, and discarded walmart furniture, to make free treatment. so far ive got tons of particle board from cheap computer desks etc.. cieling tiles, lengths of hardwood flooring, and ive been finding alot of r-19.

i already made a 2-d QRD with 2x2's cut up into different lengths, but its like 30 pounds, and took a whole day to make, and is only like 18" x 18"..

my goal is to make a bunch of RPG hemiduffusor style diffusors cause they look cheap to make and lighter than the solid block style i made earlier.

any help with the exact specifications of such a diffusor would be greatly appereciated! also, any word as to whether im on the right track.
thanks in advance!

[Jason Jones]

Hey,

You are right about picking the depths. For the widths, I woiuld not go more narrow than 2".

As to the material, you need to use material that is reflective, not absorptive.

Let me know if this doesn't answer your questions.

Jason

[asmonica]

thanks for answering my queries! i appreciate it, as most of the people in this forum seem to be here only to prove their wit... so you are saying that the steps of the quadratic sequence are all multiples of the same base measurment, ie: if 1", then 1=1" 2=2", etc... and in reference to you saying i shouldnt go under 2" inch width; in alton everest book, he refers to an RPG diffusor- the QRD 1911 - which utilizes a prime-19 quadratic sequence and a 1.1" width(hence the model name). is the book outdated, or wrong? in this field of personal fancy over fact, i have a hard time discerning between the legitimate physics based facts with overinflated opinions..(totally not referring to you, just making an umbrella statement)
again, thanks in advance!

[andrebrito]

Those are numbers that are multiplied by a specific constant to obtain different thickness. 1 can be one inch or not.

Everest is not wrong.

[Jason Jones]

Originally Posted by asmonica

thanks for answering my queries! i appreciate it, as most of the people in this forum seem to be here only to prove their wit... so you are saying that the steps of the quadratic sequence are all multiples of the same base measurment, ie: if 1", then 1=1" 2=2", etc... and in reference to you saying i shouldnt go under 2" inch width; in alton everest book, he refers to an RPG diffusor- the QRD 1911 - which utilizes a prime-19 quadratic sequence and a 1.1" width(hence the model name). is the book outdated, or wrong? in this field of personal fancy over fact, i have a hard time discerning between the legitimate physics based facts with overinflated opinions..(totally not referring to you, just making an umbrella statement)
again, thanks in advance!

1. Yes, that is correct. So for a 1-D, base 7 QRD:

Base or reference depth of 2", then:

0 = 0"; 1 = 2"; 4 = 8"; 2 = 4"; 2 = 4"; 4 = 8"; 1 = 2"

2. As to your second point, I have two comments:

a. The point of narrowing the wells is to increase high frequency performance. At 2", you already have great high frequency performance. In fact, the Auralex Space Array has 3" wells and a .84 scattering coefficient at 4Khz.

I pick the Auralex for example, because I know the lab it was measured at and how it was done - their data is real, no BS.

At a 2" well width you will have great performance beyond 10Khz. The more narrow your well widths go you have three problems:

1. Absorption increases.
2. The work to build them increases.
3. Your sequence will repeat more often, which will decrease its effectiveness.

So, it is not that 1" is a disaster and 2" is the holy grail. Everest is not an idiot or anything like that. A 1" design will have some performance. But, a 2" design will take less work to build, with less absorption and no performance compromise.

b. Make sure you are using a 2-D sequence. I don't remember the one that you are referencing, but I think I remember that book having some earlier designs that were using 1-D sequences that were not properly translated to 2 dimensions. I am pretty sure that book also has some that are proper 2-D sequences in it, though. Most proper designs will reference the chinese remainder theorem. That is the most popular way to make it a real 2-D sequence.

Hope that helps. Feel free to fire away any more questions.

Jason

[asmonica]

thanks somuch for the info! very helpful. i was reading the bbc paper on diffusion, and they said you just set the same sequence across adjacent corners, and to get the remainder of the well depths, you just add the row and column and if its a higher # than your maximum well depth, you subtract from the prime to get a remainder..
also, which do you think works better: the omniffusor type with concave wells, or the skyline type with no well dividers?

again, thanks!

[Jason Jones]

Originally Posted by asmonica

thanks somuch for the info! very helpful. i was reading the bbc paper on diffusion, and they said you just set the same sequence across adjacent corners, and to get the remainder of the well depths, you just add the row and column and if its a higher # than your maximum well depth, you subtract from the prime to get a remainder..
also, which do you think works better: the omniffusor type with concave wells, or the skyline type with no well dividers?

again, thanks!

I am not sure that I am following what you are saying about the BBC paper. Could you post a link to the paper?

The difference between the skyline style and the omniffusor style is minimal. The skyline type can go deeper with a little less absorption than the omniffusor style...but not a big difference either way.

Jason

[avare]

Originally Posted by Jason Jones

I am not sure that I am following what you are saying about the BBC paper. Could you post a link to the paper?

I think this it 1990-15 pdf page 9.

Andre

[asmonica]

yeah thats it Avare thanks

[Seamus TM]

Originally Posted by Jason Jones

2. As to your second point, I have two comments:[INDENT]a. The point of narrowing the wells is to increase high frequency performance. At 2", you already have great high frequency performance. In fact, the Auralex Space Array has 3" wells and a .84 scattering coefficient at 4Khz.

I pick the Auralex for example, because I know the lab it was measured at and how it was done - their data is real, no BS.

At a 2" well width you will have great performance beyond 10Khz. The more narrow your well widths go you have three problems:

1. Absorption increases.
2. The work to build them increases.
3. Your sequence will repeat more often, which will decrease its effectiveness.

So, it is not that 1" is a disaster and 2" is the holy grail. Everest is not an idiot or anything like that. A 1" design will have some performance. But, a 2" design will take less work to build, with less absorption and no performance compromise.

Hey Jason,

On the absorption as a product of well width: isn't this in direct relation to the depth?

I would think that if you had a width of 1" and a depth of 6", you would be absorbing a lot more than with a depth of 3".

Just thinking out loud to you,
Seamus

[Jason Jones]

Originally Posted by Mumblesound

Hey Jason,

On the absorption as a product of well width: isn't this in direct relation to the depth?

I would think that if you had a width of 1" and a depth of 6", you would be absorbing a lot more than with a depth of 3".

Just thinking out loud to you,
Seamus

Deeper does mean more absorption and shallower does mean less. I don't think it is exactly a straightforward linear relationship, though.

[Seamus TM]

Originally Posted by Jason Jones

Deeper does mean more absorption and shallower does mean less. I don't think it is exactly a straightforward linear relationship, though.

Thanks, Jason.

No, I doubt that it would be that simple, either.
However, I wonder if it would be safe to assume that if the well width was more than the deepest well, you would have very little absorption...
...maybe...

EDIT:
...of course, that would approach a flat surface, making the diffusor kind of worthless, so it, obviously, would have less absorption.

Don't mind me, I like reasoning with myself.