In theory and in practice, if you put multiple drivers together as in a line array, you are increasing the surface area of the cones. That in itself should be enough to push more air as if one were using a singe larger driver. However, according to one speaker designg program, this configuration does not afford any lower frequencies as if it was a single large driver.
So, could some one tell me why this would be so?
Thanks in advance.
So, could some one tell me why this would be so?
Thanks in advance.
Interesting. In this case you are dividing the ‘work’ of producing a certain frequency at a certain spl over multiple drivers. Correct?
Regards,
Dan
Regards,
Dan
If you add cones to the room, you will get more SPL in all frequencies, assuming they are time & phase aligned for net summation. If they are within a quarter wavelength of each other, they can be considered co-located in bandwidth that is within that distance as a quarter wavelength. If you stack sub drivers high and place them at 1/4th distance from the horizontal plane on the wall and stack them so they cover top to bottom on this same well, or at least 1/4th points on that wall in vertical plane, you'll have effectively a single bass array too which is excellent for mode handling (except one axial mode).
Very best,
Very best,
Please help me understand this.
By combining identical drivers within 1/4 wavelength in time and phase of each other, you can effectively "add" all their frequencies. Correct?
But does that also mean that if the driver's lowest frequency is (for example) 40hz, then multiple drivers will effectively "lower" the lowest frequency because it you have increased the size of cones' surface area?
Is this correct?
Let's say you're using 12" drivers and they are stacked on top of each other with about 2" between each driver just for material room, etc, separate boxes even. The first and fourth drivers are farther away from each other than the rest. They're 28" away, plus 6" to the center of each, so let's say it's 38" center to center from the first and fourth driver cone. That's 0.96 meters. That's the quarter wave distance, so 4 x 0.96m = 3.84 meters. 343m/s / 3.84m = 89hz. So they're essentially one source from 10hz to 89hz. Above 89hz, each driver is not acting as a single source, but you get the idea. Extrapolate with real numbers in your intended array.
If you double cones, you get 3db in all bandwidth if they're the same cones.
If you increase overall SPL, then yes, you will get access to lower frequencies, as you build overall SPL to be higher, but also the frequencies above them too, the whole thing just moves up by Xdb based on the number of cones. Doubling cones = 3db. Double power = 3db. So if you double in power and cones over and over, each doubling of the two is 6db. That's only if they are time and phase aligned.
Just take a typical sealed response from 10hz to 200hz and look at it. Let's pretend 20hz is too low, well under F10. But let's say you went from that one sub to 16 of those subs, and you raised the entire response bandwidth SPL, same response, just more SPL, by 24db by going from 1 to 16 cones with the same power to each one. Your F10 from before is now +24db, so you could cut output above that value to flatten it to whatever lower frequency you want. This is how house curves are made.
It's not because you increased cone size, specifically, its because you raised overall SPL and so you can shape the response to whatever you want from the higher SPL to frequencies that were previously too low SPL to work with.
Very best,
True.
There is no mechanism in physics to make lower frequencies from multiple identical subwoofers. It will just make higher SPL.
This has all been extremely helpful. Thank you for all your contributions and explanations.
Given the above, I have more questions:
Is there anything in the Thiel/Small parameters which would indicate that one speaker is better suited over another for this kind of design?
More to the point, is there anything in the T/S which would indicate that one speaker is better "detailed" or does not store energy over another when used in this way?
Finally, given an Fs of a driver, does this necessarily mean that with x number of drivers increasing the overall spl of an array, that one could in theory (and in fact) increase the spl of the frequencies below the fs of the individual driver?
Given the above, I have more questions:
Is there anything in the Thiel/Small parameters which would indicate that one speaker is better suited over another for this kind of design?
More to the point, is there anything in the T/S which would indicate that one speaker is better "detailed" or does not store energy over another when used in this way?
Finally, given an Fs of a driver, does this necessarily mean that with x number of drivers increasing the overall spl of an array, that one could in theory (and in fact) increase the spl of the frequencies below the fs of the individual driver?
There's really not a T&S value that makes it meaningful for stacking subs like this. It's just a matter of total system design. I would think you'd want high sensitivity, high excursion, high power handling where you can for the least cost, when it comes to subs. And then the rest is based on design between sealed vs non-sealed geometry.
There's no T&S for detail.
The Fs is free air, the Fb is in the enclosure. And yes, if you build up enough cones and thus SPL, you comfortably command frequencies below Fb of the subs. My fleet of subs peak at 118db at 8~10hz. Way below Fs/Fb of any driver in my fleet. I cut everything above 10hz.
Very best,
There's no T&S for detail.
The Fs is free air, the Fb is in the enclosure. And yes, if you build up enough cones and thus SPL, you comfortably command frequencies below Fb of the subs. My fleet of subs peak at 118db at 8~10hz. Way below Fs/Fb of any driver in my fleet. I cut everything above 10hz.
Very best,
Again, thanks.
So my next question is slightly different: all things being equal, is it therefore possible to design a line array with the use of smaller speakers (ie smaller than 8" speakers)?
So my next question is slightly different: all things being equal, is it therefore possible to design a line array with the use of smaller speakers (ie smaller than 8" speakers)?
Yes of course. You could use any size. It's just a cone area vs SPL system. Smaller needs a lot more cones. Scale to your taste. And it depends on the bandwidth you want. If you want high sub-bass, it's gonna take a LOT of small cones though. A lot. I would use 10's or 12's as a starting point. But ideally bigger if you want bass.
Very best,
Last edited:
Example:
Standard 8 foot ceiling, 96 inches, you can fit 14x 6.5" drivers in a line from top to bottom with minimal spacing between them.
Cheap 6.5" woofers, with say 3mm xmax and that's it, 14 of them at 20hz at 1 meter:
96db.
So a 2nd line array gets you potentially up to 102db at 20hz at 1 meter. -4 to -6db at 2 meters depending on the room. Anda gain -4 to -6db at 4 meters. Extrapolate to your needs.
That's 28x6.5" drivers.
Compared to a handful of very cheap 10" drivers with moderate xmax. Let's look at 10" drivers, 10mm xmax.
8 x 10's gives you way more output down at 20hz fore example. Probably cheaper. And 8 of them comfortably fit in the 8 foot ceiling height. Two arrays of these would put you well into the one-teens at 20hz at 1 meter.
Very best,
Standard 8 foot ceiling, 96 inches, you can fit 14x 6.5" drivers in a line from top to bottom with minimal spacing between them.
Cheap 6.5" woofers, with say 3mm xmax and that's it, 14 of them at 20hz at 1 meter:
96db.
So a 2nd line array gets you potentially up to 102db at 20hz at 1 meter. -4 to -6db at 2 meters depending on the room. Anda gain -4 to -6db at 4 meters. Extrapolate to your needs.
That's 28x6.5" drivers.
Compared to a handful of very cheap 10" drivers with moderate xmax. Let's look at 10" drivers, 10mm xmax.
8 x 10's gives you way more output down at 20hz fore example. Probably cheaper. And 8 of them comfortably fit in the 8 foot ceiling height. Two arrays of these would put you well into the one-teens at 20hz at 1 meter.
Very best,
Yes it's done all the time but these aren't subwoofer arrays.
Here is the thing, reproduction of low frequencies indoors is a very different challenge than with higher frequencies. When the wavelength of the sounds being produced is very long.. often larger than the room dimensions, it becomes impossible to remove the room from the equasion. With higher frequencies you are listening to the direct radiating energy from the speakers, reflected energy can be absorbed if you like a more damped environment. It's generally not possible to do that with low frequencies however, the dimensions of the room and the position of the sub/s in it setup a complex frequency dependant standing wave pattern(nodes) that can only be changed by altering the subs position or the room dimensions. So in home listening environments sub clusters/arrays are generally not done because to get a more even in-room low frequency coverage multiple sub positions need to be used. Sub arrays are generally only found in larger outdoor professional applications where there are no boundaries to create nodes but there may be a need to steer the LF energy in one direction or another.
Woofers with high Qts sound boomy. But, majority of quality professional woofers have (very) low Qts, so no problem here.
No, just the SPL is higher - frequency response curve is the same but shifted up in SPL, with the same low frequency cutoff F3. No free lunch here.
Only with professional setups with big horn subwoofers, tight coupling of boxes will help.
