Addressing a Null

[gshelley]

Hello,
Have been reading and learning a lot on room acoustics and this forum has been a BIG help. Currently challenged with ~20dB null at 30Hz. Which you can see in the graph. How do I isolate the cause, and treatment, for the null? (e.g. SBIR, Room mode, other?)

Have moved the main speakers forward/back (along with the listening position) and find the null gets deeper as I bring the speakers forward from the front wall.
The frequency of the null seems to be the same irregardless of the speaker position. Have moved the speakers between 75" and 83" from the front wall. Again with the null becoming more severe as I move the speakers out from the front wall.

This measurement is taken in a dedicated listening room that is 17'9" x 25' x 10'. Currently listening to Wilson Alexandria's with a JL Audio Gotham.

Very interested in hearing about your experiences and, if possible, and recommendations you may have.

 

[amirm]

You have a room mode (null) along the dimension of 17'9" at 32 Hz. Is that the width of the room and do you sit in the middle? If so, that is the cause.

 

[amirm]

Here is the location of all of your room modes:

Where the sub located right now? In the corner?

 

[gshelley]

Amir,
Thanks for the response. The sub is currently centered between the speakers ~14" off from the front wall. Have tried different positions along the front wall with no real impact on the null.

 

[Nyal Mellor]

Amir,
Thanks for the response. The sub is currently centered between the speakers ~14" off from the front wall. Have tried different positions along the front wall with no real impact on the null.

Are the mains crossed over (if so what crossover point) or are you running them full range?

 

[NorthStar]

Amir,
Thanks for the response. The sub is currently centered between the speakers ~14" off from the front wall. Have tried different positions along the front wall with no real impact on the null.

Hi,

Adding a second sub would/could help. ...And also the listener's main position; if only by moving few inches sometimes (sideways or/and forward/backward, horizontally and vertically). ...The rule of thirds, fifths, sevenths, etc. applies ...For the loudspeakers, the subwoofers, the height of the listener's ears, the room's width and length at which the main listener's ears are intersecting (main position), in relation to everything else.

 

[microstrip]

gshelley,

I think that most of the time the critical nuls are caused by differences in length between direct path and the paths of reflected sounds in non absorptive walls. In a 7.5m long room you will typically have a deep nul between 30 and 35 Hz, depending on distances from front and back wall. The best way to fill them is using large bass traps tuned to the null - not an easy thing, or a sub far from the main speakers adjusted to compensate it.

As you have seen when you approach the speakers of the front wall the difference of the length of the paths become smaller and null becomes less deep - but imaging and depth of soundstage can suffer. I went through this process with a X2 recently.

A good point of the XLF is that you can change the orientation of the bass port, allowing more flexibility in speaker positioning.

 

[audioguy]

I vote for multiple subs placed appropriately in the room.

 

[NorthStar]

In post #3, from Amir, you have a very useful graph for your room's modal nulls & peaks. ...Great basis point to start with.

 

[j_j]

Looks to me like the right thing is to consider appropriate room treatment from the start. That's a pretty powerful notch.

 

[NorthStar]

Looks to me like the right thing is to consider appropriate room treatment from the start. That's a pretty powerful notch.

A 'null' is tough to fill. :b ...Best is to sit outside of it, and by re-positioning some stuff around. ...I think.

 

[DonH50]

Getting enough room treatment to fix a 30 Hz null is tough. I have a similar null in my room; my solution was to move the listening position to get out of the null.

 

[gshelley]

Nyal,
The mains are running full range with the sub filling in at ~47hz. I prefer having the speakers out from the front wall for a better imaging effect. Of course, I lose bass coupling when doing so thus the 'need' for the subwoofer.


RE: listening position

According to the graph Amir provided, the null is due to a room mode that is centered on the width. Sitting in the middle of the room (width) seems the only logical spot for best L/R symmetry. So, how does one avoid the null? Assuming this is the cause.

Microstrip,
if I read your post correctly, you are suggesting cancellation due to reflections from the wall (SBIR?). I recall you built a membrane absorber. Were you dealing with a similar issue?

 

[j_j]

A 'null' is tough to fill. :b ...Best is to sit outside of it, and by re-positioning some stuff around. ...I think.

Thing is, even if you move outside of it, the room storage is going to be there at that frequency, and the hangover (in time) is going to be ginormous.

And that by itself can mess up articulation and sound kind of bad to wretched.

Some absorbent material with an open back set out at least 6" from the wall will help more than you might imagine. Putting a membrane absorber at a pressure point would help, too. But that is a whopper of a room mode, for sure.

 

[NorthStar]

1. The room is 17.75' (W) by 25' (L), and 10' high.
2. 17.75' divided by 2 = 8.875'
3. 8.875' multiplied by 3 = 26.625'
4. Not bad at all.

5. Try the main listening position at 2/3 (16.66') the length of your room. ...Then try again at 3/5 the length (from the front wall).
...And 4/7, 5/9, 6/11 ... experiment. ...But thirds, fifths, or sevenths; one of them should do.
6. Try slightly to the left, and right of the center width line; main listening position. ...Half/quarter foot or so, and position your mains in consequence.
7. Measure (SPL meter) each time from a test disc, and from all the frequency bands of the audio spectrum; say from 20Hz to 16kHz.
...And in particular from 25Hz to 35Hz (your null issue). And also at the crossover point between your mains and sub.
8. Make sure you compensate for the inaccuracy of the SPL meter at the lower frequencies (compensation chart).

9. Subwoofer: proper phase; again with a good audio test disc.
10. Experiment with the Low Pass filter. ...A good place to start is twice what your speakers' lowest bass frequency is (minus 2dB).
So, for example they register -2dB at 30Hz; just try the LPF at 60Hz, and also 50Hz, and 70Hz (because your sub probably doesn't accurately trace the numbers from the x-over (LPF) knob's dial.
11. And experiment too with sub's positioning.

* Others might jump in to rectify, elaborate, or simplify.

** And like I said before; two subs are better than one.

*** Patience and experimentation (positioning and measurements) goes hand & hand.
{It is time consuming, and you'll get as far to the best as your stopping limit will bring you; when you have enough to be satisfy with.}
-> Perfection is the best compromise. :b

 

[microstrip]

(...)

Microstrip,
if I read your post correctly, you are suggesting cancellation due to reflections from the wall (SBIR?). I recall you built a membrane absorber. Were you dealing with a similar issue?

gshelley,

Please see the sketch bellow (I hope there is no big mistake in it, I am sure any of our experts will point any inaccuracy).
The problem is mainly moving the triangle in the room minimizing nulls and getting good stereo image. Unfortunately the thirds rule creates horrible nulls at low frequencies in long rooms with strong, non absorptive walls - mine are 2 feet stone!

I built a membrane absorber to decrease the RT60 time at low frequencies with success. It also ameliorated the null depth by a few dBs. But in order to deal with the null I will have to build a second tuned wall at the back of the room, just moving speakers and listener could not suppress it. I already designed it, but I have to remove a very large book shelve with all its contents and another side one filled with LPs to built the wall ...

 

[Nyal Mellor]

Nyal,
The mains are running full range with the sub filling in at ~47hz. I prefer having the speakers out from the front wall for a better imaging effect. Of course, I lose bass coupling when doing so thus the 'need' for the subwoofer.

What does a single speaker measurement look like?

 

[Bjorn]

Running one subwoofer and the fronts as fullrange will often cause a more uneven response. Measure both separately first and post the graphs.

 

[gshelley]

All,
Thanks a million for jumping in here and giving your input. Certainly one of the best things about this forum!
Really like the 'custom' graph... Thanks Microstrip for taking the time to write and post it. From, this it looks like I have a triple play; Front & Back wall SBIR, with a mid-room null that Amir posted. Can this be?

If so, no wonder the null is so deep and wide.

Here are additional plots of the speakers response. Unfortunately, I am not confident of the speaker location for the individual L, R graphs. Although the 30Hz null is showing itself in both. The subwoofer plot is with its current position.

 

[gshelley]

Sorry, the graphs did not include the description of the graph. They are, in order;
-Left speaker only, no subwoofer
-Right Speaker only, no subwoofer
-Subwoofer only
-L & R, no subwoofer.