Nonparallel walls

Posted: Sat Aug 30, 2008 6:21 am

I know that minimizing the number of parallel flat surfaces is a typical trick in studio construction.

1. Can someone give me an explanation of what this does and why it works? (I know there was some discussion of this in Rod's book, but I've lent my copy to someone.)

2. Can someone give me a reference to a mathematical analysis of this phenomenon?

Posted: Sat Aug 30, 2008 6:39 am

Non-Parallel walls are useful to create a Reflection Free Zone but they are not mandatory. You can build a professional studio with parallel walls (of course by having proper acoustical treatment).

Search for Reflection Free Zone for a start. The angle of the walls depends if your monitors are soffit or not, size of room. You can use typical geometry to check this.

Posted: Sat Aug 30, 2008 6:57 am

andrebrito wrote:

Non-Parallel walls are useful to create a Reflection Free Zone....

I don't think this is correct. Any smooth wall will have at least one point of reflection between the monitor and the listener. It is possible to make curved walls with multiple reflection points. In fact, you can construct an elliptical wall where every point is a reflective point.

Posted: Sat Aug 30, 2008 7:58 am

Of course, but you are supposing a loudspeaker radiates sound with the same directionally where in fact it does not at medium and high frequencies. So in terms of pure mathematics it seems the same but in reality it is not. Keep in mind that RFZ rooms do have acoustical treatment as well.

I forgot to say that RFZ also acts in the time domain, trying to delay the first reflections at the listening point

Posted: Sat Aug 30, 2008 8:02 am

They can help stop flutter echo - and (if designed correctly) WILL stop first reflections........

to quickly prove your assumption that any smooth wall will have a first reflection - lets begin with this......

you have a wall 15' in width........ you sit dead center of the wall and 10' from it's face.

you have speakers sitting 2' from the wall and placed in an equilateral triangle in relation to the listeners position -

now - to the left and right of you - place walls that intersect the original walls - and place them so that they create a 170 degree angle in relation to the original wall.

I can promise you that no matter how long those walls are you will never see a reflection of your speakers on those walls.......

OK _ so you begin closing up the angle - and we know that with an interior angle of 90 degrees you will see the reflection - and at 170 degrees you won't - somewhere in between the 2 is the point where the speakers dissapear........ and that is what you can calculate with geometry.......

Posted: Sat Aug 30, 2008 8:05 am

Here's a picture !

Posted: Sat Aug 30, 2008 8:09 am

Non-parallel walls cut down on the ping-pong effect from direct reflections and helps minimize standing waves.

Stand in a square room and clap your hands - that reflection you hear is what you want to avoid in a studio space.

A non-parallel space reflects the waves at offset angles, delaying the reflections from each other to a certain degree, making them less obtrusive and lower in amplitude when they enter a mic capsule.

Posted: Sat Aug 30, 2008 8:11 am

Looking at Andre's pic you can probably figure out the math involved...

Posted: Sat Aug 30, 2008 8:18 am

Quote:

Non-parallel walls cut down on the ping-pong effect from direct reflections and helps minimize standing waves.

Non-parallels walls shift standing waves but they still exist. Newell's book has a nice picture to confirm this.

About flutter echo, I'm not sure if we have a room with non-parallel walls and non-parallel ceiling BUT the room is highly reflective if the "ping-pong" effect really disappears. This could be tested easily in a reverberation lab room but I can't recall checking this the only time I was inside one of those rooms. duh !

If you non-parallel walls but a parallel reflective ceiling/floor situation you will still have flutter echo.

Posted: Sat Aug 30, 2008 8:33 am

One would hope that the ceiling was taken into consideration along with the rest of the build...

Posted: Sat Aug 30, 2008 11:39 am

It could be, they built the studio and thought...

"Why is it wet and windy in here?"
"Oh ****, we forgot the ceiling!"

Posted: Sat Aug 30, 2008 6:21 pm

Thanks for the picture, Andre. I had read you were supposed to treat the speakers as omnidirectional radiators when treating for first reflections (for instance putting treatment behind the speakers) but your picture probably gives a more accurate model for what is going on.

Rod, I see what you are saying about my claim, and it is certainly incorrect as I stated it. What I meant to say is that I thought I had a proof that (treating the speaker as an omni source) you had to have a reflection point on at least one wall in any convex polygonal room. (For instance in your example there is a reflection point on the wall behind the speaker.) However, there is a hole in my proof, and I'm not sure the statement is right. Moreover, you can certainly avoid reflections from the side walls, so if that's the goal the theorem doesn't really matter.

Of course, the flutter echo issue certainly makes a lot of sense.

What I was really fishing for was to see if there was some reason based on the resonant frequencies and mode shapes. (More fodder for my PDE class.)

Posted: Sat Aug 30, 2008 6:40 pm

BobRogers wrote:

What I was really fishing for was to see if there was some reason based on the resonant frequencies and mode shapes. (More fodder for my PDE class.)

No. The only significant, easily predicted effect of splayed surfaces on modes is that they reduce/remove axial modes, so that the highest strength modes are reduced. The tangental and oblique modes are in not easily calculated patterns then.

Andre

Posted: Sun Aug 31, 2008 3:26 am

You can use the wave equation to find mode ressonances, shapes, etc etc...

Get the book Room Acoustics by Kuttruff. This is a great starting point in terms of math. This is not a book for small room acoustics, it is in fact a generic room acoustics book with a larger emphasis on large room acoustics.

You can also find tons of Info on JASA papers. There is also another book "Formulas of Acoustics", which I have never read but it has an interesting name Smile

Posted: Sun Aug 31, 2008 7:20 am

Bob,

reality is that the splaying of walls makes it much more difficult to predict modal activity - although (in theory) it should tend to smooth out axial modes in relation to the splayed walls (although it does not remove axial modes with relation to those walls).

This is because axial modes do not just exist in relation to parallel walls (although they do take place between 2 oppposing wall surfaces) - and thus you will tend to develope more modes with less energy in each mode in relation to those walls.

This might (or might not) be a good thing depending on the room. And still you have to take into account that if any given dimension in that splayed wall assembly is equal to any other dimension between 2 parallel surface within the room - you will still have some reinforcement of that particular axial mode due to that point.

As I noted - it may have it's benefits (which will typically grow smaller as the rooms becomes smaller) but predicting the room does become that much more difficult in a lot of respects.

Sincrely,

Rod

This topic sponsored by: