Axial Mode Spacing

[watchnerd]

I was using this calculator to try to get a sense of my room acoustics:

http://www.acoustic.ua/forms/room-modes-eng.html

I get this table, which I think I understand:

But then I read this footnote:

And have this bigger table:

Which I don't really understand very well.

My interpretation is:

• I will have a dip in the in-room response curve between 77 Hz and 99 Hz, and another between 232 Hz and 257 Hz.
• The gray area of the table, which covers 99 hz to 165 Hz, will have choppy peaks and dips.

Do I have it all wrong? Am I missing something critical?

 

[RayDunzl]

Am I missing something critical?

Is your new room a nice tight recta-cube?

 

[watchnerd]

Is your new room a nice tight recta-cube?

No, not really.

The ceiling is high, but slopes upward from about 10' to 18'.

There are multiple openings, a fireplace that juts into the middle.

 

[Fitzcaraldo215]

No, not really.

The ceiling is high, but slopes upward from about 10' to 18'.

There are multiple openings, a fireplace that juts into the middle.

So, isn't the calculated table therefore totally pointless?

But, even if you had a nice perfectly rectangular room without openings or obstructions, a calculator could only give you an estimate, since specific materials, floor springiness, etc. etc. are not part of the simple minded, idealized calculation. Actual acoustic measurements, including the furnishings which can also make a difference, are the only things worth doing, in my opinion.

A few years ago, I think it was Linn that brought out a DSP room correction tool that avoided the need to take those messy and troublesome measurements. Instead, you typed in room dimensions and a few other details, and, voila, it would calculate some nice correction filters for you. Sooo much better than all those other clumsy tools that required mike measurements. But, too bad if your room was not rectangular, had carpets/rugs on the floor, openings, windows, pictures, etc. What a joke.

 

[watchnerd]

So, isn't the calculated table therefore totally pointless?

But, even if you had a nice perfectly rectangular room without openings or obstructions, a calculator could only give you an estimate, since specific materials, floor springiness, etc. etc. are not part of the simple minded, idealized calculation. Actual acoustic measurements, including the furnishings which can also make a difference, are the only things worth doing, in my opinion.

A few years ago, I think it was Linn that brought out a DSP room correction tool that avoided the need to take those messy and troublesome measurements. Instead, you typed in room dimensions and a few other details, and, voila, it would calculate some nice correction filters for you. Sooo much better than all those other clumsy tools that required mike measurements. But, too bad if your room was not rectangular, had carpets/rugs on the floor, openings, windows, pictures, etc. What a joke.

Okay, theory vs reality aside, if, hypothetically, I did have a perfectly rectangular room with average absorption and reflectivity, what would the calculated results mean?

 

[Blumlein 88]

Hey REW also has a room mode simulator function. If you do have a closed rectangular room it works well enough in the lower frequencies. Not perfect due to absorption, reflection and springiness never been exactly known. Enough to sometimes be helpful about where to put speakers (especially subs) that might not have struck you as places they would work well.

 

[Fitzcaraldo215]

Okay, theory vs reality aside, if, hypothetically, I did have a perfectly rectangular room with average absorption and reflectivity, what would the calculated results mean?

In the table you provided, it is a mystery. I cannot figure out what they are trying to say. I have seen calculators that do a better job at estimating modal behavior in terms of problem frequencies, including the modes generated non-axially. I just have none handy, because I never use one.

 

[watchnerd]

In the table you provided, it is a mystery. I cannot figure out what they are trying to say. I have seen calculators that do a better job at estimating modal behavior in terms of problem frequencies, including the modes generated non-axially. I just have none handy, because I never use one.

Okay, glad I'm not the only one who is perplexed by their presented results.

 

[watchnerd]

Hey REW also has a room mode simulator function. If you do have a closed rectangular room it works well enough in the lower frequencies. Not perfect due to absorption, reflection and springiness never been exactly known. Enough to sometimes be helpful about where to put speakers (especially subs) that might not have struck you as places they would work well.

Do you have insights into the tables above?

What do you get if you use the calculator and does it compared with what you know about your own room?

 

[March Audio]

Seriously, just try REW. It is surprisingly accurate for simple shaped rooms. You can move speaker and listener locations and see the effect

 

[watchnerd]

Seriously, just try REW. It is surprisingly accurate for simple shaped rooms. You can move speaker and listener locations and see the effect

Thanks, nothing against REW, but I want to understand the theory behind the calculator.

 

[Blumlein 88]

Do you have insights into the tables above?

What do you get if you use the calculator and does it compared with what you know about your own room?

It seems like a poor interface without giving predicted results of the various modes in terms of levels. I mean so two modes are 20 hz or more apart, and therefore will have interactions so that they combine to form a wide enough dip to matter. If closer then they never interfere destructively over a wide enough band to be heard as a problem. Or so I gather reading it. But where? In the middle of the room, along each edge or what?

You could do a spreadsheet that took rectangular room dimensions and calculated modes along the 3 axis, and even combine them for different locations across the room. Modes that cause a deep null dead center of the room will cause peaks a few feet away in the same room. The value of the table like above might be to pick room dimensions with very few modes that would interact over a wider band of 20 hz or so. It won't be incredibly useful if you aren't able to change the room.

Parroting an earlier post, seriously, I mean SERIOUSLY, give REW a try. It is useful, and you still can puzzle out why it works like it does.

If you are just trying to get it into your mindspace how it works, think about sound on an open flat plane (or plain as in Kansas flatlands). Think about reflections if you added one wall behind your speakers. Now think about modes if you add a second wall behind the listener. Think of how that will constructively and destructively interfere. Then add two sidewalls and add a ceiling. Of course the REW Room Sim lets you do this and provides you with instant graphics of what the result is. You can look at one mode at a time and select which one. It even allows tangential and oblique modes. All you have to do is set all walls to 100% absorption except the one or pair of walls you wish to understand. Sure it is a shortcut having it all done so easily for us. Sometimes that keeps you from getting lost in the weeds when you really want to think about room modes.

 

[March Audio]

Thanks, nothing against REW, but I want to understand the theory behind the calculator.

Well, there isnt a lot to understand really. The walls in your room reflect sound. The room dimensions dictate what frequencies (wavelengths) and positions in the room where the reflected sound wave constructively adds and destructively cancels. Absorption will affect the amplitudes of the nodes/antinodes. There are lots of calculators on the web for this. REW gives a very clear picture of whats going on.

I like th3d pressure zones on this calculator. Its purty.

https://amcoustics.com/tools/amroc?l=600&w=450&h=270&re=EBU listening room

Some random videos.

 

[DonH56]

I too find their presentation a little hard to follow and the spacing and cancellation depends on frequency and room size as well.

Info about room modes here: http://www.linkwitzlab.com/rooms.htm The equations in that web article are the same as found in any acoustics text (Everest, Kinsler, etc.) I have implemented them in Mathcad and Matlab.

f = ( c / 2 ) [ ( l / L )^2 + ( w / W )^2 + ( h / H )^2 ]^(1/2) [Hz] (5)

l, w, h = 0, 1, 2, 3 etc.
L, W, H = room dimensions (in feet, or same units as c)
c = speed of sound (1127 ft/s)

There are a myriad of other online articles such as http://www.acousticfrontiers.com/room-modes-101/ and many online calculators. A good one is from Harmon: http://www.harman.com/room-mode-calculator but there are others e.g. on Bob Golds site (http://www.bobgolds.com/Mode/RoomModes.htm) or Ethan Winer's site (http://realtraps.com/modecalc.htm).

HTH - Don

 

[watchnerd]

I too find their presentation a little hard to follow and the spacing and cancellation depends on frequency and room size as well.

Info about room modes here: http://www.linkwitzlab.com/rooms.htm The equations in that web article are the same as found in any acoustics text (Everest, Kinsler, etc.) I have implemented them in Mathcad and Matlab.

f = ( c / 2 ) [ ( l / L )2 + ( w / W )2 + ( h / H )2 ]1/2 [Hz] (5)

l, w, h = 0, 1, 2, 3 etc.

There are a myriad of other online articles such as http://www.acousticfrontiers.com/room-modes-101/ and many online calculators. A good one is from Harmon: http://www.harman.com/room-mode-calculator but there are others e.g. on Bob Golds site (http://www.bobgolds.com/Mode/RoomModes.htm) or Ethan Winer's site (http://realtraps.com/modecalc.htm).

HTH - Don

Thanks the equation is helpful.

 

[DonH56]

Thanks, but I just noticed it did not copy clearly, see edited version (or grab straight from the website or another reference).

 

[watchnerd]

Thanks, but I just noticed it did not copy clearly, see edited version (or grab straight from the website or another reference).

Yes, I got it from the website, thanks. Equations never seem to copy/paste very well.

 

[DonH56]

I was a beta tester for COMSOL at one point (boss knew the principals); really wish I had kept in touch as that would be a really cool tool to use for room modeling when the rooms aren't symmetric (mine is not).

 

[Brad]

Quite a bit if FDTD modelling of rooms in the literature too. Getting appropriate complex reflection coefficients is a challenge though.
COMSOL has examples of speaker models as well.

 

[watchnerd]

Quite a bit if FDTD modelling of rooms in the literature too. Getting appropriate complex reflection coefficients is a challenge though.
COMSOL has examples of speaker models as well.

Really? I've never seen / heard of those.