Acoustics of corners

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Some acoustics :
we know that corners are places where all low frequency modes are present with max pressure and no velocity.
For this reason, many people (and many books) say that basstraps are preferably placed in corners. This is also often said for fibrous materials (despite no velocity in corners). For sure, at opposite corner, the effect will be heard. But at listening place ?
So let's imagine a room with holes at corners (a hole gives total absorption): with or without this hole LF modes will still be present, the effect of an axial mode at listener point is due to walls and not corners and should barely be changed.
Am I missing something ?
 
Some acoustics :
we know that corners are places where all low frequency modes are present with max pressure and no velocity.
For this reason, many people (and many books) say that basstraps are preferably placed in corners. This is also often said for fibrous materials (despite no velocity in corners). For sure, at opposite corner, the effect will be heard. But at listening place ?
So let's imagine a room with holes at corners (a hole gives total absorption): with or without this hole LF modes will still be present, the effect of an axial mode at listener point is due to walls and not corners and should barely be changed.
Am I missing something ?

A hole will have some mass, and this hole will form a Helmholtz resonator with the room. This could be a boost or a loss depending on the details. If the hole is well damped and several are used (the effect on the room will depend on the total area of the holes) of different sizes and locations then you could achieve some broadband absorption inside of the room. Of course these holes have to go somewhere and on the other side of the hole the sound level goes way up. Hence such a technique does not work in a room meant to maximize sound isolation (like mine).

I have however, used various in-room features (no leakage to the outside) to make Helmholtz resonators as LF absorbers - like a seat riser for example.

By far the best solution to LF absorption is damped and hung walls. As the walls flex they become strong absorbers. In my room there are basically no modes evident above the very lowest axial one - its along the length of the room and has the least wall area as an absorber.
 
Refer to Ethan Winer for a source. He has written numerous articles on room acoustics and attenuation.
But this is typically the litterature I refered to in my introducing post : it says "place basstraps in corners" without any justification.


My hole example was not well understood. So I'll try another way : imagine no room, and just a front wall and a parallele rear wall (in empty space, no floor, no ceiling, no side walls). It is like having infinite corners fully absorbant (no reflection at all). Despite this, there will be axial modes between the two remaining walls, front and rear even with 100% losses of all others modes. Those axial modes can only be damped with the structure and absorption of the two walls. This simply means that the corner can have practically no inluence on the axial modes seen from a centered listening position.
 
Last edited:
But this is typically the litterature I refered to in my introducing post : it says "place basstraps in corners" without any justification.
In theory a corner is the worst place to put an absorber.

My hole example was not well understood. So I'll try another way : imagine no room, and just a front wall and a parallele rear wall (in empty space, no floor, no ceiling, no side walls). It is like having infinite corners fully absorbant (no reflection at all). Despite this, there will be axial modes between the two remaining walls, front and rear even with 100% losses of all others modes. Those axial modes can only be damped with the structure and absorption of the two walls. This simply means that the corner can have practically no inluence on the axial modes seen from a centered listening position.

There is always some leakage of modes, something that one finds in exact solutions of room acoustics, but in the simpler models it doesn't exist. So this axial mode will dissipate to the sides and top in your example. It takes perfectly reflecting side walls to keep the axial modes axial. And absorption on the side walls will dissipate an axial mode, but it takes some serious math to show this. Way beyond the simple modal model usually used.
 
But this is typically the litterature I refered to in my introducing post : it says "place basstraps in corners" without any justification.
This simply means that the corner can have practically no inluence on the axial modes seen from a centered listening position.

I'm SO disappointed to hear that the corner isn't the best place for absorption. I'll have to take mine down. Before doing so, let me note that they do appear to work. I have (somewhere) before and after APL_TDA measurements that show their effect at my central LP.

How can this be? Well in the corner they cover some of the front and side walls and ceiling. The corners are also a very convenient and visually benign location and going across the corner gets the absorber out to where there is some velocity. Of course they can't compete with turning the entire wall(s) into absorbers as does Gedlee.
 
Thanks Earl, I was suspecting something like this. But is there any room acoustics litterature on such a subject ? Morse ?

There are three approaches to modeling a room so let me elaborate here:

First - by far the most common approach can be found in almost any text on acoustics. It models the room as a summation over the modes with a given source and receiver location. How this simpler approach handles damping is its limitation. In order to get a simple equation, the damping is assumed to be uniform allowing for a single damping term in the denominator and modes that don;t change with the damping. This is fine for rooms with very low damping, but it yields some unreal results. The energy flow in the room will be entirely circular with no energy loss at the boundaries, its as if all the loss is from the air. This is unrealistic since in real room virtually all energy loss is at the boundaries.

Second - in Morse and Ingard one can find a rough outline of how the room modes would change with boundary absorption. The exact solution is not worked out or shown. Some time ago I did work out the solution and programmed this model. I did not do much with it however and I still plan to write a paper comparing the two methods to see how much difference there is. In this model any value of boundary absorption, real and/or complex is allowed and the mode shapes will change depending on this value. The limitations here are that the impedance must be over the entire surface of the wall. It can have any value, but it must be uniform. One cannot have "patches" of absorption and the modes must remain orthogonal, by this I mean that there cannot be coupling between the modes. This kind of thing would occur fro example if the side walls had very large absorption and the end walls had none. The axial mode would be dissipated by the side walls as it progressed and this would leak some energy from the axial modes to the side wall modes. Because in this case the modes are no longer orthogonal standard math techniques don;t work - dead end.

Third - a finite element model can be done that accounts for everything that each of the above models cannot. However, the model size in FEA would get very large if one wanted to go to higher frequencies. Ray tracing techniques are better suited to HF modeling, FEA to LF modeling.

Absorbers in corners do work, its just not the most efficient place to put them. Hanging a dead wall over an existing wall is not too hard and much more effective. Takes up almost zero space. Not as easy as just buying traps and placing them in corners.
 
Porous absorbers need to be placed where air velocity is high, and this is not against the wall or in the corner, but at a reasonable distance from these places. Of course it's not practical - for using the room - to place them in the way so compromises are made.

On the other hand, resonant absorbers (like membrane and slit resonators) are working with pressure and they are best placed against the boundaries or even flush with them.

In reality, it's the fact that the corners are considered useless space and stacking triangles of porous sheets floor to ceiling against the corners is easy and practical. Just not effective as the effective elements are the ones far from the corners.

In reality, you can place a porous panel leaning against the corner with empty space behind it and it will perform as good as a "solid" stack. And if that panel is moved one or two "thickness" distance away from the corner or the wall, it will perform better. And intrude into the room of course.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.