[rick57]

I read MJK’s very helpful paper from http://www.quarter-wave.com/OBs/U_and_H_Frames.pdf 
This shows for a single woofer amongst those three options, the H Frame has (slightly) the lowest f3 and the most output around 250 Hz. In exchange for the lowest efficiency.

With an H frame, if the 'wings' were on all four sides, rather than just two opposite sides, would it probably make only a small difference if the two woofers to be used per channel, were side by side, rather than on top of each other?

OTOH, if the ‘wings’ were to be on all four sides, in a rectangular shape eg 16” * 32” (to accommodate two woofers) there will be a second set of quarter wavelength resonances. These would be at half the wavelength (from the added sides 32” apart) of the resonances from just a pair of wings 16” apart.
So in Martin’s example, all other things equal eg with a depth of 7.5”, there would be an added resonance at 131 Hz . .

Martin, if you are reading this . .
In the formula on page 3, what is R_effective?
And in the formula on page 10, is c the speed of sound?

Thanks

[rick57]

johnk's approach:

the theory I found: http://www.musicanddesign.com/NaO-II-U-frame.html

The material? for any folded design . . to dampen any cavity, what materials actually work down around only c 130 Hz?

[MJK]

I read MJK’s very helpful paper from http://www.quarter-wave.com/OBs/U_and_H_Frames.pdf 
This shows for a single woofer amongst those three options, the H Frame has (slightly) the lowest f3 and the most output around 250 Hz. In exchange for the lowest efficiency.

With an H frame, if the 'wings' were on all four sides, rather than just two opposite sides, would it probably make only a small difference if the two woofers to be used per channel, were side by side, rather than on top of each other?

I think what you are asking is for the following scenario. Two woofers in the same H, or U, frame side by side without a dividing wall to form a rectangular exterior geometry. Then the question is will the cabinet orientation with respect to the floor matter, long side on the floor for a short wide profile or short side on the floor for a taller tower profile. I believe it will make a difference but not a huge difference, you should think of the floor as a mirror and visualize a second cabinet under the floor. Each orientation doubles in height.

OTOH, if the ‘wings’ were to be on all four sides, in a rectangular shape eg 16” * 32” (to accommodate two woofers) there will be a second set of quarter wavelength resonances. These would be at half the wavelength (from the added sides 32” apart) of the resonances from just a pair of wings 16” apart.
So in Martin’s example, all other things equal eg with a depth of 7.5”, there would be an added resonance at 131 Hz . .

I do not believe there will be any additional resonances across the enclosure between opposite side walls of the H frame. Thinking of a simplistic model, the pressure fluctuation at the open end of the cavity needs to be zero. The open end also provides a fair amount of damping that removes energy from the enclosure sending it out into the room. For these two conditions the only way to support a standing wave between either of the two sides is for it to be a maximum at the driver baffle decreasing to zero at the open end. I just don't see this happening or having enough strength to change the H frame's performance.

My design method for dipoles is to limit the woofer contribution to below the hump or peak that exists in OB or U and H frame designs. Use the hump in the crossover design to transition to the next driver be it a full range or a mid range driver. I think it is a mistake to crossover a large woofer much above 200 Hz and leave the hump or peak in your woofers SPL response.

In the formula on page 3, what is R_effective?
And in the formula on page 10, is c the speed of sound?

R_effective is the radius of a circle that has the same area as the open end of a U or H frame. The end correction formula is based on the acoustic impedance of a circular piston source in an infinite baffle

c is the speed of sound

Hope that helps,

Martin

[scorpion]

rick57,

Have a look at Martin's H-frame, it is closed on all four sides (per definition) and Martin just explained why there is no offensive resonance doing things his way. From my simulations this will not be changed in any fundamental way by accomodating two drivers in parallel over each other instead of just one. But gaining 6 dB SPL. There is however one thing to remember, and perhaps you cannot solve it without MJK's simulation software, you will have to choose crossover frequency and crossover characteristics with care.

/Erling

[rick57]

Thanks Martin, you understood the scenario.

In a rush . .

I understand and agree with most most of your points, excepting

> you should think of the floor as a mirror and visualize a second cabinet under the floor.

I understand that . .

> Each orientation doubles in height.

 . . but what do you mean by that?

More importantly 

> The open end also provides a fair amount of damping that removes energy from the enclosure sending it out into the room

You don’t think that damping boosts (iirc) output below Fs, as johnk advocates?

Cheers

[rick57]

erling
I’m going to parallel two 12 inch woofers with a Qts of 0.8, so there should be ample output. And yes the XO will need care and time . .


Rather than an H frame, I’ll likely be using a U frame, or possibly a W.
(The W frames with the drivers in push-pull cancels some (even order driver?) distortion).

Cheers

[MJK]

Thanks Martin, you understood the scenario.

In a rush . .

I understand and agree with most most of your points, excepting

> you should think of the floor as a mirror and visualize a second cabinet under the floor.

I understand that . .

> Each orientation doubles in height.

 . . but what do you mean by that?

If you have a 16" by 32" U or H frame and you place it on the floor then you have the following effective sizes at low frequencies because of the floor reflection. Lying it on the long side the effective dimensions are 32" wide by 2 x 16" tall. Lying it on the short side the effective dimensions are 16" wide by 2 x 32" tall. If the software being used to design OB, U, or H frames does not include the influence of the floor at low frequencies, you are getting a result that is understating the potential bass performance.

More importantly 

> The open end also provides a fair amount of damping that removes energy from the enclosure sending it out into the room

You don’t think that damping boosts (iirc) output below Fs, as johnk advocates?

My models do not show a boost in low frequency output when fiber damping is added into an U frame. In fact it is just the opposite, the low frequency output sags by a couple of dB.

[rick57]

Thanks again Martin , you're a fountain of information

 

[MJK]

you're a fountain of information

A fountain of something, different people might have other opinions about the something.

You are welcome,

Martin

[scorpion]

rick57,

As soon as you mount the speakers opposite on a baffle you will have 2nd order distortion cancellation, have a look at SL's Phoenix prototypes. From the W you will not have any workable reply over 100 Hz and you will have to take care of the fundamental resonance with a noth filter. The W is quite a Sub-solution, which can work wonders if you know what you are doing.

/Erling

[rick57]

Erling

> W you will not have any workable reply over 100 Hz

Thanks, as I want to go to 200-250 Hz, the W is out.


With a combination of two speakers facing in opposite directions (wired out of phase), and mounting in a U - output to 250 Hz isn't I would think diminished?

Thnaks