or Connect
AVS › AVS Forum › Audio › DIY Speakers and Subs › Double Bass Array (DBA) - The modern bass concept!
New Posts  All Forums:Forum Nav:

Double Bass Array (DBA) - The modern bass concept! - Page 5

post #121 of 321
First, I'm always surprised this thread doesn't get much traction...

------


Quote:
Originally Posted by MBentz View Post

That said, I would argue that there are some psycho-acoustic benefits to not having pressure gain effects.

Interesting, care to elaborate?



Quote:
Originally Posted by MBentz View Post

The cool thing is you could set up your processing to choose your flavor with the flip of a switch...

Yep, easy way to explore optimization approaches/options. This would certainly aid in the examining of your contention above.

-----

The lack of the benefits of PVG/Room Gain, which is one of the rare instances of a free lunch in audio, seems to be the leading reason not to pursue a DBA. Given the option of proceeding without the loss of these reciprocal gains, seems quite attractive and maybe would be the way to get more interest in similar pursuits.


Thanks
post #122 of 321
I'd be more likely to do a single bass array which would still have room gain, and simply treat the rear wall with pressure traps to target depth modes. In a HT room you can then sit closer to the rear wall, the cost is halved also.
post #123 of 321
Quote:
Originally Posted by paulspencer View Post

I'd be more likely to do a single bass array which would still have room gain, and simply treat the rear wall with pressure traps to target depth modes.

Yes, exactly,......like PNW and the Octagon; (18)15's launching toward a massive, complex, rear wall limp-mass membrane rear wall surface that's transparent and dissipative below "x" frequency.

The driver spacing addresses height and width modes, and the faux rear boundary addresses front to back modes.



Thanks
post #124 of 321
This is the most interesting thread regarding linear and time-correct bass I've read since I learned about Geddes'/Welti's/LeJeune's theories of Distributed Bass Arrays.

THIS DIY recipe employs two identical stand mount monitors in an L-shaped array for ambiance and bass mode cancelling effects above the sub woofer range. Seems like a good match for DBA. The promised spatial effects and sound intensity are frosting on the cake.
post #125 of 321
Hello all.. New Member Here.. Have Much interest in Taming rooms

Any chance at seeing the full range 1/24oct response measurement.. Or did I miss it?
post #126 of 321
Thread Starter 
Hi folks,
time have passed and I've got a house now with a dedicated homecinema room in the cellar. smile.gif

Right now I'm in the planning phase, but I can tell you that I will not install a DBA, but a SBA with a damped back wall instead. A Single Bass Array (SBA) consists only of the front array. The plane wave is still formed and all modes beside the length modes are eliminated. The length modes are excited to the max, so an effective absorber for the back wall is a must!

The room size is 6 x 4,8 x 2,2 m and it was completely empty when I did this experiment. I calculated the correct flow resistivity for the porous absorber with a maximum depth of 60 cm to eliminate the first length mode of 27 Hz. I bought 15 packages of Rockwool Sonorock which has a flow resistivity of 6 kPa*s/m². This is suitable for thick bass absorbers. I measured with only two of the old DBA subwoofers, because two drivers are enough to form a plane wave beyond the second length mode. Let's see the results. smile.gif

Front:


Back:



I measured at three microfon positions: 2m, 3m and 4m distance to the front wall.

Frequency response
No treatment:


60 cm rockwool:


Decay:
2m:


3m:


4m:


Spectrogram at 4m:
Empty room:


SBA:


DBA:


As you can see this temporary absorber works pretty well. The first length mode is still there, but the decay time is greatly reduced. And a little boost at 27 Hz can make much fun when watching movies! wink.gif

There are several reasons I chose the SBA over the DBA.

1. Costs! While saving the money for the back array (and amplifiers) I'm now able to install 18 x Peerless XXLS12 in closed boxes. When the room is finished I will post more details. smile.gif
2. The satellite speakers also benefit from the absorber above 100 Hz.

On the other side the contras are:

1. Response is not as flat as a DBA.
2. More room is required for the absorber, since it shall damp the first length mode as well.

There are several SBA's out there which lack the absorber at the back wall. The response is horrible. But if well-planned a SBA will work!
post #127 of 321
How about putting the SBA on the rear wall, rather than the front wall?
I ask because then you can put all that absorbing material on the front wall instead (maybe even put the front speakers in the rockwool(front sticking out ofc)) and get absorbtion benefits for the main speakers too.
Also, on the back wall i belive it would be beneficial with diffusors rather than absorbtion for more "spacey" sound(Both for the sound coming from the front, and surround speakers). Kinda like LEDE

So lots of rockwool on the front wall, absobtion on the 1. reflections (walls and ceiling atleast) then diffusors on the back wall and maybe some in the ceiling and on the walls in the rear half...
post #128 of 321
Thread Starter 
Quote:
Originally Posted by eXa View Post

How about putting the SBA on the rear wall, rather than the front wall?
I ask because then you can put all that absorbing material on the front wall instead (maybe even put the front speakers in the rockwool(front sticking out ofc)) and get absorbtion benefits for the main speakers too.
Also, on the back wall i belive it would be beneficial with diffusors rather than absorbtion for more "spacey" sound(Both for the sound coming from the front, and surround speakers). Kinda like LEDE

Of course this would be possible. But my approach is different. smile.gif

I'll build a baffle wall which integrates my self-built loudspeakers. This has the following advantages:

1. No SBIR from behind the speakers
2. Higher sensitivity and maximum SPL of front speakers
3. localization of SBA is always in the middle of the screen

The front speakers still benefit from the rear absorber, because the modes of higher order and mid/high reflections are reduced. But I consider to build slats on the absorber if RT60 is not well balanced in terms of frequency. So the lower frequencies can pass the slats and the higher don't. And of course there will be absorbers and diffusers on the side walls and the ceiling. But I have to measure the finished room first do decide the next steps.


Edited by FoLLgoTT - 8/25/13 at 11:11pm
post #129 of 321
Why not cancel the odd-order length modes by putting subs on the back wall?

I bet you could buy a lot of drivers for the cost of all that rockwool.

And contrary to the thinking (not necessarily yours) that you have to buy more amps for more speakers, the opposite is true (if you add enclosure volume w/speakers), as more Sd increases efficiency .
post #130 of 321
Thread Starter 
Quote:
Originally Posted by noah katz View Post

Why not cancel the odd-order length modes by putting subs on the back wall?

Why only the odd-order (1./3.) modes? This makes no sense.
Quote:
I bet you could buy a lot of drivers for the cost of all that rockwool.

You lose this bet. wink.gif
One package rockwool costs about 12 € here in Germany. In my case this is 180 € for 15 packages. For this price I would get one lonely driver.
Quote:
And contrary to the thinking (not necessarily yours) that you have to buy more amps for more speakers, the opposite is true (if you add enclosure volume w/speakers), as more Sd increases efficiency .

If you go for a specific SPL this is true. But if you want to reach the mechanical Xmax of each driver, amplifier power scales lineary.
post #131 of 321
Yeah but after a certain amount of drivers, you have plenty of SPL anyways, so more drivers after this is just for lower THD and smoother reponse and so on.
post #132 of 321
Thread Starter 
Quote:
Originally Posted by eXa View Post

Yeah but after a certain amount of drivers, you have plenty of SPL anyways, so more drivers after this is just for lower THD and smoother reponse and so on.

That's right, but additionally you are able to lower the frequency at which the target max SPL can be reached. Of course this is only true for closed boxes. A shelving filter + high pass will linearize the target response. And thus more amplifier power is required.
post #133 of 321
Quote:
Originally Posted by FoLLgoTT View Post

Why only the odd-order (1./3.) modes? This makes no sense.
Modes have polarity. So the modal peak on one wall will be as loud on the opposite wall but will be in opposite polarity. Which means that subs on opposite walls reproducing the same signal (same level, same phase) will cancel odd-order modes. You can try this yourself by placing a pair of your subs on opposite (front & back) walls, directly opposite each other, measuring each sub then both together. And you won't have gotten the smoother response by absorbing away much of the bass you spent money to create in the first place.
post #134 of 321
Quote:
Originally Posted by FoLLgoTT View Post

Why only the odd-order (1./3.) modes?

Further to Sanjay's nice explanation, the Welti white paper:

http://www.harman.com/EN-US/OurCompany/Innovation/Documents/White%20Papers/multsubs.pdf

If additionally to front and back you put the subs at the walls' midpoints, you don't excite the side-to-side odd-order (I think) modes.
Quote:
Originally Posted by sdurani View Post

And you won't have gotten the smoother response by absorbing away much of the bass you spent money to create in the first place.

Good point!

So rather than buy more woofers, take half from the front and put them in the back.

Though I've never been clear on how the levels work, i.e. do you more or less get 6 dB increase for 2X the subs when you do this.
Edited by noah katz - 8/29/13 at 5:48pm
post #135 of 321
Thread Starter 
Quote:
Originally Posted by sdurani View Post

Modes have polarity. So the modal peak on one wall will be as loud on the opposite wall but will be in opposite polarity. Which means that subs on opposite walls reproducing the same signal (same level, same phase) will cancel odd-order modes. You can try this yourself by placing a pair of your subs on opposite (front & back) walls, directly opposite each other, measuring each sub then both together. And you won't have gotten the smoother response by absorbing away much of the bass you spent money to create in the first place.

Ok, now I understand what you mean. I already measured this configuration a few years ago. Both arrays had the same polarity and no delay.



Yes, the first-order length mode is cancelled. But the second-order mode is there with maximum gain. One could try to absorb only the second order mode (less rockwool needed) and cancel the first order mode with subwoofers. Both walls have to absorb, so rockwool should be on the front wall, too. I think this solution does not waste less space than a damped SBA.

I have to think about it in terms of maximum SPL, because doubling the number of drivers on the front wall gives +6 dB. a second array looks like onle +3 dB in comparison to the DBA. Of course a correct comparison would be against a room without back wall. Maybe I do a few BEM simulations...

I wonder if anyone has done this already and took measurements...
post #136 of 321
not that you need it too, but wouldn't it work even better if the rockwool were unpackaged/decompressed?

and wouldn't it help to get it out away from the wall and out of the pressure zone? I thought bass absorbers of this sort were catching and damping the soundwave in the velocity zone.
post #137 of 321
Thread Starter 
Quote:
Originally Posted by LTD02 View Post

not that you need it too, but wouldn't it work even better if the rockwool were unpackaged/decompressed?

It was just an experiment. The room is still a bit wet and must be sealed professionally. The final solution will be an lightweight construction with unpackaged absorbers.
Quote:
and wouldn't it help to get it out away from the wall and out of the pressure zone? I thought bass absorbers of this sort were catching and damping the soundwave in the velocity zone.

Yes, I also placed the 60 cm rockwool with 40 cm distance to the back wall and the first-order mode was completely gone. But this wastes too much space. I want to install two rows of seats.
post #138 of 321
Great thread.

I would use two arrays, one front and rear, with as much absorption on the front and rear walls as possible, and arrange seating to avoid the first even mode as best you can. The second even mode likely will be absorbed well with enough absorption.

JSS
post #139 of 321
thanks nils. I just wanted to ensure that I wasn't missing something.
post #140 of 321
Quote:
Originally Posted by FoLLgoTT View Post

Yes, the first-order length mode is cancelled. But the second-order mode is there with maximum gain.
A peak due to a length mode would be the same across your seating area. Peaks can be pulled down using EQ, benefiting all seats across that row. The alternative is to place the seating away from a peak or null.
post #141 of 321
Thread Starter 
Quote:
Originally Posted by sdurani View Post

A peak due to a length mode would be the same across your seating area. Peaks can be pulled down using EQ, benefiting all seats across that row. The alternative is to place the seating away from a peak or null.

This is only true for one seat row. If you have more than one the frequency responses differ. But assuming the owner sits in the first row one could equalize only that. Usually the guests are too impressed by the big picture to notice anything else. wink.gif
post #142 of 321
<-subscribed biggrin.gif
post #143 of 321
Quote:
Originally Posted by FoLLgoTT View Post

Both walls have to absorb...

Why? If the wave were completely absorbed at one wall nothing would make it to the opposite wall.
Quote:
Originally Posted by FoLLgoTT View Post

This is only true for one seat row.

And there are the side wall modes as well.
post #144 of 321
With a well implemented bass array (1/4 points on front wall in both dimensions), you have to go to the 2nd even mode to notice a difference between seats in one row. All odds are canceled, as well as first even in both horizontal (along row) and vertical directions. The only better setup would be 1/4 points floor and ceiling.....then only the even vertical modes would be left....

JSS
post #145 of 321
Thread Starter 
I did a few BEM simulations with ABEC and came to the conclusion that a multi sub configuration can not work for more than one seat row. I simulated with two identical arrays without delay and same polarity.

Let's take alook at the first-order length mode. With two subwoofer arrays is gets cancelled, because when one wave reaches the opposite wall the polarity of the second wave is inverted. Since the mode is cancelled, the efficiency in this frequency range ist the same as from a DBA. The second array only cancels the first-order mode, but doesn't give any additional gain.



The second order mode is a completely different beast. The polarity of both wave is inverted when they hit each other at 1/4 of the room length. This is always the case and completely independant of the wall's absorption coefficient. Even under free air condition both waves cancel each other at these positions (for a real sinus). For a time limited signal the first half-wave of the less distant array and the second half-wave of the more distand array are not cancelled.



Here you see the frequency responses at 1/2 (3 m) and 1/4 (4.5 m) of the room. Without absorption on the front and rear wall you get a strong peak at 1/2. On the opposite with absorber at the front and back wall there is a strong dip. But the seat rows always have different frequency responses. A DBA or a well-damped SBA is better in this regard.

Without absorption:


With frequency selective absorption (> 50 Hz):


Quote:
Originally Posted by noah katz 
And there are the side wall modes as well.

With all three configurations all seats in the same row share the same frequency response. Side and vertical modes are completely cancelled. The density of drivers in the arrays must be high enough, of course. My measurements confirmed the simulations.
post #146 of 321
Choosing the position of the rows carefully can make it work.

A true DBA has no room gain. A non-delayed DBA (all woofers receive same signal with no delay to rear array) cancels the first length mode and all odd modes) but maintains room gain. But the 2nd order length mode is left to be dealt with absorption or proper row location choice, so that one LF EQ solution yields a good result.

JSS
post #147 of 321
Quote:
Originally Posted by FoLLgoTT View Post

This is only true for one seat row. If you have more than one the frequency responses differ.
Sure, subwoofer placement is not going to solve all frequency response problems from row to row AND across all seats in each row. The best you can hope to do is maximize seat-to-seat consistency in order to make later equalization (auto or manual) easier. If an auto-EQ system, like Audyssey, sees very similar response in all seats, then it can correct more problems.

If you want better row-to row consistency, then move your front and back subs to the quarter points of room length, which will cancel the first 3 length modes. If you move those subs to the quarter points of room width as well, then the first 3 width modes will be cancelled as well, for better seat-to-seat consistency in each row. 4 subs at the quarter points of room length and width takes care of the first 3 length and width modes.

That's a pretty good head start, using placement alone, before even further tuning with treatments and EQ.
post #148 of 321
Thread Starter 
Quote:
Originally Posted by maxmercy View Post

A true DBA has no room gain. A non-delayed DBA (all woofers receive same signal with no delay to rear array) cancels the first length mode and all odd modes) but maintains room gain.

According to this article room gain exists only under 20 Hz in normal living rooms. Since most subwoofers have a high pass at 20 Hz there is no real benefit from room gain. Maybe I do a few measurements myself to check at which frequency the gain starts in my room (very solid stone walls).

Btw a damped SBA has room gain, too.
Quote:
Originally Posted by sdurani View Post

If you want better row-to row consistency, then move your front and back subs to the quarter points of room length, which will cancel the first 3 length modes. If you move those subs to the quarter points of room width as well, then the first 3 width modes will be cancelled as well, for better seat-to-seat consistency in each row. 4 subs at the quarter points of room length and width takes care of the first 3 length and width modes.

I know that this configuration works, but this is no option for me. One primary requirement of my home theater is a mxaimum integration of all electrical components. I want to see only the big picture and no loudspeaker. Placing the subwoofers near the seat rows is not useful to me. A DBA/SBA can be integrated much better and inside a baffle wall it is completely invisible.
post #149 of 321
Quote:
Originally Posted by FoLLgoTT View Post

Placing the subwoofers near the seat rows is not useful to me.
When I said quarter points of room length and width, I didn't mean placing the subs on the floor.
post #150 of 321
Thread Starter 
Quote:
Originally Posted by sdurani View Post

When I said quarter points of room length and width, I didn't mean placing the subs on the floor.

What did you mean? Shifting the complete array to 1/4 of room length? This is even less useful in real life. Imagine 18 subwoofers between the first seat row and the screen...



I don't thnik that placement alone is the answer. If the goal is a constant response at both seat rows, high max SPL and invisibility, there must be some sort of absorption or active cancelling.
Edited by FoLLgoTT - 8/31/13 at 1:32am
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: DIY Speakers and Subs
AVS › AVS Forum › Audio › DIY Speakers and Subs › Double Bass Array (DBA) - The modern bass concept!