Room modes, calculated versus measured

Posted: Mon Mar 21, 2005 2:09 pm

FWIW, i know that some folks who promote walls as bass traps suggest staggered studs.

as it were, staggered stud walls tend to exhibit resonance across a much broader frequency range than 2x4 walls. they exhibit the combination of the same resonance behavior as a 2x4 wall (but less severe), and a mass-spring behavior, and by serendipity these resonances sort of kind of nearly overlap across a reasonably broad range of frequencies ...

the same would hold for a double wood stud wall, but the lower spring resonance point spaces them out more.

wall absorption = panel trap absorption = resonance

at least unless you manage to build a wall with some amazingly low TL at low frequencies, and the common sealed reasonably well built 2x4 wall ain't it. Very Happy

Posted: Mon Mar 21, 2005 2:14 pm

there, i'm not trying to be a pain, folks, and if that is tangent #741 then i'll volunteer to be flogged, but i was just chatting about this with a theater designer and so my blood was brewing, so to speak, lol

i'll be still now

Posted: Mon Mar 21, 2005 2:16 pm

Brian...

This one looked somehow appropriate.

Posted: Mon Mar 21, 2005 2:22 pm

awesome! mind if i use that guy from time to time?

Posted: Mon Mar 21, 2005 2:23 pm

Brian,

Your thoughts are always welcome. Just be sure to number your cans o' worms using prime numbers! (Get with it already!!! Confused

)

Posted: Mon Mar 21, 2005 2:27 pm

Brian, it ain't mine, so I guess feel free!

Posted: Mon Mar 21, 2005 2:32 pm

ok, this is jeff's fault, he said i was welcome:

if i was you fellas, i wouldn't worry about this measured reality -vs- theoretical predictions much at all

speed of sound is going to be marred by various bodies and absorption things here and there

and even beyond that:

modes of even somewhat close spacing can interact to create weirdness, two modes can look as one, two modes can look as none, two modes can look as two. Three neighboring modes? four?

they can look like anything. in a well damped drywall panel, 3 modes spaced across a blooming octave can look like one super-damped mode, two oddly shaped modes, 3 low-damped modes, etc... when processed with a simple FFT

so if i was party #1, wondering why things didn't add up, i wouldn't worry about it

if i was party #2 trying to show that things did add up, i wouldn't be so quick to correlate a null at some frequency with a mode that lined up there when the situation is perversely complex at any given spot there

and if i was party #131 (prime) touting wall effects causing this shift, i wouldn't be so quick to look for a theory to explain what is just plain & simple a wildly complex situation. i'm not sold that things aren't going according to plan, so to speak. Twisted Evil

and regardless of what party i was, i'd only look at the fundamental axial modes - the modes that have the least neighboring modal activity.

AT LEAST UNLESS i wanted to spend 6 months measuring and analyzing one single room.

see, modal analysis for other things, like various industrial structures... things where something is going to be mass produced, and where modal behavior is important, get more man*time of attention that you might ever think to make sure the situation is understood.

Posted: Mon Mar 21, 2005 2:37 pm

I accept full blame for inviting you here to pontificate with your unique flavor of creativity, Brian.

FWIW, I also tend to agree with 97% (prime) of what you write, including the above. Cool

Posted: Mon Mar 21, 2005 2:39 pm

that's awesome, jeff, as i happen to disagree with 3% (prime) of what i write

Posted: Mon Mar 21, 2005 8:33 pm

Ethan, you may find it useful to revisit Floyd Toole's article "Getting the Bass Right" (http://www.harman.com/wp/pdf/Loudspeakers&RoomsPt3.pdf) where he addressed some of your main questions in this thread head on. For example, he says on p 26:

"If a room surface absorbs sound, it will do so differently at different frequencies. It is the nature of such surfaces to slightly alter the timing of the reflected sounds (phase shift) depending on the frequency of the sound relative to the preferred frequency for the absorbing surface....This explains why, very often, it will be found that the measured resonances do not exactly correspond with the calculated ones."

Pardon me if this reference was already brought up before in this thread, which is so long and diverse I just can't remember any more.

Lee

Posted: Tue Mar 22, 2005 1:49 am

Edited:

I did something wrong here.
Deleted it.
Correct message below

Posted: Tue Mar 22, 2005 2:16 am

Brian R wrote:

ok, this is jeff's fault, he said i was welcome:

and regardless of what party i was, i'd only look at the fundamental axial modes - the modes that have the least neighboring modal activity.

AT LEAST UNLESS i wanted to spend 6 months measuring and analyzing one single room.

Brian,

Have you in fact read this thread?
I even can't see: Is your answer related to the boundary impedance or room acoustics? (I know they interact, that's not the question)
Or did you just want to show how good you are?
Humor is good.
Being clever is good.
That also calls for some responsibility.
Not just throw in some general statements.

One has to look at modes which clearly appear whether they are axial or non-axial.
In room acoustics one can only apply edge damping to solve problems to go in your terms, one can NOT damp a room as panel where one can glue 2 panels completely together with a damping material from the point of view: it doesn't matter the highest damping factor is the best.

One NEEDS to know what one is dealing with.
And don't start about chaotic behavior.
This thread goes about those things which clearly show as individual resonances.

And were they are closely spaced, it doesn't matter, because that's what indeed all those theories about modal overlap and density and Schroeder is about.

So can you please be careful and humble enough that your panel measurements and room acoustics while having lots of similarities still have some own specific goals?

If resonant peaks or dips in rooms are measured one just can't fill up the whole room with wool you know.

Or if you just want to transform this thread into some fun it's OK, but please describe it as such.

When, where, what, should you only look at fundamental axials?
What an incredible superficial answer.
Or did you just like the sound of it?
Are you talking about panels, room acoustics as covered by this thread, Boundary imedance? Or just talking??????
Is this a new definition of room acoustics: Look at the fundamental axials, and you're done, the rest is chaotic anyhow?

Can you believe that there are people taking this thread serious and putting a lot of energy in it?
And it starts over, and over, and over, and over, and over again.

If peaks or dips appear disrupting the room acoustics one has to look and try to identify those problems.

We are NOT busy with panels. And we can't glue rooms together here.
This thread is about Room acoustics.

Posted: Tue Mar 22, 2005 3:28 am

z60611 wrote:

Ethan

Ethan Winer wrote:

Sorry for the confusion. That particular graph is from a different room!

Opps

But it looked good (had also nulls and peaks), was the closest picture available. But I """"forgot"""" to tell.

Ethan,
I spent time to find logic in this deviating data. I even digitized it to bring it in one graph.

Posted: Tue Mar 22, 2005 6:03 am

eric_desart wrote:

Brian R wrote:

ok, this is jeff's fault, he said i was welcome:

and regardless of what party i was, i'd only look at the fundamental axial modes - the modes that have the least neighboring modal activity.

AT LEAST UNLESS i wanted to spend 6 months measuring and analyzing one single room.

Brian,

Have you in fact read this thread?
I even can't see: Is your answer related to the boundary impedance or room acoustics? (I know they interact, that's not the question)
Or did you just want to show how good you are?
Humor is good.
Being clever is good.
That also calls for some responsibility.
Not just throw in some general statements.

One has to look at modes which clearly appear whether they are axial or non-axial.
In room acoustics one can only apply edge damping to solve problems to go in your terms, one can NOT damp a room as panel where one can glue 2 panels completely together with a damping material from the point of view: it doesn't matter the highest damping factor is the best.

One NEEDS to know what one is dealing with.
And don't start about chaotic behavior.
This thread goes about those things which clearly show as individual resonances.

And were they are closely spaced, it doesn't matter, because that's what indeed all those theories about modal overlap and density and Schroeder is about.

So can you please be careful and humble enough that your panel measurements and room acoustics while having lots of similarities still have some own specific goals?

If resonant peaks or dips in rooms are measured one just can't fill up the whole room with wool you know.

Or if you just want to transform this thread into some fun it's OK, but please describe it as such.

When, where, what, should you only look at fundamental axials?
What an incredible superficial answer.
Or did you just like the sound of it?
Are you talking about panels, room acoustics as covered by this thread, Boundary imedance? Or just talking??????
Is this a new definition of room acoustics: Look at the fundamental axials, and you're done, the rest is chaotic anyhow?

Can you believe that there are people taking this thread serious and putting a lot of energy in it?
And it starts over, and over, and over, and over, and over again.

If peaks or dips appear disrupting the room acoustics one has to look and try to identify those problems.

We are NOT busy with panels. And we can't glue rooms together here.
This thread is about Room acoustics.

Hi Eric,

my point and my analogy aren't bad in this case, although it's regrettable that you didn't enjoy the posts or perhaps that i didn't communicate adequately what i was trying to say. or perhaps you misunderstand.

i express the sentiment that 3 modes close in frequency (applicable to basically every frequency region above the fundamental axials in the graphs presented in this thread) might add/subtract to create a wide variety of things.

FWIW, as i mentioned above, Terry Montlick illustrates this here: http://archive.avsforum.com/avs-vb/showthread.php?postid=3653594#post3653594 , as bob linked to above.

hence my comment about fundamental axial modes has this basis: they are the only modes that aren't considerably overlapped in the graphs in this thread (or in room calcs modes2vp gives for various hypothetical rooms)

and as such, i think it's something of a stretch to point at a dip or peak in any given measurement and say "this is due to this mode, which is aligned with it" without a considerably deeper look into the situation (the six months).

3 room modes within a hz or two or four... do we know they are all 3 nulled at this point in the room, hence explaining the dip in the graph? or do we know that they are all 3 peaked? or that their sum should create this null at this given point in this given room with this given speaker location, etc.?

if you respond "yes" to that, i'd ask that you substantiate it and not just say "yes" and let your expertise in this area be the proof.

You systematically indict every comment i enter above, lol, and i won't reply to each point. But perhaps if you opt to post again in this exchange, you could go lighter in the personal jive, and heavier on the counterpoint. Would make for a more productive chat, i think.

Brian

Posted: Tue Mar 22, 2005 6:06 am

so to sum what always wind up overly wordish replies to these sort of things:

here is my overly verbose response in a nutshell:

i do not think that it is prudent to state offhand that a peak is realted to mode X just because they line up on a graph, and i do not think that it is prudent to say a null is related to modes K and Y and Z because of the same. And the reason is that the interactions of modes (acoustic or mechanical) can get complex when they overlap in frequency. Also, any given mode might measure flat somewhere in the room.... making it seem not there, and so on.

This topic sponsored by: