What is the cause of this room null on Room Sim?

[neRok]

I can't work out what the cause of this null is on the REW Room Sim. I think it is something "room length" related, because absorption on the other surfaces doesn't affect it. As the screenshots show (taken from this thread/discussion), the frequency decreases as you move away from the back wall.



It's not related to the half cycle of the problem frequency, because the wavelengths are too long, particularly as the problem approaches 40Hz. I did up a quick spreadsheet and can see that the distance to backwall does hover around 26%-29% of the problem wavelength though, so maybe the answer is there?



Below are some individual speaker measurements in the room (these are OCA's measurements). I wonder if the problem is 52Hz in these measurements? The right speaker measurement doesn't seem to show the issue, however there is a small blip on the group delay. The Left speaker however does show an issue around 52Hz. But that measurement is also peculiar because on the phase plot you can see the steep section at the problem frequency, and the predicted phase plot actually shows it "switching". I think this might be an "artifact" of how REW has processed the data, because the spectrogram also shows how REW has interpreted the "second peak" as the main peak, and that baseline is then what the phase plot is based upon?

Edit: Maybe the null in question is actually at 48.6Hz in the measurements. Both group delay plots have a null there, and both have an SPL low point there too.

 

[sam_adams]

These are the primary modes of the room imaged by AMROC:

Why do you have the surface absorption coefficients set so high?

 

[ozzy9832001]

I think half of one side of your room will be a null at 50hz and the other not.

 

[neRok]

These are the primary modes of the room imaged by AMROC:

Yes I have seen those, REW Room Sim shows the same at the bottom of the graph, but this frequency is none of those. It isn't the width mode of 50Hz because if it was, I believe it would stay at 50Hz when moving the listening position forward and back.

I increased the absorption to confirm which wall(s) were causing the problem, because high absorption effectively negates the reflection (afaik). The side walls and ceiling/floor do not affect this null, only the ends, like this confirms;

Edit: And moving the sub over where a speaker would be doesn't change it either;

I think half of one side of your room will be a null at 50hz and the other not.

Further to above, I believe any problems from the side wall should be consistent across the rooms length (like the green arrow).

 

[sam_adams]

The modes indicated in the simulation are multiples of the primary length mode at 31.5 HZ. So, 63.0 Hz, 94 Hz, and 125.8 HZ; 2nd, 3rd, and 4th multiples respectively.

 

[neRok]

The modes indicated in the simulation are multiples of the primary length mode at 31.5 HZ. So, 63.0 Hz, 94 Hz, and 125.8 HZ; 2nd, 3rd, and 4th multiples respectively.

Yes but I'm wondering what the null is, the null that decreases freq as you move from the back wall.

 

[sam_adams]

Yes but I'm wondering what the null is, the null that decreases freq as you move from the back wall.

Because you move through the nulls created by the multiples of the primary axial mode:

Don't let the above fool you because reality—even simulated—is much less precise:

 

[neRok]

Because you move through the nulls created by the multiples of the primary axial mode:

View attachment 304645

"At the walls there must always be a sound pressure antinode." Your graphic shows this. It makes sense when you think about it, because at the moment the wave is at peak amplitude and arrives at the wall, it is then reflecting on it self and boosting itself. And thus at half wavelength distance from the wall you have +dB reflection on top of -dB direct wave, and thus cancellation (node = null).

As an example of one the room sims I posted, the LP is 2.43m from the back wall, which equals the half length of ~70Hz. However the sim is telling me the null is at ~40Hz, and the half length of that is actually ~4.29m. These dimensions suggest it does not relate to a node from the back wall reflection.

 

[dasdoing]

"At the walls there must always be a sound pressure antinode."

but it's nodes you are looking at. look at Sam's graph again.

or look at this

everytime you move away from one node, you are approuching another. that's what you are seeing

 

[neRok]

but it's nodes you are looking at. look at Sam's graph again.

or look at this

I've looked, they show the same thing = a node half the wavelength from the wall.

LP is 2.43m to backwall, that's half wavelength of 70Hz. 70Hz != 40Hz REW Room Sim is showing.

 

[neRok]

Worked out my problem. I wasn't considering the wave is the "up and down" (crest and valley), I was just looking at one or other. So actually all those 1/2's I drew are technically 1/4's, and now it makes sense. Also it lines up with the spreadsheet, which was always showing ~1/4's.

 

[dasdoing]

Worked out my problem. I wasn't considering the wave is the "up and down" (crest and valley), I was just looking at one or other. So actually all those 1/2's I drew are technically 1/4's, and now it makes sense. Also it lines up with the spreadsheet, which was always showing ~1/4's.

View attachment 304681

yea, the fundemantal is omited in the graph cause it has no node. this stuff gets confusing fast lol.

EDIT: actually it should be in the graph, no? it is a standing wave, too. someone update the wiki lol