Nulls

[audioguy]

There are spreadsheets available that will allow you to predict room modes based upon room dimensions. Anyone know of something that will allow one to predict nulls?

I have one (a null) that is VERY detrimental to the sound in my room and am trying to determine from which dimensions it is generated (the null is not in the bass region -- it is located in the sweet spot of between about 200 hz and 350 hz). I'm pretty sure I know but am looking for confirmation.

 

[dougsmith]

This one does:
http://www.harman.com/EN-US/OurComp...ocuments/Calculators/Room Mode Calculator.xls

 

[DonH50]

I have a Mathcad program and can run the numbers for you and email (or post here) a table if you wish -- I need the room dimensions.

Ethan's website (www.realtraps.com) also has a modal tool on it.

 

[RUR]

Don, does your program handle rooms which are not strictly rectangular? My room has a small ell at one end for the entryway/bath entrance.

 

[audioguy]

This one does:
http://www.harman.com/EN-US/OurComp...ocuments/Calculators/Room Mode Calculator.xls

II guess I don't understand it. I cannot determine based upon entering room dimensions, where the nulls are located! Modes yes. Nulls. No. Maybe I'm not reading it correctly.

 

[audioguy]

Don--thanks for the offer.

Length is 22.5 feet. Width is 17.5 feet and height is somewhere between 9 feet and 9.75 feet. The ambiguity of the hieght dimension is that we have angled pieces between the ceiling joists that run from the bottom of the joist to the top so I'm not sure what dimension the various frequencies "see". The ceiling joists are covered with fabric.

Any help you can provide will be greatly appreciated.

 

[dougsmith]

II guess I don't understand it. I cannot determine based upon entering room dimensions, where the nulls are located! Modes yes. Nulls. No. Maybe I'm not reading it correctly.

The curves on the "Waves" page indicate the sound level across the dimension of the room resulting form the modes at various frequencies. Presumably you have a primary or secondary mode at the frequency of your null in both the length and width dimensions, and one of those null points corresponds to the location of your sweet spot. I'm not sure if I am explaining this very well, but that is the way I understand it.

 

[DonH50]

My program only handles rectangular, sorry... I have some angles in my listening room and entertained the idea of making the program more general, but it got ugly quick and I ran out of time. Below are the first 12 modal frequencies for each dimension of your room using 22.5 x 17.5 x 9.5 ft^3. The program does not indicate peaks or nulls but it's usually obvious from SPL measurements.

L W H
25.111 32.286 59.474
50.222 64.571 118.947
75.333 96.857 178.421
100.444 129.143 237.895
125.556 161.429 297.368
150.667 193.714 356.842
175.778 226 416.316
200.889 258.286 475.789
226 290.571 535.263
251.111 322.857 594.737
276.222 355.143 654.211
301.333 387.429 713.684

HTH - Don

Aaarrgghhh! The table looks fine in the preview, but all the tabs go away when I save it, and when I used spaces instead the same thing happens. Anybody know how to format the table so it stays formatted?

 

[amirm]

  L       W        H
25.111	32.286	59.474
50.222	64.571	118.947
75.333	96.857	178.421
100.444	129.143	237.895
125.556	161.429	297.368
150.667	193.714	356.842
175.778	226	416.316
200.889	258.286	475.789
226	290.571	535.263
251.111	322.857	594.737
276.222	355.143	654.211
301.333	387.429	713.684

How is that?

 

[audioguy]

The program does not indicate peaks or nulls but it's usually obvious from SPL measurements.

Don:

Thanks for running the numbers. The excel spreadsheet I have shows these numbers as well. The one I have also shows where all of the peaks might be which is what you can get from the data you provided. And yes, an SPL meter will show the null (as does my TacT) but I am trying to determine which dimension(s) are causing it.

 

[DonH50]

Amir -- great, thanks! So I just need to use the "code" tag?

audioguy: I thought my program showed all the modal frequencies; are you saying your null is not among them? I indicated L (22.5'), W (17.5') and H (9.5') in the notes, and made sure I grabbed enough to cover up to 300 Hz.

 

[Kal Rubinson]

If one wants/needs to model non-rectangular rooms with both graphic and time-correlated views of peaks and nulls (both modes), there is CARA (http://stereophile.com/miscellaneous/438/) . It is not plug-and-play as one has to first enter lots of parameters into its CAD-like database but it works. For me, however, it is easier to just measure the room. ;-)

Kal

 

[Nyal Mellor]

It is probably phase interference related:

If the difference between the direct and indirect path lengths (where path length is the distance the sound has to travel) is equal to half a wavelength, then the two sound waves will combine destructively and a notch in the frequency response will occur. The frequency that interference occurs at can be calculated through application of the wavelength formula as follows: cancellation frequency = speed of sound / (2 * path length difference). All of the boundaries in a small room can cause interference and audible suck-outs – the ceiling, front wall, back wall, side walls and floor.

Real life application:

Measure the direct distance from speakers to the listening position (e.g. 9 ft). Measure the distance from the speaker to a 'mirror point' (i.e. reflection point) AND BACK to the listener (e.g. 18ft). Calculate the path length difference (e.g. 18-9=9ft). Put this figure into the formula above (e.g. 1140/(2*9)). See the frequency where phase interference occurs (e.g. 63Hz).

Note that generally you only have to worry about the path length difference from the first reflection, since the sound decays with distance according to the inverse square law and by the time it has reached and reflected from a second boundary it does not have much energy.

 

[Nyal Mellor]

Also see this on Ethan's site http://www.realtraps.com/art_modes.htm

 

[Ethan Winer]

There are spreadsheets available that will allow you to predict room modes based upon room dimensions. Anyone know of something that will allow one to predict nulls?

I appreciate all the links to articles on the RealTraps site, but the only way to know what frequencies have nulls and where they exist in a room is to measure the room. I believe a room made of foot-thick cement will behave as predicted, but normal construction varies a surprising amount. I once measured a modal peak that was 20% different than what was predicted for its dimension!

Even if you could predict this stuff exactly, it's almost irrelevant because the solution is bass traps anyway. Well, that and optimizing speaker / sub / listener positions too.

--Ethan

 

[audioguy]

Nyal. Since in a rectangular room there are 6 surfaces that provide first reflection points, I assume one would need to make that calculation for all 6 surfaces. That gets complicated.

I would attach an Excel spreadsheet from the SGTH web site (but don't know how apparently) that does show where nulls will show up based upon the input. Unfortunately, it does not match reality, so I am back where I started.

And Ethan: the null I am concerned about is between ABOUT 200Hz and 350Hz (or so) and my room is very heavily trapped already (all four corners, floor to ceiling and three of the 4 wall/ceiling intersections with soffit traps. I'm not trying to predict where the nulls will be but rather to determine what surfaces need to be modified (moved) to get rid of one that has been measured. If I move the seating position forward or backward, it does change the amplitude of the dip and a bit of the Q but is is basically still there.

 

[Ethan Winer]

I'm not trying to predict where the nulls will be but rather to determine what surfaces need to be modified (moved) to get rid of one that has been measured.

Ah, gotcha. I have two suggestions: One is to use my bassy pink noise file to see where those frequencies are building up:

Pink noise aids placing bass traps

In your case, a better solution is to use REW as a live RTA as explained in Part 4 of my Room Measuring article:

http://www.whatsbestforum.com/showt...nd-Acoustic-Treatment-series-Part-4-using-REW

Then with the microphone in one position, you and a friend can walk around with a bass trap and hold it up in the different places and see where the trap has the most affect on that null.

--Ethan

 

[Nyal Mellor]

Nyal. Since in a rectangular room there are 6 surfaces that provide first reflection points, I assume one would need to make that calculation for all 6 surfaces. That gets complicated.

I would attach an Excel spreadsheet from the SGTH web site (but don't know how apparently) that does show where nulls will show up based upon the input. Unfortunately, it does not match reality, so I am back where I started.

And Ethan: the null I am concerned about is between ABOUT 200Hz and 350Hz (or so) and my room is very heavily trapped already (all four corners, floor to ceiling and three of the 4 wall/ceiling intersections with soffit traps. I'm not trying to predict where the nulls will be but rather to determine what surfaces need to be modified (moved) to get rid of one that has been measured. If I move the seating position forward or backward, it does change the amplitude of the dip and a bit of the Q but is is basically still there.

It is not really a lot of work, it will take you 30 mins max! Do you have Excel? Then create a new spreadsheet:

Column A | Column B
Direct Distance | <enter distance in ft>
Indirect Distance | <enter distance in ft>
Cancellation Frequency | =1140/(2*(Indirect Distance - Direct Distance))

Job done!

 

[audioguy]

Then with the microphone in one position, you and a friend can walk around with a bass trap and hold it up in the different places and see where the trap has the most affect on that null.

--Ethan

What would a temporary bass trap have to look like (physical dimensions) to trap frequencies in the 200hz to 300hz range. Based upon Nyals recommendation, the ceiling seems to be at least one of the culprits.

 

[RBFC]

What would a temporary bass trap have to look like (physical dimensions) to trap frequencies in the 200hz to 300hz range. Based upon Nyals recommendation, the ceiling seems to be at least one of the culprits.

Here's some specs that may assist you....

http://www.gikacoustics.com/absorption.html

Lee