Compromises with a "boxed" dipole speaker?

[Mazeroth]

I purchased a boatload of Elemental Designs EHQS8 8" woofers a few months back for $5 each and am actually very impressed with them for what I paid. An honest 6mm xmax and some very impressive distortion measurements above 40 hz (mostly below 0.05%) that I took with Arta. I had an idea to build a pair of very inexpensive dipoles that could be crossed to a subwoofer and here's what I'm thinking. Four EHQS8s in the front and back of an enclosure wired out of phase. The front baffle will extend an additional 10"+ to allow for a dipole MT or MTM. Using a box instead of a typical U or H frame I'm hoping to gain the ability to cross over higher and avoid the peaks the other two formats are prone to. If I can get a healthy cross around 300-400 hz I could use smaller mids and alleviate some crossover component costs. Another nice thing I'm hoping this will give me is deeper bass, as I can build the enclosure quite large, increasing my effective path between sets of woofers. I'm thinking somewhere along the lines of 11-12" wide and up to 20" deep.

With that, can you see any flaws with doing this? Will I need to include some type of sound absorption for the internal waves?

Thanks.

It will look something like this, only larger:

[Mazeroth]

81 views and no replies. Come on, guys! :B

For a little bit of clarification these will have 4 8" woofers in the front and back, for a total of 8.

[Jed]

What is the senstivity of the woofers? I'm wondering if you'll have a mismatch because of the quad set as opposed to the 1-2 mids running open baffle. Just a thought.

Also, I assume you are running the woofers Bi-pole? I don't have much experience with that configuration, but I'm wondering about the logistics of running them up to 350-400hz like you propose and how that would integrate with the dipole rear wave. Might have some interesting phase issues, but you never know.

[Mazeroth]

The woofers are 85dB efficient so the quad will be 91 dB. I will be using at least 6dB of "BSC" so they will be down to the mid to lower 80s when I'm done. I will definitely have to pad the mids and tweeters down to match the sensitivity of these. The good thing is these won't need to be played very loud so their low sensitivity shouldn't be an issue.

The woofers will be wired out of phase so they will be dipolar and not bipolar. The only reason for me doing this is because the woofers were so cheap ($5 each) that using 8 per side will only run me $40. Using a box should eliminate (I hope) the dipole peak associated with U and H frames, hopefully allowing me to cross over much higher than the typical U or H frame. This will also give a near perfect front and rear output unlike a typical dipole that has an inhibiting frame and magnet.

I did do a test last night on my test U frame with a single 12" version of these (EHQS12) with a 14" wide baffle and 6" side panels. From 1m, with a 2.0mH inductor and a 4ohm resistor (BSC circuit) I was ruler flat from 80 to 500 hz. and only down 3 db at 50 hz. The proposed alignment with the 8 8" drivers will give me an additional 20% output capabilities over the single 12" I used in the test. The output from the single 12" was more than enough for moderate music levels and that was only coming from 1 speaker. I can add 6 dB if I had two playing.

All I'm trying to find out is if there are any drawbacks to using a large box (depth-wise) in a dipole configuration with drivers wired out of phase? To me it would seem more ideal than the U or H frame but before I get to cutting a lot of wood for test boxes I'd like some input.

Thanks. :T

[Johnloudb]

The woofers are 85dB efficient so the quad will be 91 dB. I will be using at least 6dB of "BSC" so they will be down to the mid to lower 80s when I'm done. I will definitely have to pad the mids and tweeters down to match the sensitivity of these. The good thing is these won't need to be played very loud so their low sensitivity shouldn't be an issue.

The woofers will be wired out of phase so they will be dipolar and not bipolar. The only reason for me doing this is because the woofers were so cheap ($5 each) that using 8 per side will only run me $40. Using a box should eliminate (I hope) the dipole peak associated with U and H frames, hopefully allowing me to cross over much higher than the typical U or H frame. This will also give a near perfect front and rear output unlike a typical dipole that has an inhibiting frame and magnet.

I did do a test last night on my test U frame with a single 12" version of these (EHQS12) with a 14" wide baffle and 6" side panels. From 1m, with a 2.0mH inductor and a 4ohm resistor (BSC circuit) I was ruler flat from 80 to 500 hz. and only down 3 db at 50 hz. The proposed alignment with the 8 8" drivers will give me an additional 20% output capabilities over the single 12" I used in the test. The output from the single 12" was more than enough for moderate music levels and that was only coming from 1 speaker. I can add 6 dB if I had two playing.

All I'm trying to find out is if there are any drawbacks to using a large box (depth-wise) in a dipole configuration with drivers wired out of phase? To me it would seem more ideal than the U or H frame but before I get to cutting a lot of wood for test boxes I'd like some input.

Thanks. :T

Will, the front and rear 8 inch woofers have there own sealed, or ported enclosure? If they are all in the same box and wired out of phase (dipole) then I'd think you get some early roll off. How much I don't know. I have some Nelson/Reed dipole subs (four 12" inch woofers) in one box that have second order slope in the pass band. They corrected it with a 2nd order low pass filter to boost the low frequencies. I'd guess you have the same issues.

You know dipoles, so you probably took that into account. Other than that speaker placement might be difficult with the extra depth of the cabinet putting the woofers back closer to the rear wall. Just thoughts.

[Dennis H]

Interesting project. I must have missed it when you originally posted it.

Using a box should eliminate (I hope) the dipole peak associated with U and H frames, hopefully allowing me to cross over much higher than the typical U or H frame.

You'll avoid the U or H cavity resonance but you'll still see the dipole peaks and nulls based on the "D" distance which should be about half the baffle width plus the full box depth. Make sure you don't make it so deep that the first null gets down into the passband.

Assuming the front and rear drivers share the same box (best for efficiency -- no pressure inside the box) I'd loosely stuff it to prevent resonances based on box dimensions.