Subwoofer position with least nulls? (1 Viewer)

[Chris PC]

Let me get this straight. Every room is a little different, but I was wondering about subwoofer corner placement. What is it about corner placement that is beneficial and what is poor about it? I know it can sound "boomy". Is that simply nodes , peaks or standing waves at various frequencies? Or does corner placement have nulls too? If there are less nulls with corner placement, and you have a subwoofer and an equalizer, does it not make sense to put the subwoofer in the corner and tame the peaks? From what I understand, nulls are harder to fix, and require adjusting placement mostly, but with peaks, you can trim them, correct? So proper corner placement using a good EQ means fewer or smaller nulls, and tamable peaks? Since you are reducing the peaks with an EQ, would this result in higher efficiency and/or greater headroom since you are reducing the power requirements? At least for those peaking frequencies, anyways.

How many people do this and do you notice a difference in headroom or overall better frequency response?

 

[Saurav]

From a purely layman's POV (there are many many people on this forum who know much much more than me on this) - corner placement excites all 3 room modes. This can be a good thing if your 3 room dimensions are fairly different, because, like you said, that's easier to equalize, and it may result in an overall flattish response due to the 3 peaks. This can be a bad thing if your room is almost a cube, because now you have a huge peak at one frequency.

 

[Chris PC]

Right. But corner placement does not necessarily mean no, or fewer nulls, does it?

I guess you can just try it and see, as long as your room is not a cube? My room is 12.5 by 14 by 8 feet so not really a cube, but the horizontal dimensions are a little on the square-ish side. My room is weird though, with flimsy panel walls on two walls and two concrete walls and a concrete floor. I guess without an EQ, I can try the corner again out of curiosity and see what the response is.

 

[Saurav]

It's not really a cube, but close enough, maybe. What are the resonant frequencies for those 3 room dimensions? I don't remember the math now.. it should be 1/2 a wavelength, and the speed of sound in air is.. what?
OK, roughly 345 m/s, which is 1138 f/s. Now, speed = wavelength * frequency. For your room dimension to be 1/2 a wavelength, you're talking about 25', 28' and 16' waves. What are the frequencies for those? frequency = speed / wavelength, so you get:
45.5Hz
40.6 Hz
71.1 Hz
I don't know what your nulls would be (it's possible to calculate that), but I can see the first 2 peaks combining into a bigger peak.
Again, I am a complete layman at this stuff, so take this for what it's worth

 

[Harold_C]

Corner placement is generally best because most people really don't have enough subwoofer to play at Dolby reference levels. Corner placement will give you most output and, therefore, the best chance of getting sufficient subwoofer levels.

As a general rule, moving the sub out from the corner by a foot or two will probably reduce the level of one or more of the peaks in output. If you have enough subwoofer, this may make the bass a little easier to tune.

I think that lower resonant peaks (say in the 30 - 50 Hz range) aren't going to cause big problems. The ones that make a subwoofer sound "muddy" are the peaks in the crossover region -- say 80 to 100 Hz. If you set the subwoofer levels for those peaks, then you'll probably be losing a lot of output down very low. However, if you don't set the subwoofer levels for these upper bass peaks, then they will overpower the harmonic structure and make the bass output seem really boomy and muddy.

 

[Chris PC]

Actually, my room shape is probably not too bad. I remeasured it. 14' 4" Long x 12' Wide x 8' high ceiling (drop ceiling at 7' but underneath (heheh) I have egg cartons glued to the full 8' ceiling. (from my drumming days!). So I guess my room is pretty decent for not being a square cube or having a 1.5:1 dimension. It does have weird walls though. I bet drywall or solid wood walls would change the sound of the room, but who knows if it'd worse or not.
In my room, I currently have my Mirage BPS400 sub located 7 feet from the 14 foot wall. Its not quite half way, because the subwoofer is 2 feet long. Plus its bipolar, so thats another difference. So really, the center of the subwoofer is about 3/5 along the wall one way or 2/5 along the other way, depending which wall you measure from. I'll try the corner again out of curiosity though But I've no EQ to tame any peaks. I'll try the corner and measure the response to see.

 

[Dennis B]

Chris,
The peaks and nulls positions are really a function of the room dimensions and not of speaker/listener placement, which in turn determine their amplitude.
For a simple approach, the axial harmonics for your room are determined by n*c/2*L, where c=343 m/s (1130 ft/s) is the speed of sound, n is the n-th harmonic and L is the room dimension under consideration.
If you notice how these resonances or modes build up, you'll see that for a given dimension between two parallel walls, there will always be a null right in the center for all odd harmonics, including the fundamental, and a peak for all even harmonics.
Now the real frequency response of your room in the low frequencies will actually be the result of the combination of all the modes (including tangential and oblique) and how much they are exited by each speaker (and sub).
The corner produces the maximum acoustic coupling between the sub and the room, therefore peaks are at maximum, but watch out for the nulls. If you room dimensions are large enough, most of your problems go away as the peaks and nulls are very close together and tend to blend in a nice way.
Your first resonances are relatively spaced apart, which on one hand isn't very good, cause you'll notice them more clearly, but on the other hand are easier to tame with a BFD or other parametric eq:
39.4 Hz, 47 Hz, 70.5 Hz, 78.7 Hz, 98.4 Hz, 118.1 Hz, 141 Hz (this one a double peak).
So theoretically there are 'optimum' positions in any room where you should place your sub (and speakers) for best low frequency performance, but that would most likely turn the sub into a table or another piece of furniture...
Best approach IMO if you're really into it is getting ETF5 or another software that can easily give you the room's frequency response, from which you can really understand what's going on in your room. It's a lot easier than plotting all modes across a graph for all dimensions, plus it will take into account all aspects of the room, including those egg cartons...

 

[Chris PC]

Cool. Thats darn helpfull advice.

Your first resonances are relatively spaced apart, which on one hand isn't very good, cause you'll notice them more clearly, but on the other hand are easier to tame with a BFD or other parametric eq:
39.4 Hz, 47 Hz, 70.5 Hz, 78.7 Hz, 98.4 Hz, 118.1 Hz, 141 Hz (this one a double peak).

Those are resonances, right? So nodes or peaks? What about nulls?
Did you quickly calculate those for my room? Are you solving for the axial harmonic? What are you using for n? Could you show me an example of a calculation? Where can I read more about this stuff. I want to play with it as a way of getting interested in math so I have a desire to sharpen my math skills
I downloaded the demo ETF program so perhaps I'll toy with it. How much better is the full version? I could pick up and EQ too
Thanks

 

[Richard Greene]

Great post Dennis B.

You should write a book.

Please include this simple formula for estimating the frequencies of axial room modes in rectangular rooms:

(1) 565/room dimension in feet = Room Mode in Hz.

(2) Room mode in Hz. x 2 = second harmonic room mode

(3) Room mode x 3 = third harmonic room mode

Use (1), (2) and (3) for room length and width and height

Assume room modes are 5Hz. wide (usually 5 to 10Hz. wide)

to predict whether any modes are likely to be stacked or adjacent (makes them easier to hear & more annoying)

Sine wave measurements are needed to identify the exact

room mode frequencies. Especially if the walls are not parallel (it is a common myth that non-parallel walls eliminate room modes -- non-parallel walls just make it difficult to calculate room modes)

A room corner (tricorner) is a high pressure zone for all room modes. Placing a subwoofer in a corner will fully excite all room modes. That results in maximum bass energy entering a listening room. If a room mode(s) results in a frequency response peak(s) heard at your listening position, the solution is bass traps or parametric equalization. The only room modes that matter are the ones heard/measured at your listening position. Most likely you will hear one or more frequency response peaks (nodes) and one or more troughs (antinodes or nulls) in the bass frequency response.

Peaks are much easier to hear than troughs ... and that's a good thing because parametric equalization can only reduce peaks .. and only at one listening position.

There is no optimum listening position and subwoofer

location to eliminate all these peaks and troughs.

Even if there was an optimum listening position for bass frequency response, it would not make sense to sit there if was not an optimum location for the main speakers.

It is also very unlikely that an optimum main speaker

location would also be an optimum subwoofer location.

Two subwoofers located on opposite sides of a room can be used to tame one specific room mode and its odd order harmonics ... but eliminating one room mode usually makes the remaining room modes easier to hear (our ears try to smooth the frequency response -- that works best when there are lots of room modes (such as above 200Hz.) that are not stacked (as in a square room) or adjacent (as in a

near-square room). Since the primary acoustics problem below 100Hz. in home listening rooms is insufficient modal density, eliminating one room mode can be counterproductive.

 

[Chris PC]

Even if there was an optimum listening position for bass frequency response, it would not make sense to sit there if was not an optimum location for the main speakers.

So it makes sense to place the subwoofer where it sounds best, where it has the best response when you are seating in the best location in terms of your main speakers. This makes sense of course. I guess the idea is to achieve the best flattest response. Thanks for mentioning that eliminating only one mode will not necessarily be a good thing. I will keep that in mind and try to go all or nothing. Definitely requires an EQ.

 

[Kwang Suh]

Try placing the subwoofer where you normally sit, then walk around the room on all fours and listen to the sub. You'll find that different areas sound better than others. Once you've found an area that sounds reasonable, put the sub there.
Of course, there are practical limits to this. My sub sounds best right between the living room and the kitchen about 1 meter from the wall. Can't place it there

 

[Dennis B]

Chris,

I downloaded the demo ETF program so perhaps I'll toy with it. How much better is the full version? I could pick up and EQ too

The full version will allow you to save your work and overlay multiple plots and positions you've recorded for A/B comparisons. The SW can do a lot more things than just low frequency testing, though. The "demo room" on their site is well worth a visit.
Did I hear BFD...?

 

[Dennis B]

Richard,
Thanks. Yours wasn't bad either...
While I second all your thoughts, I'd like to comment on one of them:

There is no optimum listening position and subwoofer
location to eliminate all these peaks and troughs.
Even if there was an optimum listening position for bass frequency response, it would not make sense to sit there if was not an optimum location for the main speakers.
It is also very unlikely that an optimum main speaker
location would also be an optimum subwoofer location.

This is perfect, and I just wanted to make clear that when I referred to an 'optimum' position I really meant the one with the flattest response, and not the one with a flat response, beacuse it's just not going to happen.
Chris also noted this well.

 

[Chris PC]

Cool. I think thats the book I was thinking about