The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Gear for sale  Search Gearslutz Go Advanced
2 ch listening room, aiming for best possible.
Old 19th July 2009
  #1
Gear maniac
 

Thread Starter
2 ch listening room, aiming for best possible.

Hi all

I am aware that this forum is mainly into 'pro' stuff, studios etc. But hey, acoustics is acoustics right? and so I hope you don't mind me taking a slight tangent and asking for your help with my listening room.

Maybe it will be a nice diversion from the usual??

I have lots of questions, mainly about diffusion (once you 'master' the basic concept of absorption, it's pretty well covered really...as much as you can, as thick as you can type stuff. Was I close?? ha ha).

You can't help too much if I don't give data, so for starters here is a sketch of the plan of my room.



It seems large to me as a listening room (ie I compare it to most peoples room) but maybe it is small in terms of diffusion etc etc??

The ceiling height is 4850, which is very close to the 4760 'width' of the room. Have not bothered with a mode calculator, but every measurement I have done has a big hump somewhere between 90-110 hz, so feel confident a mode calculator would spit that back at me.

The rectangles ( for eg in the bay area) are windows that are 3180 high, so no treatment possible there.

Hopefully I can load that sketch into paint and draw circles etc to show where the speakers and lp are located.

I have only just recently finished my speaker build, and now on to treatments. It is something I have 'always known' was important, so just as a quick experiment I opened a few bags of f/glass and threw them around in the room, including the old 'un-opened bag of f/glass in the corners' trick.





Raw batts, not even covered yet!! That is to the wall behind the speakers, basically behind me it is untreated at the moment. I just started and got bored at one stage, so not planned in any way but I guess I have atm LEDE.

My impressions?? Very favorable indeed. The room is well over 100 years old, and the walls are two foot thick solid brick. You can imagine the (what I presume to be) slap echo. I have been in this room for five years or so, and hung rugs on the wall to at least get rid of some of that (can just make it out on the left in the last pic). Still, plenty of bare wall left yeah!

So last weeks experiment really made a positive difference, and the funny thing is that each time I added some, I (possibly imagined) could hear where more needed to go.

Downside, even tho only half the room is done it is bordering on too dead for me. For sure, it's (obviously) fine when music is playing, but too much like a carpet showroom otherwise (when they have carpet hanging everywhere etc)

So, diffusion has strongly entered into my thinking as a result of that. Hey, even if I kept it as LEDE, am I not supposed to have diffusion behind me??

So whatever the case, I'm investigating diffusion.

Part of the enjoyment I get from diy is the 'how to do' side of it, and enjoy working out easier ways of doing things.

I have quite a few ideas in that regard, which I hope will be of interest to some of you when I get to the building, think of it as my exchange for your help!!

For example, I have a technique (to be tested tho) where I reckon I could knock up half a dozen diffusors on a saturday of any diffusor shape similar to this

GIK Acoustics: GIK D1 Diffusor

Note, I am not advocating industrial espionage here!! But, do any of you know of a shape along these lines that is known to work well as a diffusor?? If so, could you please post a link or cross section for me? Ie it is already in the public domain, intellectual property theft does not appeal to me.

This sketch appeared in another thread in this forum



presumably that was just a mock up to illustrate an idea. Such a simple shape like that you'd just bend a bit pf ply really, rather than my 'new idea', but if it had a bit more curve complexity etc then my idea could be very suitable for quick production.

So if you know of any such shapes I'd like to know too, thanks.

That is enough for a first post, hope I did not bore you all to tears, and hope that it would be an interesting side track for you all to give advice on something other that a studio.

Pending confirmation that it is OK to post a strictly 2 ch listening room thread on this forum, I guess I should ask a question or two.

Ok, what is the philosophy behind all this? That prob sounds like a dumb question, but what I'm getting at is this. If I came on and said 'I want LEDE room and here are the speaker locations and lp', then that would dictate whatever advice was to follow yeah?

An alternative approach is 'This is my bare room, I can put the speakers etc anywhere', well then we just 'treat the room' and make the room as good a performer as we can.

Did that make sense?? Ie, just because (by accident) I have something approaching an LEDE set up (and it is quite good from such a slapdash effort) does not necessarily mean I am wedded to it. Even from an aesthetic viewpoint an LEDE approach will be an 'unbalanced' look. A big part of my problem is getting the looks right in this historic and unique house (and as I hinted at earlier, that is part of the appeal in making it all work, fixing the aesthetic problems and being effective at the same time).

And, maybe I'm alone in this, absorption (tho effective) also looks pretty dull!!

So, diffusion and absorption and aesthetics, are you able to help??

Will start to ask some of the more 'technical' questions soon, this was a simple way of introducing myself and saying 'hi'.

Hi!!
Old 19th July 2009
  #2
Gear Guru
 
Ethan Winer's Avatar
 

Lightbulb

Quote:
Originally Posted by terry j View Post
do any of you know of a shape along these lines that is known to work well as a diffusor??
This program helps you design QRD type diffusors:

Acoustic Calculator

These videos featuring diffusion may also be of interest:

All About Diffusion
Hearing is Believing

--Ethan
Old 20th July 2009
  #3
Gear maniac
 

Thread Starter
cheers Ethan

have the link thanks. I certainly like the look of qrd's, and yes that is part of the palette that no doubt end up as the final package, if only to use it to get around the adverse effects of pattern repetition.

May as well ask that as part of this whole process, what are some of the methods we can use to avoid the 'nullifying' of a single diffusor that multiple diffusors bring about? Is it sufficient to use a 7 well diffusor followed by another size, would that avoid the repetition?

Or using semi circles say, just use different diameters? There is a diagram in one of the antonio pdfs on the net, where simply using an absorption trap between two diffusors also helps provide diffusion, yet surely must also avoid pattern repetition.

I have seen at least one of your videos, that you have linked to two means there is one I possibly have not seen yet, so thanks will check it out.

Still, there is no doubt that no matter the effectiveness of qrds, they are very time consuming and the level of difficulty is much greater than polys for example. (indeed, much to my surprise as a new guy to this, there is a great deal of debate whether or not the extra effort of qrd's is even worth it!)

So what I would like to see is a diffusor cross section (same sort of family as the gik one I linked to) as that would lend itself to my 'trick' in construction I mentioned first up. As I said, I reckon I could knock up half a dozen in a few hours, a far greater level of productiveness than building qrd's!!

AND, if my trick does indeed work, then I could easily do a you tube that shows how it is done so that all can benefit. (funnily enough, the trick has come from a totally disrelated field, a traditional (lost??) method used in building from long ago to do cornices etc. I learnt it as part of this house restoration so it's possible cross-over application to making diffusors required that unique blend of interests!)
Old 20th July 2009
  #4
Gear Guru
 
Glenn Kuras's Avatar
Hey,
I sent Jason Jones, our diffusion expert a email to have him chime in to give you a little help. He was the one that come up with the D1



Glenn
Old 20th July 2009
  #5
Gear nut
 

Quote:
Originally Posted by terry j View Post
May as well ask that as part of this whole process, what are some of the methods we can use to avoid the 'nullifying' of a single diffusor that multiple diffusors bring about? Is it sufficient to use a 7 well diffusor followed by another size, would that avoid the repetition?
You can use diffusers based upon different numbers (i.e. 7 and 5 or 7 and 11) or you can use a diffuser and for your "second" diffuser you flip the diffuser so it is oriented the opposite way of the first, or you could use diffusion and an absorber.

Once you have your two pieces picked you put them up based upon a pseudorandom sequence like the Barker code. That sequence will look something like this 1 -1 1 1 -1 1. Decide which diffuser (or the absorber) to use for 1 and then the other will go in the -1 spot.

I'd be happy to walk you through it when you get to that stage of planning.

Jason
Old 20th July 2009
  #6
Gear Guru
 
Ethan Winer's Avatar
 

Lightbulb

Quote:
Originally Posted by terry j View Post
there is a great deal of debate whether or not the extra effort of qrd's is even worth it!
The only solution I can think of, which is not practical, is to get a roomful of QRDs and also a roomful of polys, and see which group sounds better.

--Ethan
Old 21st July 2009
  #7
Gear maniac
 

Thread Starter
Ethan, the video I had not seen was the small room (narrated by someone other than yourself) with diffusion all round. thanks for that (I assume the other is you in front of polys and qrd's?? TBH, have my doubts about the validity of that one, I know you wanted to demonstrate worst case scenario, but still quite an invalid test I feel as they were both being used in a way that is simply inapplicable).

Got it about testing a room twice, once with polys and once with qrd's. Yeah, quite impractical really!!! I only really mentioned the 'dispute' (well, it IS a good point regarding the relative work levels!!) as it kinda made me laugh at my reaction (confusion to a degree).

What I mean is, (and traditionally I used to inhabit normal hi fi/audiophile forums..) a new guy to audio gets on and asks 'which cable should I buy?? *he knows that cables are important, all the reviewers and mags say so yeah??* and a guy like me says 'if it conducts, then it will do' and it spins him as it rocks his stable datum.

So I get on a 'scientific/pro' type forum and ask 'which qrd should I use?' *I know qrds are the rolls royce, all the studios and all the scientific literature say so yeah??* and some guys get on and reply 'not really better'....threw me into a spin as it rocked my stable datum!

So, in all of these fields, you gotta be able to sort the wheat from the chaff and get a sufficient understanding of it all that you can work it out reasonably well for yourself.

Else just pay a bloke and do what you're told! (and you will be happy, cause at least it WILL sound different!)


Quote:
Originally Posted by Jason Jones View Post
You can use diffusers based upon different numbers (i.e. 7 and 5 or 7 and 11) or you can use a diffuser and for your "second" diffuser you flip the diffuser so it is oriented the opposite way of the first, or you could use diffusion and an absorber.

Once you have your two pieces picked you put them up based upon a pseudorandom sequence like the Barker code. That sequence will look something like this 1 -1 1 1 -1 1. Decide which diffuser (or the absorber) to use for 1 and then the other will go in the -1 spot.

I'd be happy to walk you through it when you get to that stage of planning.

Jason
thanks Jason (and Glen). At times there is enough of the cox and d'antonio book available on google books that half of this stuff can be worked out. Re your 'flipping', (and what you mention is enough for me to have a better check tonight) is not, for example, a seven well qrd (heck, all qrd's, as I say I will check) symmetrical? Ie, flipping will end up presenting you with the same profile? In the cox book when they give the example of the -1 of the barker code, well I guess a good enough word would be 'the inverse'..and stangely enough it starts behind the normal ones, ie goes deeper. Heck, may as well make all of them deeper and diffuse lower! if you follow.

When I get sensible about it I realise it was a silly question for me to have asked. There is no need for me to have that problem in the first place (if diying). That problem really only arises if you are buying commercial stuff, AND you have decided to 'fill that wall in of length 'x'.

For commercial practical reasons, a company prob only has (at most) a few different models, and so to get length 'x' it rerquires a few of them (and hence we get the repeat problem.

Me? meh, just build a single diffusor of a longer length (greater base number, dunno 29 instead of four sevens or whatever) and no repetition problem. Maybe it will be slightly more work doing 'one 29' than 'four sevens', but better results.

But thanks for your help offer, it WILL be gratefully accepted as we proceed!

NOW, on to one of those 'technbical' questions hinted at, and this is really to start cleraing up some of my confusions...or was that diffusions heh heh...in this area.

Was going to quote and link (and will if needed) but maybe all will follow if I do this from memory.

For most of us, and a few of the skyline calculators out there, we get the idea of the BBC model when we use that term. In the BBC paper, they quite clearly mention that it is simplified in many ways to help ease manufacturing costs and effort, and some of those measures include quantising the heights (four was it) and only having a 12*12 grid. And here is the question, they also make it clear that the well dividers are removed, accepting that for floor and ceiling slap echo amelioration that will not lead to too many compromises.

Ok then, often it comes up that 'having wells in skylines' is not a bad idea at all, and it seems to stem from that type of thing in the BBC paper (?), and indeed I used to think that this



was simply a 'skyline with wells', except of course it isn't! (it's qrd based rather than prd based)

But nonetheless you get the idea.

So my confusion?? It was only last night that I finally printed off the patent for the skylines and went thru it.

There is NO mention in it at all about well dividers! (will double check tonight), no mention that for ease of manufacture 'we leave them out' or anything.

So is the BBC paper talking about something different yet 'we' have comingled them?

Again, the reason I'd like to know is I think I have a way where I could make an RPG skyline (with wells that is) in an afternoon...no wells and then it's back to the very long and tedious method. So it would be good to clear that up!

that's enough for this post, thanks for your time and hope to start moving soon.

Glen and Jason, any thoughts about a profile 'similar' to yours that works? Dunno, maybe you have a profile that is better than the one linked to earlier, but impractical to manufacture or that would be commercially too expensive or sumthin. Or one that you found 'good enough', but still not as good as the one you DID manufacture.

I could easily test my trick on 'any old shape', but no point is simply making a complex shape (ie proves it works) on a shape that is no good for diffusion. May as well get a product if I can out of the test.
Old 21st July 2009
  #8
Gear addict
 
Magickman's Avatar
 

Quote:
Originally Posted by terry j View Post
Ethan, the video I had not seen was the small room (narrated by someone other than yourself) with diffusion all round. thanks for that (I assume the other is you in front of polys and qrd's?? TBH, have my doubts about the validity of that one, I know you wanted to demonstrate worst case scenario, but still quite an invalid test I feel as they were both being used in a way that is simply inapplicable).

Got it about testing a room twice, once with polys and once with qrd's. Yeah, quite impractical really!!! I only really mentioned the 'dispute' (well, it IS a good point regarding the relative work levels!!) as it kinda made me laugh at my reaction (confusion to a degree).

What I mean is, (and traditionally I used to inhabit normal hi fi/audiophile forums..) a new guy to audio gets on and asks 'which cable should I buy?? *he knows that cables are important, all the reviewers and mags say so yeah??* and a guy like me says 'if it conducts, then it will do' and it spins him as it rocks his stable datum.

So I get on a 'scientific/pro' type forum and ask 'which qrd should I use?' *I know qrds are the rolls royce, all the studios and all the scientific literature say so yeah??* and some guys get on and reply 'not really better'....threw me into a spin as it rocked my stable datum!

So, in all of these fields, you gotta be able to sort the wheat from the chaff and get a sufficient understanding of it all that you can work it out reasonably well for yourself.

Else just pay a bloke and do what you're told! (and you will be happy, cause at least it WILL sound different!)




thanks Jason (and Glen). At times there is enough of the cox and d'antonio book available on google books that half of this stuff can be worked out. Re your 'flipping', (and what you mention is enough for me to have a better check tonight) is not, for example, a seven well qrd (heck, all qrd's, as I say I will check) symmetrical? Ie, flipping will end up presenting you with the same profile? In the cox book when they give the example of the -1 of the barker code, well I guess a good enough word would be 'the inverse'..and stangely enough it starts behind the normal ones, ie goes deeper. Heck, may as well make all of them deeper and diffuse lower! if you follow.

When I get sensible about it I realise it was a silly question for me to have asked. There is no need for me to have that problem in the first place (if diying). That problem really only arises if you are buying commercial stuff, AND you have decided to 'fill that wall in of length 'x'.

For commercial practical reasons, a company prob only has (at most) a few different models, and so to get length 'x' it rerquires a few of them (and hence we get the repeat problem.

Me? meh, just build a single diffusor of a longer length (greater base number, dunno 29 instead of four sevens or whatever) and no repetition problem. Maybe it will be slightly more work doing 'one 29' than 'four sevens', but better results.

But thanks for your help offer, it WILL be gratefully accepted as we proceed!

NOW, on to one of those 'technbical' questions hinted at, and this is really to start cleraing up some of my confusions...or was that diffusions heh heh...in this area.

Was going to quote and link (and will if needed) but maybe all will follow if I do this from memory.

For most of us, and a few of the skyline calculators out there, we get the idea of the BBC model when we use that term. In the BBC paper, they quite clearly mention that it is simplified in many ways to help ease manufacturing costs and effort, and some of those measures include quantising the heights (four was it) and only having a 12*12 grid. And here is the question, they also make it clear that the well dividers are removed, accepting that for floor and ceiling slap echo amelioration that will not lead to too many compromises.

Ok then, often it comes up that 'having wells in skylines' is not a bad idea at all, and it seems to stem from that type of thing in the BBC paper (?), and indeed I used to think that this



was simply a 'skyline with wells', except of course it isn't! (it's qrd based rather than prd based)

But nonetheless you get the idea.

So my confusion?? It was only last night that I finally printed off the patent for the skylines and went thru it.

There is NO mention in it at all about well dividers! (will double check tonight), no mention that for ease of manufacture 'we leave them out' or anything.

So is the BBC paper talking about something different yet 'we' have comingled them?

Again, the reason I'd like to know is I think I have a way where I could make an RPG skyline (with wells that is) in an afternoon...no wells and then it's back to the very long and tedious method. So it would be good to clear that up!

that's enough for this post, thanks for your time and hope to start moving soon.

Glen and Jason, any thoughts about a profile 'similar' to yours that works? Dunno, maybe you have a profile that is better than the one linked to earlier, but impractical to manufacture or that would be commercially too expensive or sumthin. Or one that you found 'good enough', but still not as good as the one you DID manufacture.

I could easily test my trick on 'any old shape', but no point is simply making a complex shape (ie proves it works) on a shape that is no good for diffusion. May as well get a product if I can out of the test.
Just wondering: have you actually ever tried to read one of your own posts?

Sheesh...
Old 21st July 2009
  #9
Lives for gear
 
Nordenstam's Avatar
 

Quote:
Originally Posted by terry j View Post
So I get on a 'scientific/pro' type forum and ask 'which qrd should I use?' *I know qrds are the rolls royce, all the studios and all the scientific literature say so yeah??* and some guys get on and reply 'not really better'....threw me into a spin as it rocked my stable datum!
Do believe the hype! Polys are great for what they do but they're crude and simple devices compared to number theoretical diffusers.

Quote:
Originally Posted by terry j View Post
In the cox book when they give the example of the -1 of the barker code, well I guess a good enough word would be 'the inverse'..and stangely enough it starts behind the normal ones, ie goes deeper. Heck, may as well make all of them deeper and diffuse lower! if you follow.
The reason is that the prime 7 QRD have a nice "free depth" bonus. The calculator says the sequence is 1422410, while the numbers in the math actually goes from zero to seven. So you get the performance of a seven deep diffusor with only 4 as the real depth limit. When you invert it, you'll have to invert it with respect to the real performance depth - the number seven.

Quote:
Originally Posted by terry j View Post
Me? meh, just build a single diffusor of a longer length (greater base number, dunno 29 instead of four sevens or whatever) and no repetition problem. Maybe it will be slightly more work doing 'one 29' than 'four sevens', but better results.
That's the way to go!

Quote:
Originally Posted by terry j View Post
So my confusion?? It was only last night that I finally printed off the patent for the skylines and went thru it.

There is NO mention in it at all about well dividers! (will double check tonight), no mention that for ease of manufacture 'we leave them out' or anything.

So is the BBC paper talking about something different yet 'we' have comingled them?
PRD's are much less sensitive to the lack of well dividers than the QRD's. Having dividers gives better performance of the PRD array, but then again, you'd need to have a bigger distance between the array and the sweet spot. Well-less also represents less of a reflective side area on the array, which may be an issue in some rooms.
Old 21st July 2009
  #10
Gear nut
 

Quote:
Originally Posted by terry j View Post

is not, for example, a seven well qrd symmetrical? Ie, flipping will end up presenting you with the same profile?
No they are not symmetrical. RPG takes the last well of their 7 well qrd and splits it in half - putting half on the front and half on the back. So their design is symmetrical. But a QRD is not symmetrical. (BTW, there is no performance increase to their slight design change).

Quote:
In the cox book when they give the example of the -1 of the barker code, well I guess a good enough word would be 'the inverse'..and stangely enough it starts behind the normal ones, ie goes deeper. Heck, may as well make all of them deeper and diffuse lower! if you follow.
The barker code has nothing to do with it being deeper. Just implementing the barker code will keep all diffusers at the same depth. In the D'Antonio and Cox book, they make a "big QRD" by using a "normal" 2 ft wide QRD for each well. This will put some of them deeper.


Quote:
Me? meh, just build a single diffusor of a longer length (greater base number, dunno 29 instead of four sevens or whatever) and no repetition problem. Maybe it will be slightly more work doing 'one 29' than 'four sevens', but better results.
That will solve your repeat problem. But, you will have no performance advantage over an array of smaller QRD's using the barker code implementation.

Quote:
Again, the reason I'd like to know is I think I have a way where I could make an RPG skyline (with wells that is) in an afternoon...no wells and then it's back to the very long and tedious method. So it would be good to clear that up!
The biggest issue with these smaller 2-D diffusers is absorption. With the wells you get quite a bit of absorption - especially if you make it very deep.
Old 21st July 2009
  #11
Gear Guru
 
Ethan Winer's Avatar
 

Lightbulb

Quote:
Originally Posted by terry j View Post
I know you wanted to demonstrate worst case scenario, but still quite an invalid test I feel as they were both being used in a way that is simply inapplicable
Not sure who suggested that to you. The intent of my video is to put you right there in the room, so you can hear first-hand what these surfaces sound like. And it does that very well. This is how I first "got it" when learning about QRDs, by standing right in front of one and talking into it. It's very revealing, and the best way I know to hear and convey the sound.

However, if you'd like to produce your own video that lets people hear this better for themselves, I'll definitely look forward to seeing it.

--Ethan
Old 22nd July 2009
  #12
Gear maniac
 

Thread Starter
Quote:
Originally Posted by Jason Jones View Post
No they are not symmetrical. RPG takes the last well of their 7 well qrd and splits it in half - putting half on the front and half on the back. So their design is symmetrical. But a QRD is not symmetrical. (BTW, there is no performance increase to their slight design change).
Duhh, silly me! How can an odd number be symmetrical?? What I meant is that once the first well has been divided in half then they become symmetrical..precisely as you pointed out!



But in the barker code implementation we would not do that??

Quote:
The barker code has nothing to do with it being deeper. Just implementing the barker code will keep all diffusers at the same depth. In the D'Antonio and Cox book, they make a "big QRD" by using a "normal" 2 ft wide QRD for each well. This will put some of them deeper.
Ok. I was confused by the drawings in the book

Acoustic absorbers and diffusers ... - Google Books

eg fig 9.17 'shows' the -1 being behind the plane of the normal ones. But the wording tends to back up my interpretation "This can be done by using the rear of a normal N=7 diffusor (provided the fins extend far enough)"

Oh well, your method is obviously easier!! so will go with that.


Quote:
The biggest issue with these smaller 2-D diffusers is absorption. With the wells you get quite a bit of absorption - especially if you make it very deep.
So I can conclude that (even with the RPG skylines) that wells seperated by walls is the preferred embodiment? not withstanding that no mention is made of that in the patent? (good for my method)

Hi Ethan, if I misrepresented you then please accept my apologies! Was reasonably sure I had read you saying that by being close to them it would 'exagerate' the effects.

Not much hope of me producing a video I'm afraid, best I can do is offer an alternative method of construction that others may find useful. We all contribute in our different ways, yet in hopefully complementary ways!

Thanks Lupo.

Ok, so we'd better start at the beginning (a very good place to start-or so the song tells me). We are after the best possible listening room, which may or may not include diffusion (and presumably we can get good results without it, after all 'most' treatment companies for a long time did NOT offer diffusion products, I doubt they would now claim - once they offer diffusion - that absorption only is insufficient).

So almost as a matter of course, the first step is broadband effective absorption. Best not to lose sight of that in all these diffusion ponderings.

Layout of the room



In the old setup, I satisfied the symmetry 'requirement' of listening rooms. The first reflections would be quite long (tho whether under 20 ms say I didn't ever bother to work out).

I read the article on the decware site and decided to give the diagonal alignment a go. I found it much better, clearer and more distinct etc. I have formed the theory that in the old setup, even tho I had plenty of space around the speakers (and to the side walls) I was, in effect, at the focal point of the bay behind me (and not too far from it). Dunno if that is the sole explanation, but I think it could be part of it. As that bay is basically 'all window', treatments there are out.

So big basstraps in the three main corners (one corner again is full height window, so no trap there, unless a small one under the window). No point in a six foot trap in a room with ceilings seventeen feet!, so these traps will have to be high, current thinking is either 3.6 metre or 4.8 metres. That has implications for stability and aesthetics, so not a simple job at all. Then first reflection treatment, most advocate/use absorption, others suggest diffusion can work too. Easy enough to test and compare and use what is best for my tastes.

In the pic you can see the area called 'trap'. Big diagonal traps in the front two corners 'makes sense', but would be odd in that location. Best to fill the entire section with a trap.

OR, and this is where we enter fantasy land(?), how about this for a wacky idea? It is about 400 mm deep, may as well add another couple of hundred mm, I could have a bass diffusor 6-700 mm deep, 1350 wide an make it ten foot high or sumthin. Would not look too out of place anyway, due to the natural recess it will be in.

Stupid/useless/pointless idea?? If it did achieve effective bass diffusion, what does bass diffusion sound like? Anyone have experience with that?

No bites on the solid diffusor cross section cherry, oh well. looks like that one has died. But just to test my idea, I will make a small high frequency diffusor in the next few days, fingers crossed that my idea will be a fast and effective method of production. We'll all find out soon enough.

For no other reason than it has already been designed and tested, I will try and clone this high frequency diffusor for the test

http://www.kineticsnoise.com/interio...true%20QRD.pdf

Now, what measurements would be useful to see to move this forward?? FR at the lp, waterfall, decay times??

Ethan, I think it was your video on diffusion that I saw this, you showed the difference in the room of the decay times across the frequency bands. Can you (or someone) give a bit of a rundown on what that can tell you about a room?? (we usually only see Fr or waterfalls etc, so that piqued my interest).

I can run a few quick measurements tonight and post them. (hopefully parts of this thread will be of interest to others in the future, so using measurements with the explanations is a good way to go?)
Old 22nd July 2009
  #13
Gear Guru
 
Ethan Winer's Avatar
 

Lightbulb

Quote:
Originally Posted by terry j View Post
Was reasonably sure I had read you saying that by being close to them it would 'exagerate' the effects.
Yes, though if you look at that photo from the cover of Mix magazine you'll see that diffusors can be used very close to instruments intentionally. I've done a lot of listening lately to diffusors close up, and have concluded that the "conventional wisdom" of needing to be many feet away from them is just wrong.

Quote:
Ethan, I think it was your video on diffusion that I saw this, you showed the difference in the room of the decay times across the frequency bands.
My newer video Hearing is Believing lets you hear the sound of diffusion in a room from a "normal" perspective, as if you were listening in the room. This video also shows the very large improvement in the uniformity of decay times versus frequency when a lot of diffusion is used.

--Ethan
Old 22nd July 2009
  #14
Gear Guru
 
Glenn Kuras's Avatar
Quote:
I've done a lot of listening lately to diffusors close up, and have concluded that the "conventional wisdom" of needing to be many feet away from them is just wrong.


Diffusion is always better then a flat wall, hands down. But if you really want the true effect of them you need space between the listening position, mic or what ever and the diffusor.
A few things for people to read to understand this stuff
Diffusion is a Wonderful Tool by Jeff Hedback
How Diffusion Really Works. Spotlight on MASSIVE Mastering.
It is not to say though that you can NOT use diffusion close to something you are recording to get a certain effect/sound. I really don't know of anyone that would disagree with that and that is "conventional wisdom" as far as I know it.
Old 22nd July 2009
  #15
Gear maniac
 

Thread Starter
Quote:
Originally Posted by Ethan Winer View Post
My newer video Hearing is Believing lets you hear the sound of diffusion in a room from a "normal" perspective, as if you were listening in the room. This video also shows the very large improvement in the uniformity of decay times versus frequency when a lot of diffusion is used.

--Ethan
Yes. That is the video I was referring to. The question was, can we move this forward by referring to in room measurements. If so, which do we need to see.

I don't mind going it alone on this project, and will not have much choice if things stay the way they are. I had thought that the process from go to whoa could have been a valuable resource for anyone to read later.

Maybe magickman was right, no-one can understand my posts or my 'style' turns them off. Sheesh indeed.

I can easily change my style from 'chatty' to 'hard nosed business-like' if that is needed.

Or not bother, ...whatever. The last thing I want to do is waste anyone's time, let alone mine.

I do understand that it can be a one-way street at times, someone get's all the help from others yet gives nothing back. That is why I pointed out that I felt I can give something back in return, a couple of new tricks that may help others in the future. I cannot help with the theory of course, but I may be able to help with the practical doingness.

So, assuming for now that it is worthwhile continuing this thread, is it of any use to provide measurements? Will that give us an idea of how to proceed and the map by which we will travel?

If that is of no use, then do you have any comments about the 'general plan' above, good broadband absorption as much as possible, then diffusors as desired. It may be as simple as that really, if so then thanks and away I go.

Your time is valuable, as is mine. Let me know which way to jump with this.
Old 23rd July 2009
  #16
Lives for gear
 
Nordenstam's Avatar
 

Hi Terry!

Keep on posting! Will be interesting to see how this project progresses. If you can build appreciable amounts of diffusion and measure the effect as you go, it would surely be interesting for many people to follow.

I, for one, am interested in your speedy diffuser build idea. Have built four square meters of 2D diffusion myself so far and it's going S L O W..

Quote:
Originally Posted by terry j View Post
If that is of no use, then do you have any comments about the 'general plan' above, good broadband absorption as much as possible, then diffusors as desired. It may be as simple as that really, if so then thanks and away I go.
If you plan to keep it live sounding, "as much as possible" of the absorption may not be the best way to go. Some places may be better off with a deep trap covered with diffusion.


Cheers,

Andreas Nordenstam
Old 23rd July 2009
  #17
Gear Guru
 
Ethan Winer's Avatar
 

Lightbulb

Quote:
Originally Posted by terry j View Post
can we move this forward by referring to in room measurements. If so, which do we need to see.
The three sets of measurement is shown in that video are what's needed:

Frequency Response
Modal Ringing
Reverb decay time versus frequency (RT60)

Quote:
I can easily change my style from 'chatty' to 'hard nosed business-like' if that is needed.
LOL, I can read you fine. There are so many posts that when I read through a long thread to see if I can contribute something useful, I scan for question marks. If I see no question marks I assume the post is just a bunch of chat and move on. heh

--Ethan
Old 24th July 2009
  #18
Gear maniac
 

Thread Starter
Quote:
Originally Posted by Ethan Winer View Post
The three sets of measurement is shown in that video are what's needed:

Frequency Response
Modal Ringing
Reverb decay time versus frequency (RT60)

LOL, I can read you fine. There are so many posts that when I read through a long thread to see if I can contribute something useful, I scan for question marks. If I see no question marks I assume the post is just a bunch of chat and move on. heh

--Ethan
I think I understand that ethan???????????? ha ha, just testing the new technique!!! (oops, I meant ??????)

Modal ringing, will have to look at the video again and see what that is, I should then be able to do all those measurements, thanks. (????)




Quote:
Originally Posted by terry j View Post


But in the barker code implementation we would not do that??
Ok, would like to clear this up. I can easily get the barker code (it is in the google books link above, something like 111-111 or whatever)

So, for the -1, all we need do is rotate/flip the qrd?? And for how many repeats do we need this to happen? The reason I ask will become apparent soon.

Ha, just had another look at that book, it seems there is a bit more to the Barker sequence than a set of numbers like that above, it varies depending on the N number....and they list for N=5 the sequence 1,-1,1,1,1. So 'flip' the qrd for the -1's, normal for the 1's???? (?????)



Quote:
Originally Posted by terry j View Post
So I can conclude that (even with the RPG skylines) that wells seperated by walls is the preferred embodiment? not withstanding that no mention is made of that in the patent? (good for my method)
Am very enthused now to try this this weekend, so a quick 'yes/no' to this question would be great.

More 'stupid' ideas time. Take a fractal qrd, at the bottom of the deep wells we have the smaller qrds, supposedly making the badnwidth greater. Ok then, as I will be trying my ideas on the following type of construction



has anyone ever tried having a smaller pattern in each well?? Rather than a flat face as you see here, think the skyline version of the fractal but applied to this. Again, I think very easy to do with my new method.

Quote:
Originally Posted by terry j View Post
OR, and this is where we enter fantasy land(?), how about this for a wacky idea? It is about 400 mm deep, may as well add another couple of hundred mm, I could have a bass diffusor 6-700 mm deep, 1350 wide an make it ten foot high or sumthin. Would not look too out of place anyway, due to the natural recess it will be in.

Stupid/useless/pointless idea?? If it did achieve effective bass diffusion, what does bass diffusion sound like? Anyone have experience with that?
No thoughts on this? (?????????? - for ethan hah hah) Proper bass diffusion, does it work and is it worthwhile. (tho I only have 700-800 to play with here)

Quote:
For no other reason than it has already been designed and tested, I will try and clone this high frequency diffusor for the test

http://www.kineticsnoise.com/interio...true%20QRD.pdf
Dunno if anyone looked at the qrd I wanted to test. I oringinally 'knew' that my trick would work for the type of diffusors I asked about in the first post, but got 'fed up' with waiting for some proven shape, and thought bugger it, I will try a proven qrd design (as in the link above)

Will get some time to take a photo and upload it later today, but yeah the new method worked out well. A couple of aborted attempts as I worked out the methodology...for a short while I had the horrible idea that my idea would not work after all....but then it came good.

I now have a 1.8 m length of the qrd in the link above (of course I can dock/make it to any length I want).

The downside? Well it took longer than I had envisaged (then again the first one always does), about two hours for the completed 1.8 m length.

But the good news? After having done it once, I know that I can make two of them (doing multiples speeds things up!) in about an hour.

Hence my question about the implementation of the barker code.

Hence my curiosity to see if the welled qrd version also lends itself to my ideas.


Quote:
Originally Posted by Lupo View Post
Hi Terry!

Keep on posting! Will be interesting to see how this project progresses. If you can build appreciable amounts of diffusion and measure the effect as you go, it would surely be interesting for many people to follow.

I, for one, am interested in your speedy diffuser build idea. Have built four square meters of 2D diffusion myself so far and it's going S L O W..

Andreas Nordenstam
Thanks Andreas, much appreciated. Slow eh? how slow??

Are you building a studio?

(Forgot to mention before, I estimate a cost of about $5 or so for that 1.8m length.)

EDIT just plugged the figures 1300 wide and 700 deep into the mhl calculator, threw back diffusion down to 174 hz, 1129 high. BUT, with my technique easy to throw smaller diffusors into the bottom of each well, hmm, bass diffusion may be on the cards! Just for kicks, even if it 'does not work' I will probably try it. Enjoy little learning projects like that.
Old 24th July 2009
  #19
Lives for gear
 
Nordenstam's Avatar
 

Hello!

Quote:
Originally Posted by terry j View Post
Ok, would like to clear this up. I can easily get the barker code (it is in the google books link above, something like 111-111 or whatever)

So, for the -1, all we need do is rotate/flip the qrd?? And for how many repeats do we need this to happen? The reason I ask will become apparent soon.

Ha, just had another look at that book, it seems there is a bit more to the Barker sequence than a set of numbers like that above, it varies depending on the N number....and they list for N=5 the sequence 1,-1,1,1,1. So 'flip' the qrd for the -1's, normal for the 1's???? (?????)
Yes!

The important thing is to follow the correct barker code sequence for the length you're using. It's listed in the wiki page. Or any other code that gives you the desired autocorrelation properties.

I'd still go for a continuous large prime QRD if you can, over a compound one based on many smaller. The barker code will help alleviate the repeat problems, but a larger unit doesn't have them at all.

Quote:
Originally Posted by terry j View Post
[omnifuser]

has anyone ever tried having a smaller pattern in each well?? Rather than a flat face as you see here, think the skyline version of the fractal but applied to this. Again, I think very easy to do with my new method.
I think fractalistic design is more useful with 1D diffusers. With 2D using 2" well width, the design frequency extends to about 3500Hz. Above this limit, the scattering will still be significant. If it extends 1-2 octaves above the limit, like it does below the number theoretical limit, it'll nearly reach ultrasonic.

Quote:
Originally Posted by terry j View Post
No thoughts on this? (?????????? - for ethan hah hah) Proper bass diffusion, does it work and is it worthwhile. (tho I only have 700-800 to play with here)
Bass diffusion should be just as great as high end diffusion, if you can afford the loss of space. Make sure the material used is substantial enough to reflect the LF.

Quote:
Originally Posted by terry j View Post
Will get some time to take a photo and upload it later today, but yeah the new method worked out well. A couple of aborted attempts as I worked out the methodology...for a short while I had the horrible idea that my idea would not work after all....but then it came good.

I now have a 1.8 m length of the qrd in the link above (of course I can dock/make it to any length I want).
Waiting in anticipation for the pic and explanation! =)

Quote:
Originally Posted by terry j View Post
The downside? Well it took longer than I had envisaged
Hehe.. Know the feeling!

Quote:
Originally Posted by terry j View Post
Thanks Andreas, much appreciated. Slow eh? how slow??

Are you building a studio?
Am spending something like 4 to 5 hours on each unit, each covering from 1/3 to 1/2 square meter. (so far, may build some bigger ones too) The plan is to cover at least 6 square meters. The first units where slower than that and the last ones will probably be faster. All in all about two weeks of work! Thank god I had time to schedule some "vacation" this year.. If I calculate the time spent into the price, it's about the same price as the premade ones. If it wasn't for the fact that I truly enjoy doing stuff like this and I much rather have custom sized units than the 2x2 feet only, I'd go straight for the premade units!

It's a living room, not a proper studio as such. Though it works out very well working in a residential area. The leakage is very low. It's been years since I treated it well enough for mastering use, but keep adding more acoustical treatment as time passes. Given that monitoring is alfa omega in mastering, acoustic upgrades is the best upgrades I know! It makes me happy just sit there and listen to great music, ever more amazed as the imaging gets clearer and clearer. It also translates directly into better mastering work. The starting point for diffusion is the opposite of yours. The room is trapped to death with a very dry sound. 8% of the cubic area of the room is dense mineral wool. Just added some more traps and I think I'm starting to reach the saturation point where adding more wont do much. I now want to add some life to the room without giving up on the extreme precision all the absorbers gives. That's where the neato diffusion comes into the picture..


Andreas
Old 24th July 2009
  #20
Gear maniac
 

Thread Starter
Quote:
Originally Posted by Ethan Winer View Post
The three sets of measurement is shown in that video are what's needed:

Frequency Response
Modal Ringing
Reverb decay time versus frequency (RT60)
--Ethan
Ok ethan, thanks. can you explain what the 'modal ringing' measurement is? (I use rew if that helps)

I did not bother with FR, for many reasons, but also because I use eq (deqx). In any case, you can see the FR from the waterfall.

(BTW, it would be good if you -or others- interpreted what you see. these are quick and dirty measurements)



and here is the T 30, does not seem to be rt 60 in rew??(????)




Hmm, do I need bass trapping???

hope that starts to help, thank you so much.





Quote:
Originally Posted by Lupo View Post
Hello!



Yes!

The important thing is to follow the correct barker code sequence for the length you're using. It's listed in the wiki page. Or any other code that gives you the desired autocorrelation properties.
Oh, I have missed that, I assume it is in this forum? will look, thanks for that, good data to know.

Quote:
I'd still go for a continuous large prime QRD if you can, over a compound one based on many smaller. The barker code will help alleviate the repeat problems, but a larger unit doesn't have them at all.
I would have thought that too. But this is the problem with the net really, all kinds of opinion.

What I mean is (making sure you did not think that I was casting aspersions at you!) I linked to the dutch site earlier (does not like being linked to I see) and that guy on another forum was adamant that it HELPS that the sequence is repeated.

At this stage, anyone with data on diffusion is an expert (to me with my lack of knowledge), so your point and Jason's point (with graphs) conflict with 'another experts' opinion.

So the main point is that these things are cleared up, without ruffling feathers. I am very much tending towards your opinion on this matter...and I hope this thread goes a long way in sorting these things.



Quote:
I think fractalistic design is more useful with 1D diffusers. With 2D using 2" well width, the design frequency extends to about 3500Hz. Above this limit, the scattering will still be significant. If it extends 1-2 octaves above the limit, like it does below the number theoretical limit, it'll nearly reach ultrasonic.
Funny how these things work. I asked that question as a 'what about...', thinking that there is nothing new under the sun really, so if it was not around already it must have been a silly idea.

then just tonight I found this from Room Acoustics

Golden Horn Diffusor panels: The GH is essentially a diffusor within a diffusor. Built from plastic, the GH contains a number of “bars” of seemingly random length, much like a traditional Skyline®-diffusor. The special thing with the GH is the fact that on the top of each “bar”, there is another irregular surface, with seemingly random “pits”, which makes the GH hugely effective over a very wide frequency range. The size of a GH-module is 55 x 55cm, and it is effective from 900 to 16000Hz. I used four on each side wall, giving very effective diffusion of first reflections.

So it sounds like it HAS been done!

Google Translate

Whether it is worthwhile or not is another question. (hmm, letter style just changed)

Quote:
Bass diffusion should be just as great as high end diffusion, if you can afford the loss of space. Make sure the material used is substantial enough to reflect the LF.
In the sonotube thread, alvares I think it was made a comment that diffusion was not useful under the schroeder frequency (?), still maybe worth a try at some stage eh? Will let you know if I do try that.



Quote:
Waiting in anticipation for the pic and explanation! =) Hehe.. Know the feeling!
Hopefully I can get some spare time this weekend, may as well make a useful product rather than just a test. And if I can work out how to get the date and time stamp working on my camera then we can see just how long it does take. Might make enough to end up with the opposing qrd directions in a diamond pattern, then I can try it behind me and see what I think.

But really, first things first, I think I need some serious bass trapping as the foundation for all this stuff.




Thanks for the write up of what you're doing. I gotta tell you that it is a welcome relief to talk about things that matter, I am sick of the usual audiophile garbage about changing cables etc.

As someone else said in this thread (paraphrasing) 'cables, ...sheesh'
Old 24th July 2009
  #21
Gear Guru
 
Ethan Winer's Avatar
 

Lightbulb

Quote:
Originally Posted by terry j View Post
can you explain what the 'modal ringing' measurement is?
It's the waterfall graph you already posted. heh

In this case the "mountains" come forward over time, showing for how long the ringing sustains. More here:

Modal Ringing

Quote:
here is the T 30, does not seem to be rt 60 in rew??(????)
RT30 is the same as RT60 except only the top 30 dB is measured. The other 30 dB is extrapolated, which is valid.

Quote:
Hmm, do I need bass trapping???
Every room needs bass trapping!

Quote:
your point and Jason's point (with graphs) conflict with 'another experts' opinion.
Without hearing that other expert's opinion it's impossible to respond. But you can trust me anyway. Most of the time. heh

--Ethan
Old 25th July 2009
  #22
Gear maniac
 

Thread Starter
Quote:
Originally Posted by Ethan Winer View Post
.....'content'...
--Ethan
Thanks ethan.

Ok, non chatting serious hat on.

"Every room needs bass trapping'' - accepted. Then why did you say measurements would help move the thread forward? I don't mind doing things like that, but only if it is useful and not a waste of bandwidth and peoples time.

I try and keep my threads a bit light hearted and fun, but I too dislike threads/posts that serve no purpose.

So let me critique as best I can what I see in the measurements, please grade me.

Waterfall: FR (as expected) looks very good in amplitude, due of course to the deqx. But the thing that I most notice is the very dense, discrete sharp bundling of tails/mountains.

So what does that tell us?

Presumably the broadband bass trapping will tend to fill in those ridges etc, and we should expect to see something more like a flat broad face in the sub 200 hz region. (?)

There is an obvious transition point in the RT60 graph at 200 hz. (I now wonder what the higher than 200 hz region used to look like before I hung all those batts on the wall). Above 200 hz it seems to me to be a 'perfect' response, but we need to get the sub 200 hz down to match.

Now the value of the above 200 hz RT60 looks to be between 0.3 and 0.4 ms, so call it 0.35 ms. Given that it is a largish room as far as most peoples room sizes go, is that a good value? In other words, what target would we aim for as a general rule, and how does my value compare?

Do we go for a different value in the sub 200 hz region or aim to keep it the same across the boards? (just trying to understand here) Example, let's say that we had an even but 'non-optimum' RT60 value, sonically would it be better to have an even non-optimum measurement across the entire range, or better to get the best we can in any range and sacrifice the evenness.

Obviously the best is even AND optimum values, but what are we after, evenness or absolute values? (hope that made sense, I know what I want to know, but wonder if I expressed it well)

Hope I get a B- at leats on that, but be honest. How else do we learn.

Again broadband bass trapping will do a lot to help the reverb time. (?)



No chat in this post, will ensure I get pictures of my qrd experiments up in the next day or two.

Apart from getting my score (and the pics posted), I guess this thread is on hiatus until I start moving on the bass traps. Thanks for the help so far, if anyone has more data in the diffusor questions already asked but not yet answered then that is fantastic too.

Pics soon.
Old 25th July 2009
  #23
Gear maniac
 

Thread Starter
ha! chatty again heh heh, look I thought I'd better make it clear in case I came across otherwise, I think it is fantastic that ethan and the boys come on and help us diyers!

So that was in case my serious non chatty side gave the wrong idea haha (he's a bit like that I've found)

obviously I have been thinking about my next step (serious bass trapping) and that again led me to the back corner alcove, where it is feasible to place a very large diffusor with the aim to diffuse down to about 150 hz or so.

Lupo reckons it's not a silly idea, but technically ARE there any limits or reasons to try to go that low?? Can I get that answer wrapped up if possible? That helps decide what goes there, bloody big hulkin diffusor or a bloody big hulkin bass trap.
Old 25th July 2009
  #24
Lives for gear
 
Nordenstam's Avatar
 

Quote:
Originally Posted by terry j View Post
Oh, I have missed that, I assume it is in this forum? will look, thanks for that, good data to know.
Sorry, no gearslutz wiki. There's a tips and techniques section but nothing on diffusers there AFAIK. Was thinking about the general wikipedia page on Barker code.


[repeating QRD's vs a larger unit]
Quote:
Originally Posted by terry j View Post
What I mean is (making sure you did not think that I was casting aspersions at you!) I linked to the dutch site earlier (does not like being linked to I see) and that guy on another forum was adamant that it HELPS that the sequence is repeated.

.. So the main point is that these things are cleared up, without ruffling feathers.
What dutch forum?

Agreed that it needs to be cleared up! There's a lot to be said for repetition too. Will check out what the Cox and D'Antonio book says on the subject when I'm back home over the weekend.

Quote:
Originally Posted by terry j View Post
Funny how these things work. I asked that question as a 'what about...', thinking that there is nothing new under the sun really, so if it was not around already it must have been a silly idea.

then just tonight I found this from Room Acoustics
The 2D fractal is neat! Though there are numerous things I could point out that could be done better. If you decide to do something like it, give me a word and I'll try to help you exceed the Swedish unit.

Quote:
Originally Posted by terry j View Post
In the sonotube thread, alvares I think it was made a comment that diffusion was not useful under the schroeder frequency (?), still maybe worth a try at some stage eh? Will let you know if I do try that.
That's odd. It'll break up the room modes. Or..?

Quote:
Originally Posted by terry j View Post
But really, first things first, I think I need some serious bass trapping as the foundation for all this stuff.
Yep! Quote Bob Katz in one of the GIK newsletters(?): bass is the final frontier.

Quote:
Originally Posted by terry j View Post
Thanks for the write up of what you're doing. I gotta tell you that it is a welcome relief to talk about things that matter, I am sick of the usual audiophile garbage about changing cables etc. As someone else said in this thread (paraphrasing) 'cables, ...sheesh'
Hehe.. Talking about cables can be a quick affair. Speaker cables: big enough. Interconnects: shielded twisted pair. Done! heh
Old 25th July 2009
  #25
Lives for gear
 
Nordenstam's Avatar
 

Quote:
Originally Posted by terry j View Post
Now the value of the above 200 hz RT60 looks to be between 0.3 and 0.4 ms, so call it 0.35 ms. Given that it is a largish room as far as most peoples room sizes go, is that a good value? In other words, what target would we aim for as a general rule, and how does my value compare?

Do we go for a different value in the sub 200 hz region or aim to keep it the same across the boards? (just trying to understand here) Example, let's say that we had an even but 'non-optimum' RT60 value, sonically would it be better to have an even non-optimum measurement across the entire range, or better to get the best we can in any range and sacrifice the evenness.
There's a graph in the master handbook of acoustics that looks something like this:


With a bump in the low end and a dip in the high end. While I think you can safely assume that some bass will boom more than the rest of the range, striving for a flat'ish RT in the high end is not bad goal. At least make sure it's not totally dead. Going for a flat low end can't hurt either, if it's possible! http://www.santafevisions.com/csf/im..._time_freq.gif

Though, I think there's something positive to having a dip in the high end. If the goal is to create an illusion of a room much bigger than what it really is, a dip in the high end RT is to expected from such a large room.

Quote:
Originally Posted by terry j View Post
obviously I have been thinking about my next step (serious bass trapping) and that again led me to the back corner alcove, where it is feasible to place a very large diffusor with the aim to diffuse down to about 150 hz or so.

Lupo reckons it's not a silly idea, but technically ARE there any limits or reasons to try to go that low?? Can I get that answer wrapped up if possible? That helps decide what goes there, bloody big hulkin diffusor or a bloody big hulkin bass trap.
I think you'll get better room performance with a big bass trap instead of a big diffuser. If you get the low end tight enough and still have room to build deep diffusers, they certainly won't hurt. But bass trapping should be your first priority, IMHO.

Disclaimer that goes with all these posts: I'm a DIY'er with a keen interest for acoustics. But I'm not even close to being an expert on the subject!



Cheers,

Andreas
Old 25th July 2009
  #26
Gear maniac
 

Thread Starter
hey cheers andreas! Will have a look at the wikipedia,

Quote:
Originally Posted by Lupo View Post

What dutch forum?

Agreed that it needs to be cleared up! There's a lot to be said for repetition too. Will check out what the Cox and D'Antonio book says on the subject when I'm back home over the weekend.
I should buy 'the book', boy is it pricey!! Also agree that these things need to be cleared up. Too many experts on the internet I reckon!

Ok which dutch site? this one

SoundScapeS Akoestiek Thuis -- Bondgenoten voor Reflectiebeheersing

I must have been confusing, the guy who runs/made that site says in another forum that repetition is a good thing

As I do not ever want to misrepresent someone, I feel I need to quote him and provide the link. That way we can sort this stuff out??

However there's an unwritten rule that states that when using diffusers, you must apply a minimum of 4 or 5 periods of the design (that is: 4 complete panels wide). This is the reason you sometimes want to use a narrow panel when space is restricted: better to use 4 narrow panels than 2 wider ones to cover the same surface.

and

Periodicity indeed is a condition for homogeneous diffusion. The minimum number to be used would be 4 or 5 periods (full sequences) for a given surface to be appropriately diffused; more is always better in order to suppress the 'lobing effect' of a single or double period even more.

And this section approaches what we are talking about here, a repetition of a small one vs a large single higher N number that fills the space

One other important aspect to consider when applying diffusion is periodicity: for homogenous results any diffusive surface should at least be covered by a width of 4, or better still, 5 periods. Therefore it's sometimes better to design narrow periods (lower N-numbers such as QRD-7 or 11) in order to be able to have diffusive surfaces consisting of 4 or 5 periods next to each other.
Periodicity brings homogenous diffusion over the targeted working range. Applying one wide panel (period) is not a good approach, not even when it's a QRD-19 with a wide working range. In those cases it's better to apply 5 lower numbered periods of a QRD-7.


That is enough for now, the thread this all came from is here

appropriate diffuser behind Apogee - AudioWorld's StraightWire Forums

As you can see, here are two completely different interpretations of the same phenomenon. Gut feel is to believe cox and d'antonio.....

As you say, it is best if internet 'folklore' is replaced by truth. So hopefully we can find out eh??

Quote:
The 2D fractal is neat! Though there are numerous things I could point out that could be done better. If you decide to do something like it, give me a word and I'll try to help you exceed the Swedish unit.
I give you the word, and point away!! (which word did you want????)



Quote:
That's odd. It'll break up the room modes. Or..?
again to avoid misrepresentation, here is the line I was talking about

Diffusers are not effective below the (take your pic: Davis, Shroeder #1, Schroeder #2, 3 modes/Hz) cutoff frequency of a room.


It was avares, and here is the original post

https://www.gearslutz.com/board/bass-...-diffusor.html

post 12

Quote:
Hehe.. Talking about cables can be a quick affair. Speaker cables: big enough. Interconnects: shielded twisted pair. Done! heh
Ahh, a man after my own heart!

Quote:
Originally Posted by Lupo View Post
There's a graph in the master handbook of acoustics that looks something like this:
Cheers, will have a re-look (it is ONE book I do have!)

Quote:
Though, I think there's something positive to having a dip in the high end. If the goal is to create an illusion of a room much bigger than what it really is, a dip in the high end RT is to expected from such a large room.
Ok then, if you are after a certain goal (like a drop in RT at the high end), HOW do you achieve that?

that is what I hope this thread will end up as, a documentation of the process to achieve certain goals.

Appreciate your input

On a side note, think I have managed to clear a few hours tomorrow, so will find out how many of those diffusors I can build in an afternoon with my new technique
Old 25th July 2009
  #27
Gear Guru
 
Ethan Winer's Avatar
 

Lightbulb

Quote:
Originally Posted by terry j View Post
why did you say measurements would help move the thread forward?
I never said that! You asked what measurements are useful, and I listed them.

Quote:
Waterfall: FR (as expected) looks very good in amplitude, due of course to the deqx. But the thing that I most notice is the very dense, discrete sharp bundling of tails/mountains. So what does that tell us?
Not much because I think you said you're using an equalizer in the playback path. In that case you can force any response at the one place you have the microphone, thus hiding what the room is doing. I suggest you measure again with no EQ. You could also try measuring again with the EQ, but in that case I suggest you move the microphone around and see what happens a foot to the left, a foot to the right, etc.

Quote:
what target would we aim for as a general rule, and how does my value compare?
Yeah, 350 ms is okay for a big room. But it needs to be much lower in the bass range than you have now. It's okay to have longer decays at the lowest frequencies, but that much disparity is a problem. Again, my Hearing is Believing video shows more or less ideal results to aim for.

--Ethan
Old 27th July 2009
  #28
Lives for gear
 
Nordenstam's Avatar
 

Hi!

The Book is worth it.

Regarding the repetition thing. Think it's best understood on a synthesesis basis. A small bit of the sequence is not random in itself, it does not have a flat fourier tranform. Only the total sequence works as intended at the lowest frequencies. Starting and stopping the sequence (waves striking the ends of the diffuser) makes a bit of a mess with the sound waves, those that strike both within and outside the diffuser. For the larger lower frequencies, diffusion only takes place with an "optimal effect in the middle" and broken up edges of the waveform on all other places. Repetition solves this problem. Since the sequence can be phase shifted in any direction, starting or stopping at some arbitary place in a repeating sequence is as good as having a long sequence. A repeating array will give this effect. But it introduces lobing problems in the far field scattering response. Breaking up the pattern with an inverted diffuser is good. Though I have a feeling using a different prime number for the modulation is better.

The diffusion coefficient graphs for QRD's looks like harmonics of instruments. Having different tunings seems like a good thing to me, making it more random across a larger range of frequencies. Or perhaps it makes it sound weird? Perhaps one need to tune the diffusers to each other, like musical instruments? Perhaps it would make sense to tune them to musically relevant frequencies? Hmmm... Anyone?

If the above was too confusing: the jist of it is that a single period only "works in the middle", a repeating pattern avoids this but it have lobing problems, with a modulated array being the best choice. Most of this can be found in the google book extract from The Book.

Though I still can't see why a say a modulated array of 4 prime 11 and 1 prime 13 diffusers would be any better than say a sequence based on prime 59. Anyone?


Quote:
Originally Posted by terry j View Post
Diffusers are not effective below the (take your pic: Davis, Shroeder #1, Schroeder #2, 3 modes/Hz) cutoff frequency of a room.[/I][/B]

It was avares, and here is the original post

https://www.gearslutz.com/board/bass-...-diffusor.html

post 12
Hmm.. I got a hunch it's because its unpractical to build a large enough diffuser. Haven't tried to calculate the size of these things, but by the time you get to seriously affect the room modes, the diffuser may be so large as to be impractical? Andre? Are you there?

Quote:
Originally Posted by terry j View Post
Ok then, if you are after a certain goal (like a drop in RT at the high end), HOW do you achieve that?
A drop in high end is easy. Absorption is most effective at higher frequencies and less so at lower. By the time you've got enough to seriously affect low end, high end is usually too dead. Adding back some high end response can be done by adding thin reflectors (paper, cardboard, wood etc depending on frequency) to the absorbers.
Old 27th July 2009
  #29
Gear maniac
 

Thread Starter
yet another stupid question time.

whilst I have been playing around making diffusors, I need and will get on to some serious bass trapping.

I intend to basically build super chunks, but I need to make them look good. Additionally, I'd prefer that they be free standing, well as much as possible, so that I do as little mechanical changes/fixing to the house as I can.

So I was intending to face the chunks with cloth wrapped compressed f/glass, self supporting (with a frame behind it) and looks good, have not worked it out just yet but maybe with a 600/700 mm face across the diagonal.

Now here is the silly question.

What if instead of facing the chunks with compressed f/glass, I put the chunks behind a qrd about 600/700 mm wide? I mean the bass will still 'get thru' and be trapped yeah?

Have not worked out the qrd dimensions either, but if say I could get the qrd to work down to 300-400 hz, then the trap takes over? Maybe with a face that wide, I can get the wells wide enough that I could add nested diffusors..the ones I am making now are 102 mm wide...ie fractal style.

Is that a goer? does it have any positive features? negative ones? Or just stick to broadband in as many corners as I can, then add diffusors later.

I would really love to get thoughts and opinions on this, the sooner I can evaluate all the data the sooner I can start.
Old 27th July 2009
  #30
Gear Guru
 
Glenn Kuras's Avatar
Quote:
What if instead of facing the chunks with compressed f/glass, I put the chunks behind a qrd about 600/700 mm wide? I mean the bass will still 'get thru' and be trapped yeah?
You would not put diffusion in corners.
Loading mentioned products ...
New Reply Submit Thread to Facebook Facebook  Submit Thread to Twitter Twitter  Submit Thread to LinkedIn LinkedIn  Submit Thread to Google+ Google+  Submit Thread to Reddit Reddit 
 
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get instant access to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
Divinitygeist / So many guitars, so little time!
3
The Magic Hoof / Studio building / acoustics
8
hakim / Low End Theory
3
PyroDano / Bass traps, acoustic panels, foam etc
7
andsonic / Music Computers
4

Forum Jump