If the drivers were acoustically in phase then they are both bipoles irregardless of the orientation of the 2nd driver. A diple would have all the same bass issues as a single driver on an open baffle.
dave
My experience matches this B&W whitepaper:
If your volume displacement requirement can be met by a single woofer up to 15" - 18", this is the best sounding solution. I have not built a PPSL 2-woofer counter-force design, but I guess the single slot offers the same coherent wavefront advantage plus higher efficiency and displacement.
B&W Whitepaper on Bass
Advantages of Producing a Coherent Wavefront
One remarkable fact was consistently noticed during the development of the new 380mm bass unit for the Nautilus™801, and that was that a single large and stiff bass cone always sounded better than a number of smaller cones, even though they may well have had the same aggregated properties. One possible explanation for this is the concept of the production of a “coherent wave front”. This will be produced by a single large very stiff cone, which can couple with the air in a uniform manner over the whole of its surface area unaffected by differences in loading over that area. This behaviour is to be compared with that of several cones which, even though they may be closely spaced, will still leave gaps of “uncoupled air” between them. The very stiff cone material of the large single driver, which is a thick sandwich of Kevlar® reinforced paper fibres with a very stiff skin, makes it less responsive to local changes of acoustic impedance or unbalanced modal pressures either behind or in front of the cone.
A “coherent wavefront” simply means there is either a constant or a smoothly changing phase relationship between neighbouring parts of the wavefront. So even if two drivers are relatively close together compared to a wavelength. Also, even if their contributions are equivalent to a single large driver, their different acoustic environments will mean that their outputs are slightly different, in terms of both amplitude and phase.
Furthermore, the air between the drive units is not being driven at all and this will translate into a change in phase across the resulting wavefront as the air tries to “fill in” the lost contribution. One can postulate that, at low frequencies, air can “spill off the edges” of the individual cones more easily in an array of small cones, which obviously have more edges for it to spill off, than from a single large one. For instance, two 12 inch drivers have a combined circumference of 1630mm whereas one 15 inch cone has a circumference of only 1037 mm. It is also interesting to note in this context that the radiation resistance and reactance at low frequencies of one 15 inch cone, is actually larger than that of two 12 inch cones, even though the area of the 15 inch cone is 0.02m2 smaller than two 12 inch cones. This is because the change from a steadily rising radiation resistance characteristic at low frequencies, to a horizontal one at high frequencies, occurs at a lower frequency with one large driver than with two smaller ones. Mutual radiation impedance effects will redress this imbalance to some extent, provided that the two individual drivers are close enough together for one driver to acoustically load the other, though it will only be totally redressed if the array of small drivers produces a totally contiguous surface in all directions.
If your volume displacement requirement can be met by a single woofer up to 15" - 18", this is the best sounding solution. I have not built a PPSL 2-woofer counter-force design, but I guess the single slot offers the same coherent wavefront advantage plus higher efficiency and displacement.
B&W Whitepaper on Bass
Advantages of Producing a Coherent Wavefront
One remarkable fact was consistently noticed during the development of the new 380mm bass unit for the Nautilus™801, and that was that a single large and stiff bass cone always sounded better than a number of smaller cones, even though they may well have had the same aggregated properties. One possible explanation for this is the concept of the production of a “coherent wave front”. This will be produced by a single large very stiff cone, which can couple with the air in a uniform manner over the whole of its surface area unaffected by differences in loading over that area. This behaviour is to be compared with that of several cones which, even though they may be closely spaced, will still leave gaps of “uncoupled air” between them. The very stiff cone material of the large single driver, which is a thick sandwich of Kevlar® reinforced paper fibres with a very stiff skin, makes it less responsive to local changes of acoustic impedance or unbalanced modal pressures either behind or in front of the cone.
A “coherent wavefront” simply means there is either a constant or a smoothly changing phase relationship between neighbouring parts of the wavefront. So even if two drivers are relatively close together compared to a wavelength. Also, even if their contributions are equivalent to a single large driver, their different acoustic environments will mean that their outputs are slightly different, in terms of both amplitude and phase.
Furthermore, the air between the drive units is not being driven at all and this will translate into a change in phase across the resulting wavefront as the air tries to “fill in” the lost contribution. One can postulate that, at low frequencies, air can “spill off the edges” of the individual cones more easily in an array of small cones, which obviously have more edges for it to spill off, than from a single large one. For instance, two 12 inch drivers have a combined circumference of 1630mm whereas one 15 inch cone has a circumference of only 1037 mm. It is also interesting to note in this context that the radiation resistance and reactance at low frequencies of one 15 inch cone, is actually larger than that of two 12 inch cones, even though the area of the 15 inch cone is 0.02m2 smaller than two 12 inch cones. This is because the change from a steadily rising radiation resistance characteristic at low frequencies, to a horizontal one at high frequencies, occurs at a lower frequency with one large driver than with two smaller ones. Mutual radiation impedance effects will redress this imbalance to some extent, provided that the two individual drivers are close enough together for one driver to acoustically load the other, though it will only be totally redressed if the array of small drivers produces a totally contiguous surface in all directions.
Attachments
Push-Pull is used to indicate one driver facing out, 1 facing in, acoustically wired in phase (electrically out of phase). A bipole as you have said.
My SDX10 PPP is also a bipole, drivers are wired in phase(electrically & acoustically -- ie both push out at the same time). Wired acoustically out of phase (dipole) you would have almost no bass.
Are your arrows indicating how you wired them acoustically or the orientation of the woofer cone?
dave
Then it isn't push-pull, it is a dipole, and you would expect huge difference because the bass would be cancelling.
dave
Almost all of that is only relevant at higher frequencies than a sub works at. The 2 woofers would need to be > 1/4 wl c-c to start causing phase issues. At 100 hz, that is 33 inches.
dave
Very interesting posts here and lots for me to digest!
Even if I fixed the pair of drivers together, back-to-back, very securely probably at both the magnet structures and at the bolt holes in the drivers' frames (using part-threaded rods or similar)?
The B&W whitepaper on bass reproduction gives food for thought. The comparison though seems to be mainly between a single driver vs. multiple drivers with approximately the same (total) radiating areas.
I wonder if some of B&W's arguments wouldn't apply if the multiple driver sub had massively more combined diaphragm/cone area than the single driver sub? I can understand that 'air spill' from cones will increase with increasing cone circumference (or combined circumferences) but I'd guess that 'air spill' will also become more significant with increased cone excursion - something that's likely to happen with a single driver sub.
I take B&W's point about a coherent wave front and spaces between multiple drivers. I had thought of mounting the multiple drivers from inside the cabinet then chamfering the outside of each hole in the cabinet - sort of like the flare often seen at the room end of reflex ports (or to look at it another way, a very, very short exponential horn) - so that it matched up with the adjoining one. I could increase the cabinet wall thickness at these points to give me more wood to work with. Again though, I suspect the gaps between multiple drivers becomes less significant as cone excursions reduce (which happens when radiating area is very large).
Also, B&W's paper is about bass and sub-bass. I wonder if their comments would apply to sub 80Hz only?
Even if I fixed the pair of drivers together, back-to-back, very securely probably at both the magnet structures and at the bolt holes in the drivers' frames (using part-threaded rods or similar)?
The B&W whitepaper on bass reproduction gives food for thought. The comparison though seems to be mainly between a single driver vs. multiple drivers with approximately the same (total) radiating areas.
I wonder if some of B&W's arguments wouldn't apply if the multiple driver sub had massively more combined diaphragm/cone area than the single driver sub? I can understand that 'air spill' from cones will increase with increasing cone circumference (or combined circumferences) but I'd guess that 'air spill' will also become more significant with increased cone excursion - something that's likely to happen with a single driver sub.
I take B&W's point about a coherent wave front and spaces between multiple drivers. I had thought of mounting the multiple drivers from inside the cabinet then chamfering the outside of each hole in the cabinet - sort of like the flare often seen at the room end of reflex ports (or to look at it another way, a very, very short exponential horn) - so that it matched up with the adjoining one. I could increase the cabinet wall thickness at these points to give me more wood to work with. Again though, I suspect the gaps between multiple drivers becomes less significant as cone excursions reduce (which happens when radiating area is very large).
Also, B&W's paper is about bass and sub-bass. I wonder if their comments would apply to sub 80Hz only?
"Then it isn't push-pull, it is a dipole, and you would expect huge difference because the bass would be cancelling.
dave "
yes! it is a dipole, and yes, you would expect cancelling, but...have you ever tried it and measured it? well, I did, and I posted what I got, and it surprised me too, because I expected loss of bass, on the contrary, I got higher spl in low frequencies, but it seems you did not read my post on the link, never mind
dave, now I understand what you saying, but you are mixing when one drivers orientation is turned around...and is in oposite phase
never mind, we do not seems to be talking about the same thing
dave "
yes! it is a dipole, and yes, you would expect cancelling, but...have you ever tried it and measured it? well, I did, and I posted what I got, and it surprised me too, because I expected loss of bass, on the contrary, I got higher spl in low frequencies, but it seems you did not read my post on the link, never mind
dave, now I understand what you saying, but you are mixing when one drivers orientation is turned around...and is in oposite phase
never mind, we do not seems to be talking about the same thing
I didn't have time to go rea it till now. Very counter intuitive and counter to my experience. Almost makes me think you had the switch backwards.
An easy enigg experiment when i get the amps for my SDX10 woofers.
dave
"Almost makes me think you had the switch backwards."
haha, highly unlikely
in bipolar mode it creates big impact on the box, I could feel the box vibrating, and holding cones, I could feel them moving in oposite direction
in dipole mode the box is pretty much an isobaric chamber, so no presure changes, no vibration, and putting hands on the drivers felt like one piston working together
as I said, I heard much better bass in dipole mode therefore I measured it, no mistake was made, that is what I got, I believe the size of the box must be substantial for the effect to be seen, if the box is narrow, there may be substantial cancelation
my explanation is that the cones in dipole mode and not slowing each other, fighting with each other, hence the gain in low frequencies...the situation might be different at other frequencies, but I only measured the the range used by the subwoofer, since I use active crossover up to ~170Hz. I will built two more subwoofers like this, with two 15" woofers on oposite sides, with an option to make it bipole or dipole, and see if there is the same effect
haha, highly unlikely
in bipolar mode it creates big impact on the box, I could feel the box vibrating, and holding cones, I could feel them moving in oposite direction
in dipole mode the box is pretty much an isobaric chamber, so no presure changes, no vibration, and putting hands on the drivers felt like one piston working together
as I said, I heard much better bass in dipole mode therefore I measured it, no mistake was made, that is what I got, I believe the size of the box must be substantial for the effect to be seen, if the box is narrow, there may be substantial cancelation
my explanation is that the cones in dipole mode and not slowing each other, fighting with each other, hence the gain in low frequencies...the situation might be different at other frequencies, but I only measured the the range used by the subwoofer, since I use active crossover up to ~170Hz. I will built two more subwoofers like this, with two 15" woofers on oposite sides, with an option to make it bipole or dipole, and see if there is the same effect
On a similar, but not identical note, here's a Celestion design from the eighties. As you say, box pressure doesn't change when both drivers are mounted on opposite sides of the cabinet and move in the same absolute direction so Celestion did away with the box!!!
Celestion System 6000 loudspeaker system | Stereophile.com
That's the sort of thing I was thinking of but yes, maybe a lot of effort for little reward.
As a fun project, I am converting a 3x3x3 ft part of an overhanging duplex above my music room into a subwoofer. It will have 8 10" subwoofer drivers connected in parallel (1 ohm final impedance) being driven by a set of paralleled LM3886 (8 of them). The sub will be fully sealed and stuffed with rock-wool, and the walls lined with closed cell foam.
It would have also taken 4 12" or 15" drivers, but the 10" were cheap and available on a try-before-you buy-basis, so the worst that can happen is that the woodwork will need to be redone if the project hits a flat note. I can always put in some slats and make it into a storage alcove
I am hoping with the solid construction (5 walls are brick and mortar), ceiling reinforcement and frugal outlay, I am able to achieve reasonable performance from the project. The biggest argument favouring this is the SPL, at 82dB per sub as per my estimates, eight units will be 124dB/1m, and I might be able to get about 110dB at listening position with 1 watt of power, not something to sneeze at.
The biggest issue is the baffle, with its massive size and resultant tendency to flex. I am thinking 2 layers of MDF sandwiched together, and cross-braces after every row/column of drivers in both directions. Lobing is a valid concern, though according to my earlier research it's more about higher frequencies, than lower ones. I do understand that once you spread drivers over a 3x3 ft area it's no longer a single box but multiple point sources. Not wondered enough about this I guess, but with drivers spaced 3' apart (not accounting for the ones in between yet) you could cross the subwoofer around 180Hz and not have lobing issues.
It would have also taken 4 12" or 15" drivers, but the 10" were cheap and available on a try-before-you buy-basis, so the worst that can happen is that the woodwork will need to be redone if the project hits a flat note. I can always put in some slats and make it into a storage alcove
I am hoping with the solid construction (5 walls are brick and mortar), ceiling reinforcement and frugal outlay, I am able to achieve reasonable performance from the project. The biggest argument favouring this is the SPL, at 82dB per sub as per my estimates, eight units will be 124dB/1m, and I might be able to get about 110dB at listening position with 1 watt of power, not something to sneeze at.
The biggest issue is the baffle, with its massive size and resultant tendency to flex. I am thinking 2 layers of MDF sandwiched together, and cross-braces after every row/column of drivers in both directions. Lobing is a valid concern, though according to my earlier research it's more about higher frequencies, than lower ones. I do understand that once you spread drivers over a 3x3 ft area it's no longer a single box but multiple point sources. Not wondered enough about this I guess, but with drivers spaced 3' apart (not accounting for the ones in between yet) you could cross the subwoofer around 180Hz and not have lobing issues.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.