[John Casler]

 That is my thinking too. I've always thought that

1. Equal distance + equal phase = summation of response = increased amplitude

2. Equal distance + opposite phase = cancellation of response = decreased amplitude

This certainly is the case with two front subwoofers only. I obtained this graph by measuring my two front subs, one at a time, from the listening position, one with positive and one with inverted polarity (by switching speaker wires on the in phase and one out of phase (speaker wires swapped on right channel only). Listening with this setup with Ethan's Realtraps test tones and with music, it was very obvious that there was a marked reduction in SPL at the listening position compared with both subs wired with the same polarity, or even just one sub by itself -- confirming that cancellation was indeed occuring. This is analogous to what happens at the sides of dipole bass radiators.
 :

Ah Weedhopper ( ) but in the test, you are not sitting "in between" the subs.

You are sitting in the dipolar null at the side, not in between.

So what you are getting is all the crossfire room cancellation effects.

I never said that this would work running front subs out of phase with each other for they will cancel each other.  It has to be front and rear

See when you run the fronts only out of phase the left sub loads the room in one pattern and the right loads it in the opposite and they crossfire at the listening position with both their direct and indirect waves.  These will all cancel each other.

As I alluded to earlier, What actually causes room peaks and dips????

If you have a "SINGLE" sub playing on the middle of the front wall what happens?

The sound wave leaves the sub, passes the listener (where it is heard) then crashes into the rear wall, and bounces back at the rear of the listener.

Now when it reaches the listener and the waves coming from the front sub what happens?

It collides with these waves.  Does this collision cancel?  or does this collision amplify?

It cancels.  

So the wave continues on to the front wall, and again bounces off toward the listener.  As it passes the Sub, it "joins" what ever is coming from the sub and then makes it to the listener.

Now does this "joining" or "summing" cancel?  or does it amplify?

It amplifys.

Why is it so hard to get good deep bass in the exact center of the room?

Because we have walls.  If we didn't have walls and no reflected waves, the bass would be even and smooth all over.

Reducing the pressurization of the rear wall, by having a woofer fire "in unison" (out of phase so woofs move in the same direction in relationshipt to each other) with the front brings us closer to having no rear wall.

Additionally, it fires (if all is symetrical) a "mirror image" of all the modes/nodes of its counter part on the front wall, further equalizing overall response.

If this weren't so, my bass would be pitiful, not unbelievable.

Glad this has us all thinking.

[csero]

As the front woofer pushes, the rear woofer pulls.

That wouldn't nullify the bass, it would make it more efficient.
 ...

No, it would! Wavelengths much longer than the size of the head will always arrive unattenuated and 180 degrees out-of phase at either ear and therefore will always largely cancel. What you still hear is largely stray room reflections.

[John Casler]

As the front woofer pushes, the rear woofer pulls.

That wouldn't nullify the bass, it would make it more efficient.
 ...

No, it would! Wavelengths much longer than the size of the head will always arrive unattenuated and 180 degrees out-of phase at either ear and therefore will always largely cancel. What you still hear is largely stray room reflections.

Hi Frank,

How does that relate to having the rear pair of SUBS "out of phase" or "in phase"?

Are you saying it doesn't matter which?

[JohninCR]

OK,

I think I'm getting the picture and neither are running a delay.  It does give me more confidence that with my plan I'll be able to have my cake and eat it too.

First, let me explain my understanding of why the mirrored system with delay effectively cancels the rear wall reflections.  As the waves from front subs reflect off of the rear wall, a new directly out of phase wave is started and is travelling in the same direction, so the 2 sets of waves net effectively to zero and that mode is wiped out.  How the delay affects the other modes, I don't really understand, but I would expect an overall reduction the total level of bass in the room once that second out of phase wave is introduced, but triggering it affect the primary wave hits your ears shouldn't decrease the perceived bass except related to reflections.

Now with John & Jgubman's setups, it appears that you both are getting a dipole effect on the room modes.  Sitting between the 2 out of phase outputs seems tricky at the very least.  Your output in the seating position  is dramatically dependent upon placement and seating position.  If you used a symmetrical placement in a symmetrical room and put your head dead center, you would have a net zero response.  As you move away from the center line between the out of phase sources, you essentially increase the baffle width of your dipole and change the 6db/octave dipole acoustic crossover point.

I plan to have the seating area outside but on axis with this "quasi-dipole" sub.  I won't have any reduction in output, so my seating position should have the maximum SPL and anywhere behind the coffee table sub should enjoy the same response within about 3ft either side of the center axis, unlike sitting between the subs where SPL is much lower and position of both the chair and subs is so critical.  I think I will also have the same benefits related to room acoustics, but I'll be able to set my subs at a much lower level.

John, I have a favor to ask.  I start on my room this week but I won't build my subs until the end, and sound transmission is critical for me and the main consideration in construction.   This should be an interesting and simple test for you too.  I'm sure sound transmission is greatly reduced with your quasi dipole, but what I'm most interested in is if that goes for on axis as well.  Play some bass heavy music at significant volume, then go outside the room and listen, especially from the other side of the front and rear walls.  Then with the same volume setting, turn your rear pair off and listen again from outside the room.  Is what comes through that front and/or rear wall significantly different.  I'm pretty confident that what comes through the side walls should be drastically lowered with dipole.  I'm speculating that on axis through the front and rear walls will be somewhat lower due to the absence of pressure vessel gain in dipole mode, but not like the sides and ceiling that get dipole cancellation too.

[ekovalsky]

The sound wave leaves the sub, passes the listener (where it is heard) then crashes into the rear wall, and bounces back at the rear of the listener. Now when it reaches the listener and the waves coming from the front sub what happens? It collides with these waves. Does this collision cancel? or does this collision amplify?  It cancels.  

Not necessarily.  Whether that reflected wave cancels or amplifies depends on the frequency and distance from the wall.  At frequencies where the distance between the listening position and wall equals 1/4, 3/4, 5/4, etc of the wavelength there will be significant cancellation because that reflected wave will be opposite phase to the primary wave.  A null should occur at these frequences at the listening position. But at a frequency where the distance between the listening position and wall equals 1/2, 2/2, 3/2, etc, the reflected wave will be in phase with direct sound.  Then there will be summation, rather than cancellation, and peak should occur at these frequencies.

For wavelengths not at quarter or half multipliers there will be cancellation or summation to varying degrees.  

Reducing the pressurization of the rear wall, by having a woofer fire "in unison" (out of phase so woofs move in the same direction in relationshipt to each other) with the front brings us closer to having no rear wall.

That is the whole point of the double bass array.  It's goal is to minimize the sonic contribution from the rear wall reflections by creating cancellation of the front subs' output before it reaches the rear wall.  It seems where we diverge is the on the issue of delay.  Without it, I still believe the listening position will be located in a zone where there will be a lot of cancellation of the primary bass waves from all four subs.  With delay, the cancellation is occuring behind the listening position after the primary wave from the front subs has already been heard.

Whether or not you are currently sitting at a null should be pretty easy to determine with just a Rat Shack meter.   You can measure SPL at several bass frequencies where you sit, then repeat (without changing volume levels of course) after restoring your rear subs to positive polarity and again after disconnecting the rear subs completely.  I think the SPL will be highest with the subs all in positive polarity, then with the rears disconnected, and minimum with your current arrangement.   If your current arrangement gives the highest SPLs then my understanding of wave theory is incorrect.

As I've said before, sitting in a null probably flattens frequency response by cancelling the primary waves.  Also the peaks and nulls of the reflected sound from your front and rear subs, which are asymmetric and oriented differently, will be more evenly distributed since there are so many different paths into the room.  Without adding delay or a room correction device, your current arrangement probably will provide the best sound as long as you compensate for SPL loss in the cancellation zone.

By the way, the Behringer Ultracurve Pro DEQ2496 has a digital delay function.  It is a pretty sophisticated unit and can do pretty good room correction via parametric EQ.  It actually has some functionality that the TacT gear lacks, despite it costing less than 1/10th the price.  These units can be had for under $200 on Ebay.  May be worth picking one up and tring it out  

[youngho]

If you have a "SINGLE" sub playing on the middle of the front wall what happens?

The sound wave leaves the sub, passes the listener (where it is heard) then crashes into the rear wall, and bounces back at the rear of the listener.

Now when it reaches the listener and the waves coming from the front sub what happens?

It collides with these waves. Does this collision cancel? or does this collision amplify?

It cancels.

So the wave continues on to the front wall, and again bounces off toward the listener. As it passes the Sub, it "joins" what ever is coming from the sub and then makes it to the listener.

Now does this "joining" or "summing" cancel? or does it amplify?

It amplifys.

Actually, at the modal frequencies, the wavelengths hits the rear wall and reflect, now travelling the opposite direction but out of phase. Nulls or nodes represent areas where the polarity changes.

For example, take your original single subwoofer in the midpoint of the front wall and imagine it playing at the lowest modal frequency (565 divided by the length of the room). It moves forward, alternately compressing and rarefying the air. The midpoint of the room represents a node for this frequency, where polarity changes. At the midpoint of the room, the air doesn't move at all, and on the other side of the midpoint, the air compresses and rarefies in a manner opposite the driver's original motion. When the wave hits the back wall, it reflects, now travelling the opposite direction but in the opposite phase, meaning that it constructively interferes with the "forward moving" waves, resulting in addition (or amplification, if you prefer this term). When it hits the front wall, it reflects again, now travelling the original direction and in phase.

Whether the listener experiences a relative "amplification" or "cancellation" depends on their position in the room, i.e. whether they approach a node (a point at which polarity changes for given frequencies, experienced as nulls at those frequencies) or antinode (a point at which constructive interference is maximal for given frequencies, experienced as peaks at those frequencies).

That is my thinking too. I've always thought that

1. Equal distance + equal phase = summation of response = increased amplitude

2. Equal distance + opposite phase = cancellation of response = decreased amplitude

The interactions of multiple subwoofers depends on their locations, distances from the listener and boundaries, relative phases and polarities, relative loudness, and the particular frequencies involved. For example, place a second subwoofer at the midpoint of the rear wall and have it playing 180 out of phase with the first. This would result in reinforcement of the standing wave, which might be experienced as a greater relative peak or null, depending on the listener's position.

Incidentally, the Harman paper depends on mode cancellation, which uses a driver on each side of the node (preferable equidistant from the node) to disrupt standing wave formation, at least for the odd-order axial modes.

[csero]

As the front woofer pushes, the rear woofer pulls.

That wouldn't nullify the bass, it would make it more efficient.
 ...

No, it would! Wavelengths much longer than the size of the head will always arrive unattenuated and 180 degrees out-of phase at either ear and therefore will always largely cancel. What you still hear is largely stray room reflections.

Hi Frank,

How does that relate to having the rear pair of SUBS "out of phase" or "in phase"?

Are you saying it doesn't matter which?

You were talking about having the rear subs out of phase. Without delay what you have is 2 pair of woofers playing out of phase signal. It does not matter ( in an anechoic environment) were you are and in what direction you are facing, if the woofers are equidistant, you have cancellation. In a real room of course you have lots of other interferences, so you still hear the signal at decreased level, but it is room reflection mainly.

[John Casler]

Well just did a few (Quick and Dirty) experiments

Running the SUBs only, the readings from both, (much to my surprise was withing 1-2 db at frequencies from 63Hz to 20 Hz, in phase and out of phase.  The "in phase" was slightly higher, but the out of phase was as Jon found, smoother.

The "in phase" had a "peaky" reading at 50 Hz that was 6-db higher (than the out of phase reading.  Not sure what that was about.

These were all done without any RS corrections since the numbers weren't as important as the "differences".

I then did some listening tests with a couple very bassy ref cuts, and low and behold, I didn't hear as much difference as I thought I would.

Both sounded deep and detailed which is "very" surprising to me.  I thought the the "push/pull" would certainly have the edge, but to be honest, it was a toss up, during those couple cuts.

I guess the real deal is to get a quad sub array and play with it until you get it right.

John, I can't listen to the left and right side walls since the right side wall has a double wall, and I would be on the other side of it, and the left side wall is open in the front 1/2 and has a laundry room on the rear portion which would not allow me to seal it up from the bass coming from the main room and system.

However I was able to go in my bedroom (front wall) and try the system with the rear subs off and on.

Again, contrary to what I would have thought, it "sounded" (not measured) like it was louder with the rear subs "OFF".

This was not extreme, but I think noticable.

[ekovalsky]

youngho, in describing quarter and half wave phenomena, I was referring mainly to non-modal peaks and nulls.  True modal peaks and nulls will be as you describe.

csero, I agree completely.

By turning my listening position 90 degrees I can simulate have one sub up front and one in the rear.  When I inverted the phase to one of my front subs the other day, the effects at the listening position were the same whether I was hearing two subs in front (sitting normally) or one in front and one behind (me turned 90 degrees).  

I'll run some measurements with the subs in phase, out of phase with no delay, and out of phase with delay.   I can measure at an equidistant point (no delay) and a point somewhat further back (with delay).  Obviously I will only be able to do one channel since I don't have four subs.  Edit -- just realized that doing this will require rewiring my system so the TacT amp is driving the subs instead of the K2.  I'm going to wait two weeks, when I'll be auditioning a Blue Circle 204 amp which will drive the main channels instead of the TacT amps.  It is a major pain swapping the speaker cables behind my equipment stack, and I'll be doing it then anyway.

[JohninCR]

Csero,
I don't think it's primarily just reflections that John Casler is hearing.  Just like with a more normal dipole woofer, reflections are greatly decreased.  I believe the same rules will apply as with any dipole woofer and the phase relationship of the + & - waves at the listening position determines how much cancellation occurs, which is dependent upon wavelength and the difference in distance travelled by the 2 sets of waves.  In John C's setup, not only is he closer to the rear subs, but the rear subs have the added advantage of the wall reinforcing their output.

Ekovalsky,
I think you may have the 2 Johns mixed up.  John Casler does report lower output with one of his sub pairs out of phase.  My setup will be different with one sub (out of phase) at the front wall and one built into a coffee table directly in front of my seating position with a 4m distance between their originating points.  The point at which on axis (not between the subs) SPL will equal that of a single sub unit will be about 15hz, represented by the formula  Fequal= (.17 x 343m/s) / D  , where D is the difference in distance travelled by the + & - wavefronts.  SPL will increase up to a maximum of +6db where D equals 1/2 wavelength at 43hz.  Below 15hz I expect SPL to decrease due to dipole cancellation at 6db per octave.  I don't have to worry about the first deep dipole null in response where D equals 1 wavelength at 86hz because I plan to lowpass the subs at 28hz.  I also plan to use extreme absorption behind the sub at the front wall which should eliminate that boundary gain.  I want my subs set at the same level to retain a net in room output of zero.

With my setup out in a field, I would have 2 areas of maximum SPL.  They would be on axis behind each sub.  In my room, I will have only one, right at my listening position.

The only thing I can see being a possible hole in my plan is the effect of time.  At what amount of delay does our brain start to distinguish wavefronts as being separate at such low frequencies.  It must be wavelength related.   I'm pretty sure I'm OK at 12ms based on the 4m distance, since a 25ms group delay for a ported enclosure is a good target even with subs playing over an octave higher than mine will.

[JohninCR]

Thanks John,

Bass going into the bedroom with the rear subs off was just as I had hoped, louder with the rear subs OFF.

I'm having a hard time wrapping my brain around your in room results being so comparable in phase and out of phase.  Are you using identical subs and amps with all set to the same output level ?

I think I'm going to breakdown and build some test boxes first.

[youngho]

I don't think that John Casler's results are that surprising, considering that the acoustics of his room are significantly more complex than a rectangular room would predict, due to the differences in wall construction, openings, and diagonal front corner. One might predict a significant cancellation at the frequency (and multiples thereof) whose wavelength corresponds to the distance from the listener to the subwoofer, but this is probably be high enough that the loudspeakers are also (or may be solely responsible) for reproducing this frequency, as it seems relatively unlikely that the subwoofers are more than 14 feet away from the listener.

[JohninCR]

I may have just figured out why John C is getting close to the same SPL levels both in phase and out of phase.  I don't understand it, but according to the only dipole scientist I know of, Sigfried Linkwitz, a dipole requires only 1/3 of the acoustical power output to generate the same SPL level as a monopole speaker.    See the 4th paragraph below the figure with the red and blue circles  http://www.linkwitzlab.com/rooms.htm

This equates to a 4.8db advantage for the dipole, but in John C's case he's sitting between the 2 outputs which more than offsets that gain.

In my case, that 1/3 of the acoustical power to generate the same SPL is a clear advantage because it means that there's much less to be absorbed in my wall construction.  Plus at the sides I'll get the added benefit of the dipole cnacellation.  This gets more fun all the time and will probably save me alot of money because dipole soundproofs itself to a tremendous extent.

[JohninCR]

Youngho,

It's not the listener to sub distance that is important with the out of phase, dipole setup.  It is the DIFFERENCE in distance from the listener to the in phase and out of phase subs that is key, however, that is complicated further once you get off axis because there are 4 subs involved.  If his subs were all equally distant from the wall, then a central listening position would have almost zero SPL from the subs with them wired out of phase.  Dipole takes the irregularities in room shape out of the equation to a great extent.

[John Casler]

Thanks John,

Bass going into the bedroom with the rear subs off was just as I had hoped, louder with the rear subs OFF.

I'm having a hard time wrapping my brain around your in room results being so comparable in phase and out of phase.  Are you using identical subs and amps with all set to the same output level ?

I think I'm going to breakdown and build some test boxes first.

No problem, it was an interesting experiment.

As far as my room and results:  I have to admit, I am also amazed, as I thought it would be the other way around, but with all the things that can affect the room and low frequencies, you can never tell.

But I am still interested in what some are saying.

Look at this dipole illustration:

http://www.hometheaterhifi.com/volume_2_4/dipl1.html

Am I hearing some saying that the sound "inside" of, and between, the illustrations drivers would be less than inside the next bipole illustration.

http://www.hometheaterhifi.com/volume_2_4/bipl1.html

That would be interesting physics, but stranger things happen.

[John Casler]

 I don't think that John Casler's results are that surprising, considering that the acoustics of his room are significantly more complex than a rectangular room would predict, due to the differences in wall construction, openings, and diagonal front corner. One might predict a significant cancellation at the frequency (and multiples thereof) whose wavelength corresponds to the distance from the listener to the subwoofer, but this is probably be high enough that the loudspeakers are also (or may be solely responsible) for reproducing this frequency, as it seems relatively unlikely that the subwoofers are more than 14 feet away from the listener. ...

Hi Young,

While my room "appears" rectangular, 1/2 of the left side wall opens into an entryway that is not shown, and the left rear wall also has a floor to ceiling opening that goes into the kitchen.

Generally when I listen and test, I open my patio door on the rear wall (also not shown) which is directly behind the listening area, and the window to the patio which is above the right rear sub.

I sit about 11 feet from each sub, with the front subs being "off the front wall" as shown and the rear "on the rear wall" as shown.

I now intend to experiment more with "in and out" of phase since I expected to hear a more dramatic degradation in the "in phase" and it wasn't bad at all.

It did seem to have a slightly "hollow" or "spatial" quality, I want to explore further.

Just too many variables

[ekovalsky]

Look at this dipole illustration:

http://www.hometheaterhifi.com/volume_2_4/dipl1.html

Am I hearing some saying that the sound "inside" of, and between, the illustrations drivers would be less than inside the next bipole illustration.

http://www.hometheaterhifi.com/volume_2_4/bipl1.html

That would be interesting physics, but stranger things happen.

That's what I think.

In the bipole box, as the drivers move in and out together there are huge changes of pressure in box (assuming it is sealed) as its air is rarefied (drivers move out) or compressed (drivers move in).  If you put a microphone inside the box it would record large amplitude changes.

In the dipole box, as the one driver moves in and the other out, the volume of air inside the box doesn't change.  And the rear waves of each driver would largely cancel themselves out.  Microphone inside the box would record lower amplitude.  

Outside the box, such a dipole subwoofer will behave much like a single driver on an open baffle.  The physical separation of the drivers is small compared to the wavelengths being produced.  You will still get cancellation to the sides where the out of phase waves combine.  This restricted disperison helps cut down reflections, particularly from the side walls.

[JohninCR]

There shouldn't be any sound inside that dipole illustration as long as it's only playing low frequencies.

Each of miine will be bipoles then wire one out of phase for a net dipole.  I'm using bipoles because I need 2 sources of sound at the seating position to widen my sweet spot, so they will fire out of the sides of the coffee table, not front and back.  Bipole also gives you the benefit of vibration cancellation, which is very important for a coffee table.  I will, however, arrange my drivers in a front to back configuration, so I get the added benefit of distortion reduction of a push/pull alignment.

[John Casler]

That's what I think.

In the dipole box, as the one driver moves in and the other out, the volume of air inside the box doesn't change. And the rear waves of each driver would largely cancel themselves out. Microphone inside the box would record lower amplitude...

Hi Eric,

I didn't mean to have a "sealed" box, since the room is not "sealed", I was just using the illustration to show the directions of cone movements more clearly.

So what say you if the box is not sealed (most dipole subs I have seen are not)

[John Casler]

Just thought of another interesting thing about my room/set up.  The 10" woofs of the RM30s fire toward each other, which run full range, for whatever that is worth.