[Rclark]

You think that would be better? cut strips and glue them to a backing? I might just take a step back and consider that. Gik has a QRD diffuser that is just a solid piece of EPS and so that was my inspiration. And I'd have four of them I can place together for one large pattern or pull them apart and have four 2*4 diffusers.

I can visualize the cuts, and if I take it nice and slow I think I can be very precise. Haven't decided on wall thickness yet, maybe half an inch?

[neekomax]

The issue that I foresee as a possibility is that, depending on the tool and technique you use, some of the cuts from the solid block might be kinda awkward to execute. 

[Rclark]

Well the knife is very slim, it's basically a hacksaw blade. If I can go slow I think I'll be alright. I hope. No maybe you're right.

..reappraising.

[Maritan]

Seems to be it would be easier if you cut the separate parts first. What do I know? I haven't tried to build my own.

Are you doing the fins to separate the wells too?

[Rclark]

I think I am going to change course and just cut the pieces and glue them to boards. I'll go buy some new foam and maybe some 2*4 cardboard pieces for backing. I still want to be able to seperate it rather than have it be one permanently large, single diffuser.

[neekomax]

I think I am going to change course and just cut the pieces and glue them to boards. I'll go buy some new foam and maybe some 2*4 cardboard pieces for backing. I still want to be able to seperate it rather than have it be one permanently large, single diffuser.

Seems like a decent plan.

[TRADERXFAN]

You are going to have to be pretty crafty to do this modular approach.

On one hand you would need a large # sequence for a large diffusor.  But if you break the sequence in half, from what I gather, performance suffers dramatically.

But if decided to make a large diffusor composed of a repeating sequence, I think this leads to lobing and performance suffers...

- Well,  then I guess if above is true, it would mean you need should do a mix of sequences as smaller diffusors and put together into the large diffusor. This way you could later separate...

[Rclark]

I think I read that if you have multples, say 4 like I'm going to do, you would have 3 the same, and 1 with a different pattern.

[Rclark]

"This topic is covered in the chapter entitled The curse of periodicity in the book : Acoustic absorbers and diffusers: theory, design, and application By Trevor J. Cox, Peter D'Antonio, considered to be the bible when it comes to QRD diffusers. They advise that the excessive lobing can be offset by modulating the sequence via the introduction a second style of panel, positioned as per the Barker code.
 

Barker codes - based on info from Wikipedia

Where there is a +1 in the sequence, use your standard panel.
 Where there is a -1 in the sequence, use a different panel, which has a different set of diffusion angles to the others"

http://www.subwoofer-builder.com/qrd.htm

[Rclark]

Ok, back into the fray, 4 oak hardboard panels



And some extra tools to hopefully make this go nicer.



I can just take the stack I've already glued together and just cut into it for fins too, so no waste.

A benefit of going this route is the diffusers will be a bit more photogenic, hopefully. Thanks for steering me right fellas.

[Rclark]

Temporarily detoured. Just put my foot through my laptop.

[Alex Reynolds]

Temporarily detoured. Just put my foot through my laptop.

 

This is like one of those moments where you see something in your Facebook feed and you're not sure if you should "Like" it or not.
Sorry to hear that! Hope it was an inexpensive laptop!

[Rclark]

  VGA port still works and I just pulled down my 37 in LCD tv. All good. It was the screen I broke. It's an ASUS and I will buy another. Great stuff when you don't step on them.

[TrungT]

"They Don't Make Em Like They Use To"
 

[Rclark]

 

Well, using this big screen is definitely nicer, but it sucks to have to lug it around. Oh well, at least I can get this done now.

So one thing I've noticed, and was confusing the HELL out of me, was this thing about fin width and well width. I just thought they were the same thing.

But then you see on these wooden diffusers that they have very thin fins and wide wells. Styro diffusers are just fins. Or, rather it's almost like they are just wells.. no real "fins" on a styro one.

Anyway, I think I'm going to just set it to 1" fin, 1" well, 23 wells per diffuser and just go to town. Going to get busy slicing up all my pieces into 4" long, one inch strips and then glue them to the height need per well.

I have to do it this way since I bought 2" thick panels. I suppose I could have read all this first.

Anyway, it appears that with 1" fins and wells, I can get a scatters-at-250hz and works up to around 7000Khz diffuser. Seems pretty good. And I have a feeling this project will break me in and I can do some more treatment type stuff.

(2D's for my front wall)

[Rclark]

What are viscous losses?

I am torn between a 1" well, which the calculator says I get good performance across the board with about a 7Khz cutoff, and a .5" well, which gives more like a 13Khz cutoff, but says "viscous losses" likely.

Is that because you are trying to do too much at that point? Is it best to keep it in the 500-7khz range, and then use a different sort of diffuser for ranges above 7khz?

Or should I go for it and build one that does 500-13Khz and dare the "serious viscous losses", whatever those are?



Data for One Dimensional  QRD (Quadratic Residue) Diffuser

Advanced N47+2,0 panel:    Shifted down 2 depth units

Design frequency  500 hz

Number of wells:  47

Deepest well in inches:  12.71

Fin width in inches:  .51

Width of wells in inches:  .51

Ratios       2       3       6       11       18       27       38       4       19       36       8       29       5       30       10       39       23       9       44       34       26       20       16       14       14       16       20       26       34       44       9       23       39       10       30       5       29       8       36       19       4       38       27       18       11       6       3

Well depths in inches:      .57      .86      1.73      3.17      5.2      7.8      10.98      1.15      5.49      10.4      2.31      8.38      1.44      8.66      2.88      11.27      6.64      2.6      12.71      9.82      7.51      5.77      4.62      4.04      4.04      4.62      5.77      7.51      9.82      12.71      2.6      6.64      11.27      2.88      8.66      1.44      8.38      2.31      10.4      5.49      1.15      10.98      7.8      5.2      3.17      1.73      .86

Block heights in inches:      12.13      11.84      10.98      9.53      7.51      4.91      1.73      11.56      7.22      2.31      10.4      4.33      11.27      4.04      9.82      1.44      6.06      10.11      0      2.89      5.2      6.93      8.09      8.67      8.67      8.09      6.93      5.2      2.89      0      10.11      6.06      1.44      9.82      4.04      11.27      4.33      10.4      2.31      7.22      11.56      1.73      4.91      7.51      9.53      10.98      11.84

Phase shift in degrees       15       22       45       84       137       206       291       30       145       275       61       222       38       229       76       298       176       68       337       260       199       153       122       107       107       122       153       199       260       337       68       176       298       76       229       38       222       61       275       145       30       291       206       137       84       45       22

Scatter 250 Hz   Diffuse 500 Hz   HF cutoff 13230 Hz

Period width  48.02 inches

Minimum distance to seating position  81.25 inches (3 times wavelength of lowest diffusion frequency)

Good result: Period width of  48.11 inches is not less than the design wavelength of  27.16 inches

Poor result: The well width less than the minimum  .98inch to avoid viscous losses - Serious viscous losses may be encountered

Good result: Plate frequency of  23500 hz is higher than HF cutoff of  13230 hz, so it has no effect

Good result: The HF cutoff frequency of  13230 hz is  26.4 times the design frequency of  500 hz

True diffusion frequencies:   500   1000   1500   2000   2500   3000   3500   4000   4500   5000   5500   6000   6500   7000   7500   8000   8500   9000   9500   10000   10500   11000   11500   12000   12500   13000 hz

Current strategy for recommending well width of  .98 inches: Recommend well width so that wells are at least  .51 inches wide to minimise viscous losses

[Rclark]

 Slight update

Well I just decided the heck with it, I'll go ahead and build to 1". Now that I'm all tooled up, these won't be the last diffusers I build, etc.


But playing fast and loose is now costing me. Should have read QRdude first, now the 2" panels are too thick. So I have to cut all my panels into 1" strips, and stack them into my fin sizes.

I started this morning, making some marks, and using my new chalk line to mark off the lines to be cut


my technique with the saw left a lot to be desired...





So I pulled out my power sander to grind the few pieces I cut down to 1"






So cutting with the hand saw is both slow and crappy, so needless to say, if I continued doing this to each piece of foam, I'd probably run out of fun points in a major hurry, so today I went out and upped my arsenal.


Nice Jigsaw.




Now everything should go nice and speedy. The guided, powered blade will be far more accurate. Anyway, let this be a lesson to you all, use the calculator first before you buy your foam.

[PDR]

No matter what the outcome......you have to respect a guy that just goes for it.
Its been my motto for years.....go Clark go!!.....

[Rclark]

thanks! 

I'm still excited and hopeful for a very good outcome.

[Rclark]

Oh wow. The jigsaw is like magic. Project saved. I've just named this Skilsaw "Deus Ex Machina" for its as apt a name as could be.