Destructively Driving Modes

dallasjustice

Member Sponsor
Apr 12, 2011
2,090
0
0
Dallas, Texas
#1
Is it really possible to simply "destroy" a room mode? I mean like I waived my magic wand and that nasty 13db peak at 24hz just disappeared into thin air? Yes, it's possible.

I wanted to show a quick measurement just to demonstrate how a very nasty room mode can be eliminated. Why would I call a mode "nasty?" IME, any room modes below about 35hz really cannot be reduced with any kind of room treatments. Okay, maybe if you built an enormous Helmholz resonator into your wall, it could be reduced. But anything else? Nope. There are NO off-the-shelf room treatments which will touch it.

One of two ways to deal with such a nasty mode is to use EQ. Of course, that can make a big difference. But I want the "best" solution on whatsbestforum. RIGHT?

I had some free time to measure my jl audio subs lately and I wanted to share something I think is pretty cool. I wanted to see how much I could reduce a 1st order axial room mode (one half wavelength) without using ANY DSP or room treatments. Normally, my system uses DSP, digital crossovers, four strategically placed subs for a total of 6 channels in stereo listening. The following measurement is NOT my normal system measurement. I say this upfront, because I know somebody will post something like; "man your system sux. I mean your bass is all over the place." :rolleyes: The bass measurements in my room are very flat at the seated position once everything is measured together with DSP, delay, digital crossovers and four mono subs. The following ONLY show two subs.

The concept of room mode cancellation is not new. All mathematical modeling has been done many years ago by Welti and written about extensively by Floyd Toole in his Sound Reproduction books. The following is a REW plot which shows three measurements in my room. I've labelled all measurements so the graph is pretty much self-explanatory. These are unfiltered measurements all at the same seated position with very close to the same level (I had to reduce the level to compensate for the two subs together measurement). There are only two subs used for this purpose (my system has a total of four subs). My room is 24' long. My room has a 1st order (half wave length) axial room mode at 24-25hz. All four of my subs exhibit the same 13db peak at 24-25hz when individually measured. My subs are spaced one quarter width on the front and backwalls. This is a "very good" configuration according to Welti's models.

To eliminate an odd order axial mode (eg. 1st axial mode), only a pair of subs used in opposite polarity together are needed. In this case, I used the front right and rear right subwoofers run in opposite polarity to demonstrate how this works.These measurements aren't using any EQ or time correction, so they would look MUCH better after correction. The 24-25hz 13db peak is totally eliminated when both subs are measured together in opposite polarity. That is, the axial mode is reduced by about 13db due to this combined subwoofer mode cancellation setup.

mode cancellation.jpg


I didn't post the decay plots, but things are very much cleaned up at 24hz. The time domain looks MUCH better there. Also, the even order modes weren't really increased using this method. If I had run both subs in the same polarity, the odd order mode would have been increased. That is, the 1st axial mode would be even bigger. I didn't post that plot but I did measure it; trust me I'm a lawyer. :D
 
Last edited:

dallasjustice

Member Sponsor
Apr 12, 2011
2,090
0
0
Dallas, Texas
#3
four subs vs. one sub; no DSP

Just for fun. The following REW plot shows four subs driven at the same time from the "very good" locations mentioned in the OP compared to a single subwoofer. Remember, I am not using any DSP or other EQ tricks. These are just four subs played together. I adjusted polarity and phase a little on three of the subs to get this response. Of course, I would get much better response with DSP. I am just showing what's possible without any DSP using four subs.

four subs vs. one sub.jpg
 
Last edited:

FrantzM

Member Sponsor & WBF Founding Member
Apr 20, 2010
6,469
0
0
#4
Multi subs is a practice that more audiophiles will benefit from. It is in most cases physically impossible to obtain ideal bass response in most residential rooms, even the most palatial. In my experience the subs need not be the same.. 4 <$500 subwoofer have the potential of providing more even bass response and likely higher SPL than a single $5K sub in most rooms.. Judicious use of DSP and EQ propels those cheapo subs to surprisingly good SQ and that with even (and compared to) full range speakers with deep bass reach and capabilities.. The immediate u pshoot is better clarity of sound overall.. throughout the entire spectrum, even the highs will be clearer!

I must add that I wouldn't have thought about reversing phase on one of the subs...
 
Jun 21, 2013
369
1
18
Wellington, New Zealand
#6
Taking it one step further, and assuming a rectangular room, if you were to place one sub in a corner and an out-of-phase sub in the diagonally opposite corner (ceiling mount), you could in theory kill multiple resonances.
 
Dec 13, 2010
253
1
18
#7
Active room mode treatment methods have been compared and ranked in

Fazenda et al., "Subjective Preference of Modal Control Methods in Listening Rooms", J. of the Audio Engineering Society 2012, S.338

Tests were done double-blind with music program material. The best contender was the 2-sub source-to-sink, subs on floor in centre position on front and rear walls, rear sub in opposite phase. No. 2 was CABS, i.e. 4-sub source-to-sink, all subs at 1/2 of room height, rear subs in opposite phase. No. 3 was single floor sub in centre position on front wall with amplitude EQ.

Klaus
 

dallasjustice

Member Sponsor
Apr 12, 2011
2,090
0
0
Dallas, Texas
#8
Thanks. I know Welti's most recommended four sub positions include all four subs at all four midwalls in a rectangular room. I can't really do that in room. It seems like the study you mentioned also recommends midwall position for two subs or front and back walls.


Active room mode treatment methods have been compared and ranked in

Fazenda et al., "Subjective Preference of Modal Control Methods in Listening Rooms", J. of the Audio Engineering Society 2012, S.338

Tests were done double-blind with music program material. The best contender was the 2-sub source-to-sink, subs on floor in centre position on front and rear walls, rear sub in opposite phase. No. 2 was CABS, i.e. 4-sub source-to-sink, all subs at 1/2 of room height, rear subs in opposite phase. No. 3 was single floor sub in centre position on front wall with amplitude EQ.

Klaus
 
Dec 13, 2010
253
1
18
#9
Thanks. I know Welti's most recommended four sub positions include all four subs at all four midwalls in a rectangular room.
Welti, "How Many Subwoofers are Enough?", AES paper 5602

Welti's study was concerned with optimal bass response (seat-to-seat variation) across a seating area with 16 chairs. The configuration you mention was best in the virtual room and calculated metrics. This configuration, however, was not included in the actual test room, with calculated and measured metrics. No listening was performed.

Here are some of his conclusions:

How many subwoofers are enough? Four subwoofers are enough
to get the best results of any configuration tried. Two subwoofers
at the wall midpoints is very nearly as good and has very good low
frequency support as well.

What is the optimal placement? One subwoofer at each wall
midpoint is the best in terms of Std, Max-ave and Max-min but
does not support low frequencies particularly well. Two
subwoofers, at opposing wall midpoints, performs very nearly as
well as four at the midpoints and gives a much better LF factor

In Fazenda the following methods were compared:

1.1 Subwoofer in Corner – No Equalization
1.2 Subwoofer in Corner – With Magnitude Equalization
1.3 Subwoofer in Front Wall Centre – No Equalization
1.4 Subwoofer in Front Wall Centre – With Magnitude Equalization
1.5 Controlled Acoustic Bass System
1.6 ‘Welti’ Configuration (2 subs wall midpoint front and rear, in phase)
1.7 Single Source-to-Sink
1.8 Opposite Phase-Opposite Corner

If interested in copies of those papers, drop me a mail.

Klaus
 

dallasjustice

Member Sponsor
Apr 12, 2011
2,090
0
0
Dallas, Texas
#10
Thanks Klaus. I've been curious about reading a variety of AES papers for a while. I bought a subscription so I can read them now. I appreciate the references. I want to learn more about the "sink" subwoofer setup used in the first paper you mentioned. The paper doesn't go into too much detail about how it works. Maybe I'll search the database for other papers on various subwoofer setups.

I'm particularly interested in this topic now because I've been agitating Bernt from Audiolense to improve his measurement procedure to account for setups which do destructively drive modes. He will release a new version which can and will do that. He said he's coding now so it may be a while before it's ready. If it works well, it should be a nice improvement to already excellent DSP software.

Michael.


Welti, "How Many Subwoofers are Enough?", AES paper 5602

Welti's study was concerned with optimal bass response (seat-to-seat variation) across a seating area with 16 chairs. The configuration you mention was best in the virtual room and calculated metrics. This configuration, however, was not included in the actual test room, with calculated and measured metrics. No listening was performed.

Here are some of his conclusions:

How many subwoofers are enough? Four subwoofers are enough
to get the best results of any configuration tried. Two subwoofers
at the wall midpoints is very nearly as good and has very good low
frequency support as well.

What is the optimal placement? One subwoofer at each wall
midpoint is the best in terms of Std, Max-ave and Max-min but
does not support low frequencies particularly well. Two
subwoofers, at opposing wall midpoints, performs very nearly as
well as four at the midpoints and gives a much better LF factor

In Fazenda the following methods were compared:

1.1 Subwoofer in Corner – No Equalization
1.2 Subwoofer in Corner – With Magnitude Equalization
1.3 Subwoofer in Front Wall Centre – No Equalization
1.4 Subwoofer in Front Wall Centre – With Magnitude Equalization
1.5 Controlled Acoustic Bass System
1.6 ‘Welti’ Configuration (2 subs wall midpoint front and rear, in phase)
1.7 Single Source-to-Sink
1.8 Opposite Phase-Opposite Corner

If interested in copies of those papers, drop me a mail.

Klaus
 
Dec 13, 2010
253
1
18
#11
Michael,

Thanks Klaus. I've been curious about reading a variety of AES papers for a while. I bought a subscription so I can read them now. I appreciate the references. I want to learn more about the "sink" subwoofer setup used in the first paper you mentioned. The paper doesn't go into too much detail about how it works. Maybe I'll search the database for other papers on various subwoofer setups.
Fazenda mentions 3 more sources that might be interesting in the respect:

[24] A. Celestinos and S. Nielsen, “Controlled Acoustic
Bass System (CABS)—A Method to Achieve Uniform
Sound Field Distribution at Low Frequencies in Rectangular
Rooms,” J. Audio Eng. Soc., vol. 56, pp. 915–931 (2008
Nov.).

[25] A. Makivirta and A. Varla, Sound Reproduction
Equipment and Method for Reducing the Level
of Acoustical Reflections in a Room - United States
Patent 6795557, application: 2000, published: 2004. URLhttp://
patent.ipexl.com/US/6795557.html.

[26] M.Wolff A. Goertz and L. Naumann, “Optimierung
der Wiedergabe von Surround Lautsprecheranordnungen
in Tonstudios und Abhorraumen”, Tonmeistertagung,
22, English translation available at http://forums.
klipsch.com/forums/storage/8/1485263/tmt2002 eng.pdf,
2001.

plus

Celestinos et al., “Optimizing placement and equalization of multiple low frequency loudspeakers in rooms”, Audio Engineering Society Preprint 6545 (2005)

Celestinos et al., “Low frequency sound field enhancement system for rectangular rooms using multiple low frequency loudspeakers, Audio Engineering Society Preprint 6688 (2006)

Celestinos et al., “Low-frequency loudspeaker-room simulation using finite differences in the time domain – Part 1: Analysis”, J. of the Audio Engineering Society 2008, S.772

Klaus
 

dallasjustice

Member Sponsor
Apr 12, 2011
2,090
0
0
Dallas, Texas
#12
Michael,



Fazenda mentions 3 more sources that might be interesting in the respect:

[24] A. Celestinos and S. Nielsen, “Controlled Acoustic
Bass System (CABS)—A Method to Achieve Uniform
Sound Field Distribution at Low Frequencies in Rectangular
Rooms,” J. Audio Eng. Soc., vol. 56, pp. 915–931 (2008
Nov.).

[25] A. Makivirta and A. Varla, Sound Reproduction
Equipment and Method for Reducing the Level
of Acoustical Reflections in a Room - United States
Patent 6795557, application: 2000, published: 2004. URLhttp://
patent.ipexl.com/US/6795557.html.

[26] M.Wolff A. Goertz and L. Naumann, “Optimierung
der Wiedergabe von Surround Lautsprecheranordnungen
in Tonstudios und Abhorraumen”, Tonmeistertagung,
22, English translation available at http://forums.
klipsch.com/forums/storage/8/1485263/tmt2002 eng.pdf,
2001.

plus

Celestinos et al., “Optimizing placement and equalization of multiple low frequency loudspeakers in rooms”, Audio Engineering Society Preprint 6545 (2005)

Celestinos et al., “Low frequency sound field enhancement system for rectangular rooms using multiple low frequency loudspeakers, Audio Engineering Society Preprint 6688 (2006)

Celestinos et al., “Low-frequency loudspeaker-room simulation using finite differences in the time domain – Part 1: Analysis”, J. of the Audio Engineering Society 2008, S.772

Klaus
Thanks Klaus! I've got some work to do.

I can see the benefit to using four subs. In particular, it's easy to see how out-of-polarity opposing wall setups can destroy a nasty room mode which isn't susceptible to room treatments.

My wheels have been turning now. I am trying to imagine how I would set things up in Acourate. I have a long email into to Uli now. I like to time/phase align my subs with my R/L speakers. However, if I run four subs (two out of polarity from the R/L speakers), it would be a mess to time/phase align all 6 six channels the the same crossover point with the R/L speakes. I am considering whether it would make more sense to crossover the out-of-polarity subs at a much lower frequency than the in-polarity subs. This makes sense to me because I could then easily time/phase align the in-polarity subs with the R/L at the crossover and not worry about time/phase aligning the R/L speakers with the out-of-polarity subs. This way, the out-of-polarity subs are mostly used to destroy the nasty mode and aren't really heard throughout the crossover region.


I'm just thinking . . . :D
 
Apr 3, 2010
16,022
0
0
Seattle, WA
#13
Thanks. I know Welti's most recommended four sub positions include all four subs at all four midwalls in a rectangular room. I can't really do that in room. It seems like the study you mentioned also recommends midwall position for two subs or front and back walls.
Two subs in middle walls is actually superior to four in corners from flatness point of view (at least that is what I recall from the study). The reason it is NOT recommended is that you lose a lot sub power when it is not reinforced in the corners.

Also all of Welti's recommendations rely on DSP as the idea is to cancel nulls, and let the peaks be managed by the DSP.
 

dallasjustice

Member Sponsor
Apr 12, 2011
2,090
0
0
Dallas, Texas
#14
Two subs in middle walls is actually superior to four in corners from flatness point of view (at least that is what I recall from the study). The reason it is NOT recommended is that you lose a lot sub power when it is not reinforced in the corners.

Also all of Welti's recommendations rely on DSP as the idea is to cancel nulls, and let the peaks be managed by the DSP.
Interesting Amir. Thanks.

Welti has a few different recommended setups. Obviously, midwall is best. In my setup the subs are one quarter width each on frontwalls and backwalls. They are not in the corners. I believe he refers to this setup as "very good." I would never try to integrate a sub without DSP. However, my objective is to use opposing wall sub placement to destroy a very low frequency mode, so that much less DSP is needed. It's my hope this will be a superior way to set the subs up. I don't really know yet since it may be a little more complicated once I try to time/phase align everything else as I mentioned in my previous post. We will see.
 

prerich

New Member
May 21, 2012
232
0
0
#15
Two subs in middle walls is actually superior to four in corners from flatness point of view (at least that is what I recall from the study). The reason it is NOT recommended is that you lose a lot sub power when it is not reinforced in the corners.

Also all of Welti's recommendations rely on DSP as the idea is to cancel nulls, and let the peaks be managed by the DSP.
When you say the middle wall, are you referring to the side walls or the middle of the front walls in the same plain of the main speakers?
 
Dec 13, 2010
253
1
18
#16
Amir,

Two subs in middle walls is actually superior to four in corners from flatness point of view (at least that is what I recall from the study). The reason it is NOT recommended is that you lose a lot sub power when it is not reinforced in the corners.
Keep in mind that Welti's goal was to optimize for a seating area, not a single listening spot:

"The first goal for low frequency optimization should be to minimize variation of frequency response from one location to the
next within the seating area."

"A secondary goal should be to maximize low frequency output of the subwoofers."

Klaus
 
Apr 3, 2010
16,022
0
0
Seattle, WA
#17
When you say the middle wall, are you referring to the side walls or the middle of the front walls in the same plain of the main speakers?
They are the same in theory. In practice I found in my room the two side walls worked better because I sat closer to the subs.

Just a quick theory when you put the sub in one corner, you have a null in the middle of the room. If you move the sub to the middle then by definition there is no null there. This takes care of that axis of your room. To get rid of the null in the other axis, you add a second sub and it will then fill in the center between the two subs. This all assumes perfectly rectangular room.
 
Apr 3, 2010
16,022
0
0
Seattle, WA
#18
Amir,



Keep in mind that Welti's goal was to optimize for a seating area, not a single listening spot:

"The first goal for low frequency optimization should be to minimize variation of frequency response from one location to the
next within the seating area."

"A secondary goal should be to maximize low frequency output of the subwoofers."

Klaus
That is true but the principal of mode cancellation works just as well for one seat or we would lose the sweet spot in such optimizations.

BTW, it is good to have you back Klaus. If there is someone who can give me a run for my money on these topics, it is you! :)
 

dallasjustice

Member Sponsor
Apr 12, 2011
2,090
0
0
Dallas, Texas
#19
That is true but the principal of mode cancellation works just as well for one seat or we would lose the sweet spot in such optimizations.

BTW, it is good to have you back Klaus. If there is someone who can give me a run for my money on these topics, it is you! :)
He's giving my tiny brain a pretty good workout too!
 

Nyal Mellor

Industry Expert
Jul 14, 2010
591
0
0
SF Bay Area, CA, USA
#20
To eliminate an odd order axial mode (eg. 1st axial mode), only a pair of subs used in opposite polarity together are needed.
Technical correction from acoustics geek...

There are two polarities here....

If you place the subs with same electrical polarity in opposite polarities of the room mode then you will drive it destructively. This is the best online resource that explains it: http://forum.blu-ray.com/showthread.php?t=48286. Also explained well in the Toole book.

I think what you did is different, which is what Klaus refers to as "Single Source-to-Sink" (though have not read that AES paper).

The above assuming both subs are run with no electrical delay. The Welti simulations are based on all subs fed same signal, no delay, same polarity.

With electrical delay (e.g. delay rear xx ms) then you are changing how the subs are driving the standing wave. Different delays will cause different effects. You can use this to your advantage along with individual sub level adjustments to "move" the peaks and dips in the room modes where you want them.

Some simulations: two subs on front wall (dots in middle of room are seats, this is a HT). You can clearly see the 20Hz 1st axial length room mode (low SPL center of room, high walls front and back)
20Hz.jpg

four subs, two on front wall, two on rears (rears delayed 2ms and run -3dB relative to fronts)
FINAL_20.jpg

One interesting thing about multi-subs is the debatable importance of time alignment. People like Geddes are purposely advocating NOT aligning the subs in order to control the room modes. The Welti methods also have no delay adjustments for individual sub channels relative to any listening position.
 

Members online

About us

  • Founded in 2010 What's Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals we enjoy learning about new things and meeting new people and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing