Dual woofer cabinet question

[mefistofelez]

Greetings all,

let us stipulate that to satisfy a desired low-frequency corner, a woofer requires an enclosure volume equal to V_enc. If one were to use two such, i.e., the same woofers, would it make a difference whether the two woofers share a single enclosure volume equal to 2*V_enc, or whether the enclosure is divided into two independent, i.e., isolated, volumes, each equal to V_enc?

Kindest regards,

M

 

[AllenB]

No, it won't change the alignment.

 

[Andersonix]

Whenever ADS used two of the same woofers they would separate the two compartments, and I speculate they were of slightly different volumes. Or maybe they did it for stiffness of the cabinet?

 

[tktran303]

AllenB is correct.

Andersonix is also correct. One can separate the compartments and also maintain the same alignment. Of course, internal partitioning of the cabinet also adds bracing effects. So then one wonders whether bracing alone would have been a better idea. So why they do that is unknown.

In fact, some manufacturers use two different sized woofers sharing the same cabinet, with the same low pass filter. Whatever (dis)advantage occurs before of this, it is not clear. But this is a different chapter...

 

[AllenB]

I've learned that some are interested in avoiding standing waves in the now longer cabinet, but the thing to realise is that when you mirror two woofers in a double length cabinet, the modes are no different with or without the partition.

What may sometimes change is the outside effects like baffle step and room mode interaction.

 

[planet10]

Sepearate sub-enclosures only make sense if the woofers ar n’t that well matched.

With 2 woofers you can tale advantage of push-push and the active reaction force cancelation and thus greatly reduced box load.

dave

 

[mefistofelez]

Greetings all,

thank you very much for the replies.

Hi Dave,

With 2 woofers you can tale advantage of push-push and the active reaction force cancelation and thus greatly reduced box load.

As I understand it, this is possible depending on the highest frequency, where the woofer(s) will be crossed, because the woofers must be on the opposite sides of the cabinet, correct?

Kindest regards,

M

 

[planet10]

Yes, it limits how high you can XO. We’ve used 5.25” up to 450 Hz, larger will not reach as high.

At some point the bass becomes omnidirectional and XO should be in that range.

If you are crossing really low back/front is also doable… or if you can live with a bit of bipole dip.

dave

 

[mefistofelez]

Hi Dave,

thank you for the illustration and the explanation.

Kindest regards,

M

 

[profiguy]

Based on my past experiences, running two smaller enclosure partitions equal to one of the same total volume resulted in tighter, more controlled low end.

I'm not sure why exactly that is, but the difference was rather noticeable with both sealed and vented designs. There are also subtle differences in sound with parallel vs series wired LF drivers in both partitioned and shared air space.

I've searched for a while behind the theory of this, but can't find any reasonable scientific answers.

 

[tktran303]

Suppose we take the extreme case.

10 woofers in a long rectangular prism shaped cabinet.

Now add partitions to separate all the woofers. The increase in cabinet mass alone makes for a more interesting inert or stiffer cabinet.

One fine day I’d like to measure the vibrational or microacoustic behaviour of a twin woofer cabinet

 

[planet10]

10 woofers in a long rectangular prism shaped cabinet.

Which would have a TL response. 10 separate enclosures would look sealed.

To by be structurally similar the sealed bits are generated by not putting holes in the 9 individual braces. So mass, which isn’t really an asset, isn’t a consideration.

dave

 

[profiguy]

I'd agree that horizontal opposing drivers would sound better for LF than mounted on the same operating plane. The force cancelation alone is a huge benefit. Mounting drivers in their own compartments is for me a safety feature, protecting the other driver if one fails when they're wired in parallel. The other benefit is pushing the standing wave out of operating band. I'd also do this with a 2.5 way, especially with a series network. In the practical world this may not appear to be such a big deal. Really depends on how OCD you are with things.

 

[mefistofelez]

Hi tktran303, Dave,

maybe I am misreading the intent of your posts, especially Dave's:

Which would have a TL response. 10 separate enclosures would look sealed.

but are you suggesting that the number of woofer changes the answer?

Hi profiguy,

The other benefit is pushing the standing wave out of operating band.

Does your quote not (i) directly contradict AllenB's in post #5 and (ii) yours in post #10?

Would any simulation program be able to answer the stipulation re standing waves?

Kindest regards,

M

 

[GM]

Based on my past experiences, running two smaller enclosure partitions equal to one of the same total volume resulted in tighter, more controlled low end.........there are also subtle differences in sound with parallel vs series wired LF drivers in both partitioned and shared air space.

Even drivers with identical t/s specs often aren't when heated up from both the cab net Vb & VC heating and while not normally a big deal for typical purely HIFI apps; true subs, HT LFE apps and especially prosound apps OTOH can be an expensive problem if not in individual cabs.

 

[AllenB]

Would any simulation program be able to answer the stipulation re standing waves?

When the sources are distributed (and balanced) along the length of the line they produce sound without a pressure gradient in that direction, so the length modes are not activated.

This simulation shows the fundamental situation by comparing a point source to a length long line source. Even though separate drivers only act as a line source below a certain frequency and above that new modes will creep in, they will still be the same with and without the cabinet dividers due to their symmetry.

 

[planet10]

but are you suggesting that the number of woofer changes the answer?

No. The shape of the enclosure. Anytime one dimension becomes significantly greater than the other 2, the box becomes a TL of some sort. This shape would describe 10 woofers in a linear array.

dave

 

[mefistofelez]

Hi GM,

Even drivers with identical t/s specs often aren't when heated up from both the cab net Vb & VC heating . . .

O.K., but should this condition not be averaged, regardless of the enclosure construction, i.e., single volume or two insulatedl half/volumes?


Hi AllenB,

When the sources are distributed (and balanced) along the length of the line they produce sound without a pressure gradient in that direction, so the length modes are not activated.

I interpret your sentence as asserting that the (two) sources distributed along the length, will not produce a pressure gradient and, thus standing waves, along the longitudinal axes. Is that correct?

Now, if a single source is in the middle of the longitudinal axes, and is such balanced(?), is the above also true? I interpret the term "balanced" as symmetrical and, in the case of multiple woofers, equal s to the woofers' properties.

Can you tell me what is the simulator you used to produce the picture?

Hi Dave,

No. The shape of the enclosure. Anytime one dimension becomes significantly greater than the other 2, the box becomes a TL of some sort.

O.K., that would make sense reading your previous answer. Is there some rule-of-thumb or a simulator that would illustrate this?

Kindest regards,

M

 

[AllenB]

I interpret your sentence as asserting that the (two) sources distributed along the length, will not produce a pressure gradient and, thus standing waves, along the longitudinal axes. Is that correct?

Now, if a single source is in the middle of the longitudinal axes, and is such balanced(?), is the above also true? I interpret the term "balanced" as symmetrical and, in the case of multiple woofers, equal s to the woofers' properties.

I'm considering balance to refer to both level and position.

Due to the complex nature of the question I want to use two particular simulations to demonstrate. The first shows that multiple separate sources is the same as a line source below some frequency..

The second point is that even at higher frequencies (unless you use damping material 😉 ), where length modes are produced because of the driver spacing, it's symmetrical with and without the divider.

Firstly, imagine the top half in this screenshot to be a long box with divider, and the bottom half as the same box without divider. I've drawn them mirrored simply to show the effect better. Each has two sources in it..

Now showing some modes as the animation plays out to demonstrate that the divider makes no difference...

Can you tell me what is the simulator you used to produce the picture?

https://falstad.com/ripple/Ripple.html

 

[GM]

Greets!

In theory and done with T/S when there's enough difference, but measured at worst case high power for each app? Don't see that happening for all but a prosound driver manufacturer in their recommended cabs.