Dipole bass with Adire Shivas?

I have 6 Adire Shivas laying around collecting dust and I was thinking, hey, a dipole with the Adires would be perfect for my new line arrays.

I'm designing a pair of open baffle line arrays with (9) Dayton RS150 drivers and (8) Dayton PT2-B planars per side. I was thinking of making a pair of dipole stereo subs, housing 2 Shivas per cabinet in an "H" frame. I'm open to suggestions on the frame. I did some reading over at linkwitzlab.com and found some of it quite confusing, so if anyone has any good links to dipole subs I'd really appreciate it.

I will be using a Behringer DCX2496 for the crossover, so I can add EQ boost to the subs, which I know they'll need. I'm sure the Dayton RS150s (6") will require some EQing as well.

Lastly, what parameters make a great dipole sub? I figure since I have them lying around I might as well use them. Or I could sell them and buy something else.

You'll probably be OK with the Shiva's. The builder of "Bob" discussed here http://www.doddsy.net/steve6_011.htm used 2 DPL12's per side and he say's he's flat to 20hz, but doesn't state the SPL. Your Shiva's have a bit more Xmax than the DPL12's, so they have slightly more bass potential. You will probably need to address the vented pole piece on the Shiva's as it may make noise. I wouldn't plan on taking them higher than 80-100hz because the Shiva's are purely a subwoofer driver and not as musical as a normal woofer.
The Shiva white paper shows usable frequency good to about 400-500 hz, even though I'm not so sure how good it will sound, and I have no intention of crossing over that high. I was thinking somewhere in the neighborhood of 100-150hz to my line of RS150s. I have 8th order at my disposal so I can make the frequency drop off pretty quickly :D

OB requires EQ anyway, so I'd plan on taking those Daytons down to 50 or 60 unless you plan on pushing them really hard. You can play with that once you get it set up. You've got plenty of excursion with your 9 Daytons, so think of the Shiva's as bass fill. In fact, you may want to consider using the Shiva's in mono as a real sub and use all 6 as an array as well. Since their CTC spacing can be quite large and bass needs long line lengths to achieve the line array benefits, maybe 3 WBaffle units with a pair of Shiva's in each spread evenly accross the width of your room could be your subwoofer.
In essence a dipole is very simple. It's a flat baffle where the effective width (D) determines Fequal, the point at which the output is the same as the driver in a seald box. Below Fequal there is an additional 6db/octave rolloff you must compensate for.

You can calculate Fequal:

Fequal = 0.17 x 344 / D

hence for the Linkwitz Phoenix dipole woofer, D=480mm and Feq = 120 Hz.

I'd build a W frame rather than a H frame since it gets a lower Fequal for the same size, due to the greater path difference. Also I would build a true W frame rather than the simplified version Linkwitz uses where the walls are parallel which creates a resonance. This has been discussed elsewhere on the forum - search for M or W baffle dipole.

I'd experiment with 60 - 120 Hz as the xo point for the dipole woofer since you can do it easily with DCX. Cross the Shivas too high and you use them in a range in which they don't perform well, too low and you will lose dynamic range due to the excursion limits of the mains.

You should also consider Fequal with your mains, it may be more significant than you think. Depending on how you treat the rear wave (using damping wall treatment behind or no treatment), this will impact the efficiency. Above Fequal, the efficiency rises to a point where it can reach 6db greater than a monopole.

Consider this:

Consider also:
Q20 - What is the sound pressure level at 1 m for 1 W of power under anechoic conditions?

"The voltage sensitivity has been estimated (Design Models - I) for the PHOENIX as 90 dB SPL for the tweeter, 103 dB for the midrange above 250 Hz which drops to 96 dB at 100 Hz, and for the woofer as 95 dB at 100 Hz decreasing to 85 dB SPL at 30 Hz for a constant 2.83 V across the driver terminals."

Something also to consider ...
One correction to Paul's info. "D" is not effective baffle width. D is the additional travel distance of the rear wave to reach your ears vs the front wave, so in the case of a flat baffle, it's the driver to the baffle edge. If you have folded sides or "wings", D is the distance from the driver to the rear edge of the wing plus the distance from the rear edge to the plane of the front baffle.

Regarding W or M configuration for dipole woofers, I prefer the more compact size and total cancellation of mechanical forces of using the Linkwitz layout with the parallel driver mounting baffles. The frequencies are too low to create a resonance IME. The angled baffle shape does have a the advantage of being usable to higher frequencies without the operation of the drivers interferring with each other.
Just be careful about the nasty peak that comes with the w- frame. I have built both an H-frame and a w- frame. My w-frame has a really nasty peak at about 160Hz. I built it according to Linkwitz page. The cavity for my H-frame is much shorter so the peak comes much higher in frequency. It should be considered with regards to the XO frequency. I cross my dipole woofers at 60Hz with 18 db/octave. when i cross higher the peak comes in a bit.

Just my thoughts in the subject.


Shiva in OB dipole


I actually tried the Shiva in an open baffle dipole, it was an experimental setup, just a piece of wood ca 80x80cm with a hole to fit the Shiva.

Sound was very good, i needed little eq to get good output in the 20-30Hz range. I had more than enough output (with music) for my taste.

In my final implementation I use Visaton TIW400DS drivers, they sound a bit 'cleaner' above 100Hz.

regards, Peter