Cardioid Subs in a Smaller venue.

[Jay Barracato]

Re: Cardioid Subs in a Smaller venue.

No, you would then have the same effect in all directions.

BECAUSE you are missing the physical barrier between the air that each box does work on. This changes the acoustical impedence that Bennett refered to a bunch of posts back.

As to the nonideality of air, that is a function of the wave passing through the medium, not the wave properties themselves. While there is a randomizing effect at the extremes of the energy distribution, I would estimate it to be far less than other effects that tend to move air in nonwave motions (wind etc.)

 

[Ivan Beaver]

Re: Cardioid Subs in a Smaller venue.

Place your mike where the simulation shows max cancellation, or alternatively where you want to achieve max cancellation.

Or put another way-with anything you are doing.

WHAT DO YOU WANT TO DO-or WHAT AM I HERE TO DO?

Until THAT questions is answered-you can do all kind s of things-and not accomplish anything.

You HAVE to have a defined goal.

Even a "simple" question like "which speaker sounds better" means almost nothing. At what level? at what freq response-do the extreme highs or low matter? What aobut off axis? How far is "off" axis?

Sure you can get "answers" but they probably mean nothing.

 

[Bennett Prescott]

Re: Cardioid Subs in a Smaller venue.

Place your mike where the simulation shows max cancellation, or alternatively where you want to achieve max cancellation.

Except that you've just set up your physical spacing to have cancellation in a specific direction. I would set up my mic a few multiples of array height or depth behind it and align there. Then if I want cancellation elsewhere I'll spin the array! Otherwise you're working against yourself, you're not going to have much of an effect if you're measuring at the side of the array and wondering why you can't find a timing difference between drivers.

 

[Per Søvik]

Re: Cardioid Subs in a Smaller venue.

If you are more than one unit high or wide you can steer where you want the cancellation or the max output on the plane intersecting the multiple units. Then you put your "listening"-mike where you want to steer the cancellation, if steering is what you try to achieve. The measuring mike for SMAART or something like that obviously needs to be in the "correct" position for things like phase alignment etc.

 

[harrybrilljr]

Re: Cardioid Subs in a Smaller venue.

OK thanks for the explanation. I get what you are saying.

Harry,

As I understand the cardiod sub array, it relies on three factors:
1. The delay caused by the physical spacing between the forward facing drivers and the rearward facing drivers. (Approx. 1/4 wavelength)
2. The electronic delay added to the rearward drivers (approx. 1/4 wavelength)
3. The polarity inversion.

I would add that the subs themselves are important (as opposed to considering them as an idealized point source) because they act as a physical barrier between the air that the forward facing drivers act on and the air the rearward facing drivers act on.

My idea is just that two drivers that are close together compared to the wavelength they are producing both act on the same mass of air creating a single wave which then propagates according to wave theory. If two drivers are far enough apart or have some sort of barrier between them that causes each driver to act on a separate mass of air, then two waves are formed, begin propagating, and then can interfere. An often repeated but completely wrong misconception I hear frequently is that the interaction of two waves at one point and time somehow changes the waves at other points.

I would not suggest doing this in real life, but as a thought problem, could you set up an equivalent array with all three drivers facing forward, and the delay caused by the physical spacing replaced by the equivalent amount of electronic delay?

 

[harrybrilljr]

Re: Cardioid Subs in a Smaller venue.

These simulations are useless out of context. You must look at narrow band frequency and you must look at all the frequencies you expect the subs to reproduce. You also have to take into consideration the acoustic environment.

Of course, the position will vary a bit depending on array configuration. So measuring/listening where the array is expected to have most cancellation is probably the best for fine-tuning a particular set-up. Measuring the response out front makes sense for obvious reasons too 8)~~

10:7 cardio array

Double endfire, 1.36m 4 ms delayed fronts, 3m spacing axis to axis

Double cardio 1.36m 4ms delayed and phase inverted rears -3dB, 3 m spacing

3 to 1 cardio, 1.36 m 4 ms delayed and phase inverted rear, 1.5 m spacing

 

[harrybrilljr]

Re: Cardioid Subs in a Smaller venue.

BECAUSE you are missing the physical barrier between the air that each box does work on. This changes the acoustical impedence that Bennett refered to a bunch of posts back.

As to the nonideality of air, that is a function of the wave passing through the medium, not the wave properties themselves. While there is a randomizing effect at the extremes of the energy distribution, I would estimate it to be far less than other effects that tend to move air in nonwave motions (wind etc.)

There is no need for a physical barrier. In fact the physical barrier makes things more complicated. A set of ideal point sources will work perfectly in simulation.

 

[Bennett Prescott]

If you are more than one unit high or wide you can steer where you want the cancellation or the max output on the plane intersecting the multiple units. Then you put your "listening"-mike where you want to steer the cancellation, if steering is what you try to achieve. The measuring mike for SMAART or something like that obviously needs to be in the "correct" position for things like phase alignment etc.

You really need two independent arrays to do steering. The timing and directionality of your traditional 2-3 box cardioid array was set when you physically placed it, you measure to get the electrical to match the physical. Changing the timing is just going to give you a different pattern, which maybe you want, but doesn't really move the lobe around the way it would with a flat array and multiple delays.

 

[Jay Barracato]

Re: Cardioid Subs in a Smaller venue.

There is no need for a physical barrier. In fact the physical barrier makes things more complicated. A set of ideal point sources will work perfectly in simulation.

A pair of point sources work perfectly in simulation because the simulation is written to work perfectly with a pair of point sources.

Since no real source is a point source that argument doesn't go far. A number of years ago (2010?) someone did an experiment that lead to a fairly extensive discussion on the SynAudCom list about this phenonema. Once again, if two drivers are close enough to do work on the same air but are out of polarity (i.e. one driver pushes on the air while the other pulls), the net transfer of work to the air approaches zero, the impedence of each driver drastically drops, and the energy put into the driver by the amp is mostly converted to heat. At high levels this can rapidly burn up drivers which is why I proposed it as a thought problem and said don't do this for real. If I recall the original discussion, when the phenomena was first discussed, a wave property argument was proposed but that did not hold up to close analysis. Then the more engineering minded folks proposed the change in impedence argument. On the other hand, my background is more pure science than engineering, so I approached the problem as a conservation of energy problem, which lead to the explanation based on work that I have given.

If there is no need for a barrier or a larger physical separation, then you should be able to set up the classic three box array with all three drivers facing forward with appropriate delay and polarity. Practical experience shows this doesn't work; hence, the search for a deeper understanding.

 

[Per Søvik]

Re: Cardioid Subs in a Smaller venue.

These simulations are useless out of context. You must look at narrow band frequency and you must look at all the frequencies you expect the subs to reproduce. You also have to take into consideration the acoustic environment.

Anything is useless out of context, but it is you that are reading out of context, the context was that cancellation patterns will vary depending on configuration, and thus will point of maximum cancellation. Frequency makes less of a difference on the optimal side of the array and shouldn't change the location of your quiet or hot spot (until you consider walls and stuff). Obviously the environment is important, it can turn the quietest spot in an open air simulation into a noisy place

 

[Ivan Beaver]

Re: Cardioid Subs in a Smaller venue.

You really need two independent arrays to do steering. The timing and directionality of your traditional 2-3 box cardioid array was set when you physically placed it, you measure to get the electrical to match the physical. Changing the timing is just going to give you a different pattern, which maybe you want, but doesn't really move the lobe around the way it would with a flat array and multiple delays.

And as you adjust the timing-what may be "better" at one freq makes it a lot worse for another. Or a different listening position.

What I like to do is to setup multiple microphones (in front-behind-to the side, off axis etc.

Then start to adjust the electrical time and watch all of them (Smaart 7 is good for that sort of thing), and see the entire freq response at all the locations change. Then choose a "happy medium"

Then maybe you have to change the physical alignment and start over.

But if you don't look at the entire band of interest (and a bit on either side of it)-you really have no idea what you have. But for people who only like to look at one little aspect (or freq) then you can get some amazing results-even though they may not be that useful in the overall context of the actual production

 

[Per Søvik]

Re: Cardioid Subs in a Smaller venue.

You really need two independent arrays to do steering. The timing and directionality of your traditional 2-3 box cardioid array was set when you physically placed it, you measure to get the electrical to match the physical. Changing the timing is just going to give you a different pattern, which maybe you want, but doesn't really move the lobe around the way it would with a flat array and multiple delays.

Agreed, the more boxes and the more independent channels at your disposal, the better control. However, I was amazed at how much difference some steering made to a side stacked cardio setup with six boxes a side when a little bit of delay to the inside stacks turned the lobe slightly away from the walls and got rid of a few dB of reflection. Didn't help half as much as turning the sub stacks down nearly 10 dB though ~;-)~:wink:

Edit: Sometimes, digital steering can be a place to start before rearranging physically.

 

[harrybrilljr]

Re: Cardioid Subs in a Smaller venue.

If there is no need for a barrier or a larger physical separation, then you should be able to set up the classic three box array with all three drivers facing forward with appropriate delay and polarity. Practical experience shows this doesn't work; hence, the search for a deeper understanding.

I guess your idea of a large physical separating and mine are two totally different things. I do not consider 2.75ft a large physical separation yet it is enough to make this work, barrier or not. Perhaps there is a language gap here. You are speaking science and I am speaking dumb soundguy. You are using math and I am using my ears.

 

[harrybrilljr]

Re: Cardioid Subs in a Smaller venue.

Anything is useless out of context, but it is you that are reading out of context, the context was that cancellation patterns will vary depending on configuration, and thus will point of maximum cancellation. Frequency makes less of a difference on the optimal side of the array and shouldn't change the location of your quiet or hot spot (until you consider walls and stuff). Obviously the environment is important, it can turn the quietest spot in an open air simulation into a noisy place

Obviously you didn't look at the PDF I attached.

 

[harrybrilljr]

Re: Cardioid Subs in a Smaller venue.

And as you adjust the timing-what may be "better" at one freq makes it a lot worse for another. Or a different listening position.

What I like to do is to setup multiple microphones (in front-behind-to the side, off axis etc.

Then start to adjust the electrical time and watch all of them (Smaart 7 is good for that sort of thing), and see the entire freq response at all the locations change. Then choose a "happy medium"

Then maybe you have to change the physical alignment and start over.

But if you don't look at the entire band of interest (and a bit on either side of it)-you really have no idea what you have. But for people who only like to look at one little aspect (or freq) then you can get some amazing results-even though they may not be that useful in the overall context of the actual production

Thanks you for that. I think the pdf shows this.

 

[Jay Barracato]

Re: Cardioid Subs in a Smaller venue.

I guess your idea of a large physical separating and mine are two totally different things. I do not consider 2.75ft a large physical separation yet it is enough to make this work, barrier or not. Perhaps there is a language gap here. You are speaking science and I am speaking dumb soundguy. You are using math and I am using my ears.

You are missing the OR and the fact that the physical box the driver is in is the barrier I am talking about.

2 drivers 2.75 feet apart cannot make two waves at low frequencies unless there is something separating the air those drivers act upon, like the sub itself. They must make a single wave. There is no way to arrange 2 drivers 2.75 feet apart in a cardiod array without the boxes those drivers are in creating a barrier, so I don't understand what you mean by "make this work" or how you could hear it.

Regardless of what you interpret your ears as telling you, two drivers acting on the same air mass, close enough to couple, but out of polarity are not an example of wave cancellation.

Once again, I would go back to the practical proof, if explanation of the theory is not enough for you. If what you are saying is true, then you should be able to build an array with 2 drivers in close proximity facing the same direction by just applying electronic delay and polarity change. As far as I know that can not be done.

 

[Bennett Prescott]

Re: Cardioid Subs in a Smaller venue.

Agreed, the more boxes and the more independent channels at your disposal, the better control. However, I was amazed at how much difference some steering made to a side stacked cardio setup with six boxes a side when a little bit of delay to the inside stacks turned the lobe slightly away from the walls and got rid of a few dB of reflection. Didn't help half as much as turning the sub stacks down nearly 10 dB though ~;-)~:wink:

Edit: Sometimes, digital steering can be a place to start before rearranging physically.

Sorry, I thought we were talking about adjusting the timing within a single cardioid array "cell" in order to move lobes, which will work very poorly. Adjusting delays between a pair of "point sources" that happen to be cardioid arrays themselves will naturally work perfectly well, within reason.

 

[Bennett Prescott]

Re: Cardioid Subs in a Smaller venue.

If what you are saying is true, then you should be able to build an array with 2 drivers in close proximity facing the same direction by just applying electronic delay and polarity change. As far as I know that can not be done.

I have never tried a cardioid array this close, but I don't particularly see why it couldn't be done. All cardioid arrays lose LF energy since below a certain frequency the woofers are too close and sum coherently - but of course one is out of polarity so eventually they are so close that all that energy gets turned into structure movement and heat. This is a pain to measure with most arrays since the wavelengths are so long and the maximum frequency of interest is usually kept reasonably low, but you can easily see it in a simulation that doesn't have a HPF (validity of simulation notwithstanding). Certainly some companies have products that work this way, essentially processing each driver in a "dual 18" cab differently to create directionality along the long axis of the subwoofers. Means you can pack face to face too which gets really great summation.

The only real difference IMHO having them back to back, or front to back, is the additional delay created by the waves having to travel around the front cab means that you can pack the subs tighter together than might otherwise be indicated (e.g. the sub stack, bottom sub reversed). The subs certainly aren't big enough to act as a boundary in the normal sense, but they are a big thing in the way! Either way you get what you need, separation in the axis in which you want to create directionality, and different processing - get them too close and you're just going to have maximum addition and cancellation at a higher frequency than you might like it at.

 

[Per Søvik]

Re: Cardioid Subs in a Smaller venue.

Obviously you didn't look at the PDF I attached.

Me looking at that pdf won't change the physics of cancellation/addition in the direction where this is only a relationship between time and distance. If a cardioid is correctly aligned, cancellation behind is not frequency dependant at the point of maximum cancellation. Same with the addition in front of an end-fire array, time and distance rules, and frequency only count where time and distance doesn't relate.
The thing is, we don't actually disagree about anything at all (you'll probably beg to differ), I just don't think it is necessary to tell the whole story all the time and consider all angles all the time.

 

[Jay Barracato]

Re: Cardioid Subs in a Smaller venue.

Argh, what is http://soundforums.net/ajax.php. My computer has eaten my answer four times and left me on this blank page without posting.

All cardioid arrays lose LF energy since below a certain frequency the woofers are too close and sum coherently - but of course one is out of polarity so eventually they are so close that all that energy gets turned into structure movement and heat.

This is the point I have been trying to make. In order to have wave cancellation, you must have two separate waves. Where the drivers couple, only one wave is formed. If one of the two drivers is out of polarity, most of the energy is not converted into wave form, i.e. the driver does no or little work on the air. I am using WORK in the physics meaning.

The ony real difference IMHO having them back to back, or front to back, is the additional delay created by the waves having to travel around the front cab means that you can pack the subs tighter together than might otherwise be indicated (e.g. the sub stack, bottom sub reversed).

This seems to support what I amtrying to say as well, but I am also trying to add that I believe the physical bulk of the speaker box allows you to position them closer together without coupling.

The subs certainly aren't big enough to act as a boundary in the normal sense, but they are a big thing in the way! Either way you get what you need, separation in the axis in which you want to create directionality, and different processing - get them too close and you're just going to have maximum addition and cancellation at a higher frequency than you might like it at.

I am not talking about a boundary relative to a wave. I am talking about a barrier between the masses of air each driver does work on. I believe this physical barrier allows the drivers to be closer together without coupling. While the distances at which coupling occurs is dependent on wavelegth, in physics terms coupling is not a wave function based on multiple waves.

I have never tried a cardioid array this close, but I don't particularly see why it couldn't be done.

I have been trying in DDT and still can't get anything to work. Either the drivers are in polarity and couple, or out of polarity and completely cancel.