[guest60106]

Any predictions on how this woofer module would behave? The design intent is to make an OB bass module that utilizes 4 10” woofers that is as narrow as possible (front to back) and have the exact same output (spl) front to back. That intent yielded a dual U-frame each utilizing 2 woofers that are reversed and stacked. The 2 units (acting as one) would be driven out of phase so that the cones all move in the same direction.  I am sure I am not the first person to try this. I just have not seen this configuration before. I am having trouble conceptualizing the resultant output of this design. 2 cardioids back to back equaling a dipole? A hybrid h-frame? A hybrid u-frame? Nothing?

This is a design concept so please ignore lack of footers and bracing and other missing artifacts. I am just pondering the sonic attributes/pitfalls of the design.

Matthew

[MJK]

I would have all the woofers facing the same way. I think you are adding complexity without any real benefits by facing the pairs of woofers in opposite directions. You might consider building each driver in its own U/H frame so you can add or delete woofers to mate with different efficiency and types of drivers for the mids and treble.

[Toaster]

It's essentially a shallow U-frame, FWIW. Depending on where you draw the line it might just be considered a fairly thick folded I-frame i suppose...

[guest60106]

I would have all the woofers facing the same way. I think you are adding complexity without any real benefits by facing the pairs of woofers in opposite directions.

The original design intent for the opposite facing sections is to match the SPL front to back. I have read that this idea has little benefit in OB design. Also, the depth is about 6in (15.25 cm). It is not as shallow as it looks.

[guest60106]

Here is a section view if it helps define the design.

[JohnR]

I am having trouble conceptualizing the resultant output of this design.

Hi Matthew, I think you'll find that basically behaves like a dipole. It's not deep enough to be a U frame and in any case I believe a U-frame needs to be stuffed to make a cardioid.

as narrow as possible (front to back)

I'm wondering why you want to do that?

[guest60106]

I'm wondering why you want to do that?

It's Simple. The speaker needs to be as far as away from the wall as possible. So if the back of the speaker is 3 feet from the wall, it will be less obtrusive if it is  6" thick than if it is 14" thick.

[AJinFLA]

Any predictions on how this woofer module would behave?

At LF...like a dipole. The ducts are short/symmetric and the acoustic centers are closely aligned front-back. It will behave, at LF, like a fairly symmetric dipole.

I am sure I am not the first person to try this.

Maybe, maybe not. The ones I did had only 2 woofers (vertically). Things that wide (like yours) don't fly in my household .

cheers,

AJ

[THROWBACK]

IIRC, the original idea behind the woofer-out/woofer-in concept was actually to reduce distortion, although I have never seen any data to support this.

Also, I use stacked woofers (3 on a side, GR Research 12" servo's). This results in a smaller footprint--saving real estate; also the narrower cabinet is easier to brace.

[WireNut]

It's a push-pull design.   This will go lower IMO then having all the drivers facing forward. If you only use 2 per side efficiency would be approx 3db lower compared to the same drivers facing forward and wired in parallel. I've been using four 12" drivers (2 per side) for years facing forward in a vented cabinet but after building new push-pull cabinets those same drivers now have startling bass in a push-pull configuration. 

Not sure how you plan on wiring them if you use 4 per side however. May not work as good as 2 per side. Experiment with your software.

Looks to me like using four 8 ohms drivers in that configuration and wired in parallel would be 2 ohms, wired in series would be 32 ohms, wired in series parallel would be 8 ohms.

In my push-pull setup using two 8 ohm drivers wired in parallel gives me 4 ohms, right where I like it.

[MJK]

It's a push-pull design. This will go lower IMO then having all the drivers facing forward.

Why do you think it will go lower?

[WireNut]

In my software it was easy to see that a push-pull configuration gave me -3db @ 27hz rather then -3db @ 56 hz from a vented front firing cabinet. After building the push-pull cabinets I was shocked to hear the difference. The lower 3db down point is startling at times and makes listening more exciting overall. It’s not as efficient as having both drivers facing forward wired in parallel but the improvement for me was well worth it. Try it in some loudspeaker design software with any 2 woofers. It’s pretty interesting to see the results.


 

[MJK]

In my software it was easy to see that a push-pull configuration gave me -3db @ 27hz rather then -3db @ 56 hz from a vented front firing cabinet. After building the push-pull cabinets I was shocked to hear the difference. The lower 3db down point is startling at times and makes listening more exciting overall. It’s not as efficient as having both drivers facing forward wired in parallel but the improvement for me was well worth it. Try it in some loudspeaker design software with any 2 woofers. It’s pretty interesting to see the results.

In a vented box with a constant volume adding an additional driver configured in push-pull maybe what you have claimed is the case. But the OP is showing a dipole speaker, this is not the same type of "enclosure" loading. You will not achieve a lower cut-off frequency in a dipole speaker system by operating drivers in a push-pull configuration.

[WireNut]

MJK,

 You are correct. My configuration is called a compound push-pull where both drivers are mounted at the same height vertically but facing magnet to magnet or can also be cone to cone.

Modeling the OP’s setup with 4 random woofers all being the same, and also a front firing setup with the same 4 drivers, both in identical cabinets and wired for 8 ohms shows equal 3db down points.

This would seem to raise the question what are the advantages if any of the OP’s configuration as opposed to all the drivers facing forward. My software does not show any but I would think it would have to do with room nodes and bouncing lower frequencies off the rear wall. Seems like it’s a tough one to predict in advance since driver mounting does not seem to change things.

[MJK]

It would seem logical to me that the best acoustic performance of a driver would come from the front of the cone. I cannot believe that the acoustic performance from the back of a driver is better than the front, it is probably worse. So I conclude that mouting all four drivers in the same direction, and seeing the front of the cones, is a better set-up than what is depicted in the sketches. I do not see any advantage from the different mounting directions.

[WireNut]

You might consider building each driver in its own U/H frame so you can add or delete woofers to mate with different efficiency and types of drivers for the mids and treble.

I like that idea .  If he builds four boxes per side he can experiment by rotating the boxes and rewiring if needed. Yeah it would be a pain building 8 boxes total but sure would be versatile. Hell, start out with 2 boxes per side. He could even do 2 boxes in 4 corners for a multi channel setup for even more versatility later on. 











 

[Russell Dawkins]

I like that idea .  If he builds four boxes per side he can experiment by rotating the boxes and rewiring if needed. Yeah it would be a pain building 8 boxes total but sure would be versatile. Hell, start out with 2 boxes per side. He could even do 2 boxes in 4 corners for a multi channel setup for even more versatility later on. 

That does sound like a really good idea - perhaps two boxes per side, each with two drivers. With those, all sorts of combinations could be tried - including vertical arrays of four, square arrays with drivers facing backwards and forwards, or all forwards, and even two arrays of four laid sideways on the floor, for maximum floor (and room) coupling -  in the interest of headroom or efficiency.

edit: I just remembered that none other than Klein & Hummel (now under the Neumann name) recommend this last configuration - except they talk of boxed subs - in their "Plane Wave Bass Array"™.
They say this, in part:
A benefit of multiple subwoofer systems is the possibility to reduce the side wall interaction thereby improving consistency in the side-to-side low-frequencyreproduction. This is important in studio applications where the sound engineer needs to move left and right along the mixing console, or where there are multiple listening positions along a large format mixing consol, for example in the movie industry. The subwoofers should be positioned along the front wall to generate a plane wave down the room. This is called a “Plane Wave Bass Array™” (PWBA™). The required number of subwoofers depends on the width of the room: wider rooms need more subwoofers. Two to four small subwoofers are recommended for small rooms and three to four large subwoofers for larger rooms. The subwoofers should be positioned along the front wall with a spacing of 70 cm (2.5’)
this from the "Setting up studio monitors" guide linked to from the top right of this page:
http://www.neumann-kh-line.com/neumann-kh/home_en.nsf/root/prof-monitoring_knowledge

[guest60106]

It would seem logical to me that the best acoustic performance of a driver would come from the front of the cone.

Martin,

This is the logic I used (however flawed) in the design shown. The ideology being that if there is a difference at all in the front and rear output radiation patterns of a given driver, the best shot at eliminating that difference may be to use a multiple driver arrangement that is as close to exactly symmetrical front to back as possible.
I believe this design will possibly make an audible difference than the design with all drivers firing forward (the design I have now). The question for me is....will the design change I have shown make a positive audible difference over what I am using now, will it make a negative difference by introducing other artifacts, or will it make no audible difference at all.

Like I stated before, I have read that driver output “symmetry” front to rear of an OB design makes little if any difference provided there is measurable rear output. My logic is that if this is true, then where is the dividing line in the equality of front to rear output. Can it be or does it need to be quantified?  Obviously, there has to be a line somewhere because logic suggests that the further you separate the front and rear output (SPL) of the OB design, the closer  you come to an Infinite Baffle design which is not a Dipole (at least not by my definition).

If my thinking is flawed then I am sure someone with better understanding of the subject matter will let me know (hopefully maybe you?). But, that is why I brought the subject up here to begin with.

[JohnR]

if there is a difference at all in the front and rear output radiation patterns of a given driver,

Hi, I think you'll find that a difference between front and back of the driver itself only comes in above some frequency. At low frequencies it will be the frame that will give rise to any asymmetry. You could do a measurement with your existing frame to see?

[guest60106]

Yes John that can be done. I just haven't had time to do it. I am going to try to get to it in the next couple of weeks.