or Connect
AVS › AVS Forum › Audio › Audio theory, Setup and Chat › Subwoofer/room acoustics question for Ethan Winer
New Posts  All Forums:Forum Nav:

Subwoofer/room acoustics question for Ethan Winer

post #1 of 19
Thread Starter 
Hi Ethan, how do you explain this broad suck out in the 30-40Hz region with regard to the combined in room response? Is it room related or related to the locations of the subwoofers? Could you give me some recommendations please? Thank you in advance. Of course other members can give their opinions regarding this problem.






Chris
dazzdax@xs4all.nl
post #2 of 19
Since you have 4 subs, why not placing them at different places? If you don't have the freedom to relocate the subs, I think you have to live with the bass nulls.
Else, you can try to adjust the phase controls on the subs. Phase controls may do wonders.
post #3 of 19
Thread Starter 
Hi Ethan, if you got a moment, could you give your response to the question I posted in the AVS forum? Thank you in advance.

Chris
post #4 of 19
Providing a detailled illustraion of the room showing all dimensions as well as the locations of the subs, speakers and measurement position and details of the system such as xovers and slopes, eg are the stats high passed would be a good start.
post #5 of 19
Quote:
Originally Posted by A9X-308 View Post

Providing a detailled illustraion of the room showing all dimensions as well as the locations of the subs, speakers and measurement position and details of the system such as xovers and slopes, eg are the stats high passed would be a good start.

Yes, anything else is just guessing. I will point out that nulls are often caused by reflections coming back from the wall behind the listening position. If you measure again, try moving the microphone forward and back in one-foot increments. If the null changes frequency, that's the culprit.

--Ethan
post #6 of 19
Thread Starter 
Hi Ethan and other fellow members, I have attached a floor plan of the main floor. The width of the listening room is 18 ft, length (front to back) is 42 ft, height is 10 ft.
Main speakers are Soundlab A-1's. I'm using an electronic crossover for the high pass, which is set to 70Hz and 12dB slope. The subwoofers are Rythmik Audio F-15's.
Low pass crossover frequency is about 64Hz with 24dB slope.



Chris
post #7 of 19
Quote:
Originally Posted by dazzdax View Post

Hi Ethan and other fellow members, I have attached a floor plan of the main floor. The width of the listening room is 18 ft, length (front to back) is 42 ft, height is 10 ft.
Main speakers are Soundlab A-1's. I'm using an electronic crossover for the high pass, which is set to 70Hz and 12dB slope. The subwoofers are Rythmik Audio F-15's.
Low pass crossover frequency is about 64Hz with 24dB slope.



Chris

I would guess opposite wall speaker boundary interference i.e. destructive interference of the direct wave from the speaker / sub and the wave reflected off the opposite side wall. You would need to do some distance measurements to be sure.
post #8 of 19
Quote:
Originally Posted by Nyal Mellor View Post

I would guess opposite wall speaker boundary interference i.e. destructive interference of the direct wave from the speaker / sub and the wave reflected off the opposite side wall. You would need to do some distance measurements to be sure.

This would also be my first test too -

Try pulling the subs forward a couple feet and seeing if the nulls change in frequency. If they change in frequency when you pull them towards the back wall, then you've likely found the culprit.
post #9 of 19
Much good advice given above. Ethan, Nyal and Alex ... all good smile.gif


Dazzdax, nice image/room, ... thanks for sharing. Those guys are the pros, for sure. But I would like to offer a few thoughts.

I'd strongly recommend additional placement experimentation of all the parties involved; the Sound Labs, the subs, and perhaps the listening position. I mean just a quick look at your picture, and the floorplan from above, and I see a great deal of possibilities.


You've got a beautiful room, a truly wonderful system, and both appear ideal for placement fine tuning. I'm guessing you've performed some anyway, however I'm confident essentially most all the negative attributes you're experiencing/measuring, could easily be lessened by placement optimization. Fortunately, you've got a room that's conducive to movement away from the front wall. I wouldn't be afraid to bring them out significantly into the space .. allowing more room behind them. Aesthetically, I understand you can't go too crazy,.. but I'm sure there's a combo of front placement, and seating position that would yield superior results.

Ideally, it would be great to experiment with some scattering/diffusion behind the mains. Yeah some proper diffusion products would be ideal, but there are items that would add some scattering, eliciting a better diffuse reflected energy off the back of the mains. If I recall correctly, Sound Lab enthusiast Duke LeJuene, utilizes and recommends something like a Ficus tree (real or otherwise) behind the panels to help in this regard. They're essentially invisible to the LF, yet would offer some nice scattering to the critical octaves above.

The subs; I'd use two of them midwall, on each side and somewhat rearward of the listening position. The other two, equidistant from the rear subs, yet in front of the LP. With careful relative adjustment of subs-to-mains, the sonic "wholeness" or the fabric of the LF will be perfectly interwoven into the mains output (which obviously is the key integrating stats and subs). This will require careful time/signal alignment, careful yes but it shouldn't be hard to tell when it's right. Some iteration of the four separate subs, placed accordingly, will yield an ideal final FR at the seating.

As you most likely know, with such a multi-sub set-up, you'll benefit in both the time and freq domain. The key to "bass speed" rolleyes.gif , I've found to be entirely rooted in two things; time alignment of the LF to the remaining spectrum, the LF decay characteristics of the space. You could really use a great deal of bass trapping in your gorgeous room. But, oftentimes bass damping and the high performance gains that come with bass traps is contra-indicated to aesthetic concerns. Maybe you already have some bass trapping. Remember, the subs "see" the entire volume of room air, and although some areas would offer superior performance to others, they could be carefully disguised and placed in a variety of locations. Nulls, like the one you're seeing, come from destructive interference. Bass trapping both lessen this interference, and tighten up bass decay times ... which subjectively is quite appealing.

A rooms standing waves/modal behavior is primarily established by it's dimensions. However, how significantly the standing waves are excited is a function of how the LF sources are located. With four spaced subs, along the long dimension yet well out of the corners, they're very likely going to exhibit less modal influence and more direct sound. There will be one LF source per room quadrant (as seen at the LP), and the FR will possess further smoothing merely by being closer to the LP (just like a close mic measurement, more sub, less room).

Another consideration with the current placement approach you're using, the rear wall behind the listener is enormously significant. Like Ethan mentioned, the rear wall is important in nearly any set-up. However, by spreading the subs fore and aft, left and right, and by bring the mains further off the front boundary, the rear wall's significance is diminished.


Anyway, it would be a pleasure experimenting with those stat panels, those subs, .. in that nice room. Maybe after all the experimenting/measuring, you'll discover another set of placement ideas. But what I layed out is the approach I'd pursue given the latitude to do so. Maybe I'm way off ... eek.gif


Best of luck
post #10 of 19
Thread Starter 
Hi folks, thank you so far for your contribution. FOH, could you please mark the suggested locations for the subs on the floorplan so I know exactly what you mean?
What do you think of the attached floor plan with modified subwoofer locations? The subwoofers remain stacked.



Chris
post #11 of 19
Quote:
Originally Posted by dazzdax View Post

What do you think of the attached floor plan with modified subwoofer locations? The subwoofers remain stacked.
While the subs are still stacked and at the front wall, could you try moving each stack slightly, so that the middle of the drivers are 4.5 feet in from the side walls, and then re-measuring just the subwoofer response?
post #12 of 19
Quote:
Originally Posted by sdurani View Post

While the subs are still stacked and at the front wall, could you try moving each stack slightly, so that the middle of the drivers are 4.5 feet in from the side walls, and then re-measuring just the subwoofer response?

Yes, move them from their current location inwards and remeasure
post #13 of 19
Thread Starter 
Yes, I will relocate them and remeasure. But I'm a bit sceptical about the broad and very deep suck out around 30Hz. Destructive interference at 30Hz means a quarter cycle length of 2.8 m (= 9.3 ft). The subwoofers are separated from each other about 5.0 m (= 16.5 ft), so it is more likely they would show a constructive interference at 30Hz. Please explain this phemonenon.

Chris
post #14 of 19
Quote:
Originally Posted by dazzdax View Post

Yes, I will relocate them and remeasure. But I'm a bit sceptical about the broad and very deep suck out around 30Hz. Destructive interference at 30Hz means a quarter cycle length of 2.8 m (= 9.3 ft). The subwoofers are separated from each other about 5.0 m (= 16.5 ft), so it is more likely they would show a constructive interference at 30Hz. Please explain this phemonenon.

Chris
Not sure which phenomenon you are curious about --
Modal interactions are funny things at first glance. Placing a sub at the 1/4 width ( 2nd axial modal null position) and the other 2nd axial width null (3/4 width) allows for gain matched subs to provide effective cancellation of the first 3 axial width modes.

As an aside, the plots you showed are perplexing as regards the stated crossovers. I see no indication of a forth order roll off low pass on the sub for example.
post #15 of 19
Quote:
Originally Posted by dazzdax View Post

I'm a bit sceptical about the broad and very deep suck out around 30Hz. Destructive interference at 30Hz means a quarter cycle length of 2.8 m (= 9.3 ft). The subwoofers are separated from each other about 5.0 m (= 16.5 ft), so it is more likely they would show a constructive interference at 30Hz. Please explain this phemonenon.
Your room's 18 foot width means the first width mode is at 31.4Hz. That's a possible explanation for this phenomenon (very deep suck out around 30Hz), especially if you took the measurement from the midpoint of room width (null for the first width mode). Hence my request to move the subwoofer stacks to the quarter points of room width and re-measure.
post #16 of 19
Quote:
Originally Posted by dazzdax View Post

Yes, I will relocate them and remeasure. But I'm a bit sceptical about the broad and very deep suck out around 30Hz. Destructive interference at 30Hz means a quarter cycle length of 2.8 m (= 9.3 ft). The subwoofers are separated from each other about 5.0 m (= 16.5 ft), so it is more likely they would show a constructive interference at 30Hz. Please explain this phemonenon.

Chris
It's the reflected sound bouncing off the side wall opposite the sub and interacting destructively with the direct sound I am talking about...that's speaker boundary interference. Sanjay is referring to lack of modal support, also a potential root cause. It could well be a bit of both.
post #17 of 19
Thread Starter 
Hi all, today I did some measurements after I relocated the subwoofers to 1/4 distance from the sidewalls (as suggested by Sanjay). It becomes more strange: the broad dip is still there, but the high peak at 28Hz is augmented.
More weird things: I don't see any effects of the crossover filters either with the Soundlabs as with the subwoofers (this has been mentioned earlier by Jamin). Do these crossover filters work as intended? I'm quite confused!





Chris
post #18 of 19
Aside from specifics on placement changes, etc., prior to analyzing any measurements, I'd say you need to utilize more playback level, higher SPL, in your graphs.

Also, some near-field, close mic measurements would be great to assure the measuring rig, etc., is sound. Perhaps everything is fine, but I just feel it's prudent prior to making significant advice and changes based on measuring graphs.

Many prefer 45dB-105dB vertical limits, and 10hz-200hz horizontal. With consistency, one can become more adept at quickly visualizing the issues.

Thanks
post #19 of 19
Quote:
Originally Posted by dazzdax View Post

Not sure if that's a canyon (dip) or a valley between two mountains (peaks). There is a difference. If you can make those peaks drop, then you'll hopefully get smoother response. Will be easier than trying to boost the dip.

If I divide the speed of sound by the frequency of the peak in that chart, I get the length of your room: 1130 ÷ 27 = 42. The second length mode peaks at the front wall, back wall and midpoint of room length. Is your seating around 21 feet from the front wall? If so, can you measure from 17 ft or 14 ft from the front wall?

As Nyal said, your frequency response problems could be due to a variety of factors. If you can address some of the modal problems using placement, you can then address boundry cancellations with treatments.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Audio theory, Setup and Chat
AVS › AVS Forum › Audio › Audio theory, Setup and Chat › Subwoofer/room acoustics question for Ethan Winer