[Kevin Haskins]

I've been thinking of doing an OB (open baffle) design that is similar to Dan Wiggins DDR (dipole done right).   The cool thing about Dan's design is that the crossover is fully passive.   You only need one amp to run the entire loudspeaker rather than an active multi-amp setup.    This obviously allows a significant cost savings relative to using multiple amplifiers.   

On a 12" wide baffle Dan is getting good response down to about 70Hz with a pair of Extremis.   I'd use our new Exodus EX-6.5 which is similar to the Extremis in output.   We would use the Neo3 as per Dan's design and we are thinking of a few other special "tricks" to significantly extend the low-end bandwidth.   

How would you like a complete dipole design with the same footprint as the DDR with in-room extension to 30Hz?   How about at a kit price of < $1200/pr and the crossovers assembled and built?   How about options for a CNC machined baffle in exotic hardwoods?     

http://www.adireaudio.com/Files/DDRDipoleDesign.pdf

[konut]

This has a lot of merit. How far up, in frequency, could the EX-6.5 be crossed over, at 24db per octave, before significant beaming would occur, for those of us who would like to use an active xover?

[Kevin Haskins]

This has a lot of merit. How far up, in frequency, could the EX-6.5 be crossed over, at 24db per octave, before significant beaming would occur, for those of us who would like to use an active xover?

I'd like to get a pair of them on a baffle and measure.   The dipole dispersion pattern differs from what I typically see in an enclosure so before I'd comment I'd rather have measurements to back it up.    Your going to get into vertical dispersion problems due to the MTM arrangement which is what would limit its usability upward with that configuration.

We could of course always limit ourselves to a single EX-6.5 which limits our lower-end bandwidth covered by the midbass.   With the solution I'm looking at we would have a pair of woofers to handle the bass anyway.   There is just no way to get 30Hz out of a 6.5" driver in a dipole set-up!     With a single EX-6.5 I'm assuming we could get to 3K without a dispersion problem.   The other design option is to bring in the extra EX-6.5" only at lower frequencies.   It could be an MMT arrangement with the bottom EX-6.5 coming in to counteract the dipole rolloff.   I'd have to build one and do some work though to see if that is a feasible arrangement.

[konut]

Very good explaination. Looking forward to your findings.

[JohninCR]

Kevin,

Of course it sounds interesting, but pardon me for being a bit skeptical about a 12"x12" footprint dipole[/b], using how many 6.5" drivers, that is remotely flattish to 30hz in any room.  With some kind of hybrid U-baffle I'm game, but not a dipole.

[ezeg]

The DDR's (NOW Called "Dakini" by some) baffle is 12" X 48".

I have a pair of DDR's with a pair of Linkwitz woofer-2's for
the bottom end -- four Shiva's.  I have a little over $1000
in the speakers.

I like the sound a lot!

I'm thinking about another pair for the rear surrounds.
So, yes, I would like to see Kevin's design.

[JLM]

Some good design trade-offs here (like giving up on the deep bass to avoid bi-amping and the narrow baffle to keep it domestically practical).  I'm hoping that the stated 80 dB/w/m efficiency is wrong as that would be very low (doubt that 100 wpc would cut it for most and would think that the dynamics to be very limited).  The Visaton B200 driver (an OB favorite) covers the same frequency range by itself (with high frequency beaming) and does it at roughly 96 dB/w/m while being rated at even higher wattage.

Some ideas:

1.  It would seem to be easily tipped over, needs a bigger and heavier base.
2.  I'd prefer a lower order crossover, but keep it as high as possible.
3.  If you're going to stay with a 6th order crossover, I'd push the crossover point farther from 500 Hz.
4.  I like the 2-way design, if you go 3-way see ideas #2 and 3 with a smaller midrange driver.
5.  Design a separate U-baffle powered sub to complement.
6.  I believe the name "Dakini" has already been taken for a loudspeaker (but probably not registered).

IMO dipoles/bipoles provide a more natural soundstaging (not hyper-real pin-point imaging that audionuts rave about), but are less tolerant of placement close to corners or back walls than other designs.  They also need more distance to the listener to get the sound to "gel" (nearfield applications need not apply).  Dipole/bipole can make a make nice alternative to traditional designs.

[DanWiggins]

JLM,

Dipoles are inherently low efficiency, if you want wide bandwidth.  That B200 is 96 dB efficient, but if you place it on a 12" wide baffle and want response down to 70 Hz, you're going to end up with ~80 dB efficiency at 70 Hz.  Meaning that you have to equalize the response to get it flat.

So, you can either add an active EQ before your amp to add ~16 dB of boost to the bass (noting that means you require an amp with 16 dB of headroom over what you want in the midrange), OR you use EQ to cut the high frequency output down by 16 dB.

Either way, you need the same amount of power in the amp, and the system really is 80 dB efficient.  Those claiming high efficiencies in dipoles without looking at the associated EQ required to flatten the response are deluding themselves.

Dipoles take power, if they're going to be wide-band and not have a massively wide baffle.  No way around it.  My take on the solution was different; most people use multiple amplifiers and lots of gain to get around the EQ and losses.  I chose to use a passive XO so that you can use a single amp, and just recommend a big amp (a 100W amp will get you peaks in the 100 dB SPL range).

My personal belief is that most speaker DIYers want to build speakers, not 6 channels of amps and active crossover boards.  So with that in mind, a well-designed passive XO with a big 100+WPC amp is what you need.

Now, if you can get away with a 60-70" wide baffle, then a lot of this goes away!  However, that's not too practical for most people, as many can't give up 12 feet of space across the front...

[Kevin Haskins]

JLM,

My personal belief is that most speaker DIYers want to build speakers, not 6 channels of amps and active crossover boards.  So with that in mind, a well-designed passive XO with a big 100+WPC amp is what you need.

I think we could take care of both.   It doesn't take much to do an active version with the Exodus UcD platform.   A lot of the four channel amps I've sold are used in Orions.   Hey... if people want active version I can provide active.   Once we have the acoustical design work done the resultant active filter boards are a piece of cake.   I have room in the rear of the chassis for  the boards on the inputs and I have the necessary power supply already in place.    We could do a passive filter between the tweeter & midrange because I don't see much reason (other than marketing) to use yet another amp.   

You guys let me know if I'm off-base here.   There are a number of ways to skin a cat.   I'm sure we can design something that works great no matter the solution.   I just don't want to put a lot of work into something that people are not interested in buying.   The active design pushes the price point of building the kit up by over $1250 (cost of the current Exodus EX-400-4 amplifier).   I've found for a lot of people that pushes the overall design outside of the limits of their pocketbook.   The passive version makes it something that can be tried for much less money.

[Kevin Haskins]

Kevin,

Of course it sounds interesting, but pardon me for being a bit skeptical about a 12"x12" footprint dipole[/b], using how many 6.5" drivers, that is remotely flattish to 30hz in any room.  With some kind of hybrid U-baffle I'm game, but not a dipole.

You have good reason to be skeptical.   I lied a little... the footprint in the rear direction would have to be larger to fit the drivers and we wouldn't be using a 6.5" driver for the low-end.    aa

[Kevin Haskins]

Some ideas:

1.  It would seem to be easily tipped over, needs a bigger and heavier base.

I'd agree... I was thinking the same thing myself.   Much of the idea with the DDR was to keep it simple to build.   Dan has done a great job with that but he doesn't have kids.   My kids would have that thing face-down on the floor within two weeks.   

2.  I'd prefer a lower order crossover, but keep it as high as possible.

I think what you mean is you like low-order crossovers but as high in frequency as possible.   Keep it out of the midrange if possible.   Generally I don't do designs with preconceived notions of what type of electrical filter I'm going to use.   It is dependent upon the acoustical measurements.   But, I like to use as simple of a solution as I can find.   The KISS principle is always at work. 

[JohninCR]

It's nice to see more designers attempting fully open alignments, because open mids on top of a box is not even half the story in my book.  Too bad the coax 10" that Dan helped design looks like it will have high end pricing.  I was looking forward to doing a point source open design in a smaller footprint than the DDR that would blow the socks off of 70hz, so a sub would be needed only for real bassheads.

[Kevin Haskins]

It's nice to see more designers attempting fully open alignments, because open mids on top of a box is not even half the story in my book.  Too bad the coax 10" that Dan helped design looks like it will have high end pricing.  I was looking forward to doing a point source open design in a smaller footprint than the DDR that would blow the socks off of 70hz, so a sub would be needed only for real bassheads.

From a subjective standpoint I don't know how it would all shake out.   The main advantages to my mind about why dipoles do so many things well is that they limit early reflections from the side.   They also have a lot of reflected energy off the back wall which probably helps with the whole perception of space and finally they don't excite the room to the same degree as a monopole in the bass.

It would seem pretty funky to have only part of the audible spectrum in dipole and the other in monopole operation.  It depends somewhat on the room acoustics but if you go on the same assumption that we use designing a monopole, that the off-axis spectrum is very important to correct timbre and perceived sound quality, then it only stands to reason that the front/rear power response of a dipole should match as closely as possible.   

I'm not sure how much I bite off on the entire "cabinet sound" thing.   It seems to me you can get the cabinet issues under control and the cabinet resonance down to a point where it is no longer audible.   I think most of the differences are probably due to the unique power response of the dipole.   That is just my intuition though.   I don't have any measurements to back it up at this point.

[JohninCR]

I don't have any measurements to back it up at this point.

Even Linkwitz gets into the purely subjective about some of the issues related to his preference for dipole.  Since I've gone fully open, even good boxes sound boomy to me.  I think a lot of it has to do with my room construction being mostly concrete, so directivity in the bass region has a bigger impact for me in terms of reduced reflections than a room with less rigid construction.  Another big issue for me is that I often listen late at night, and with OB's I can listen much much louder without disturbing others.  In the interest of deep bass and size, I've gotten away from pure dipole at the bottom end and use hybrids that tend toward cardioid.  This gives me the bass I want while retaining much of the room benefit of dipole.

[klh]

Kevin

I take it you're thinking of a 3-way design. Is that with just three drivers? Will it like the Adire design include the Neo 3 PDR? I inferred that but clarification would be nice. I would be interested... but I only need to go down to 80 Hz. Maybe you could offer two versions... one that's more full range and another that is more like a monitor (they could also double as surrounds or even a small CC). Very interesting stuff... and as you suggested should mate well with your powerful amps .

[Kevin Haskins]

Kevin

I take it you're thinking of a 3-way design. Is that with just three drivers?

That is yet to be determined.   I was thinking of a DDR with an additional bass cabinet feature.  I'm pretty flexible at this point though.   I'm in the "brainstorming phase".   

Will it like the Adire design include the Neo 3 PDR? I inferred that but clarification would be nice. I would be interested... but I only need to go down to 80 Hz. Maybe you could offer two versions... one that's more full range and another that is more like a monitor (they could also double as surrounds or even a small CC). Very interesting stuff... and as you suggested should mate well with your powerful amps .

Yea... we would probably stick with the Neo3 PDR used in the DDR.   I'm open to other options but I see little reason not to use it.   It performs pretty well in this application without the back cup and the proper network design. 

What we would probably end up with (once again we are early in the process here) would be a design similar to Dan's DDR with a passive solution for the crossover and an option for a bass cabinet (dipole in design) that could be easily added.    I'll probably do an active version simply to satisfy everyone.  I don't mind doing it if doesn't add to the cost significantly.   In this case an active version would just sell more amplifier kits so that is a good thing!      The Exodus amplifiers would also remove the possibility of people using the wrong amp and getting poor results.    Overall, the more of the design I can handle in-house the better the chance that the results will turn-out  as expected.   

[Russell Dawkins]

In the interest of deep bass and size, I've gotten away from pure dipole at the bottom end and use hybrids that tend toward cardioid.  This gives me the bass I want while retaining much of the room benefit of dipole.

When you say bottom end, what range do you have in mind?

I am thinking that in an ultimate system, where you want to go down to, say, 8 Hz to go along with Bag End's reasoning, to go IB or small sealed box (like Bag End) might have to be what you do from 40 Hz on down, with appropriate equalization.

I remember years ago reading, from someone whose opinion I respected but whose name escapes me, that IB is more important in the bass region than even the mids, i.e., that the bass range benefits most by being OB.

That, I think, may be a fundamental flaw in design perspective with the opposite configuration which is most often seen these days - OB mids and boxed bass. The smaller footprint is very tempting.

[Kevin Haskins]

I am thinking that in an ultimate system, where you want to go down to, say, 8 Hz to go along with Bag End's reasoning, to go IB or small sealed box (like Bag End) might have to be what you do from 40 Hz on down, with appropriate equalization.

In a normal closed box design the necessary increase in displacement of the woofer is going to limit you way before you can recognize 8Hz and any reasonable level.   Even if you have program material down there (movies like Black Hawk Down, The Haunting etc..) I wouldn't want to reproduce it.   Why?   Because you rattle everything in your house and the pictures, bookshelves and window panes rattling is hardly musical.    The dipole situation is even worse... you would basically need a large segment of a wall on a motor that you could move back and forth.   It is doable but I wouldn't dream of trying it.   

I remember years ago reading, from someone whose opinion I respected but whose name escapes me, that IB is more important in the bass region than even the mids, i.e., that the bass range benefits most by being OB.

That, I think, may be a fundamental flaw in design perspective with the opposite configuration which is most often seen these days - OB mids and boxed bass. The smaller footprint is very tempting.

Having the bass is going to help with the room modes.   I think it is more than just the bass though.   It is the entire power response of the dipole and how it interacts with the room.

[Russell Dawkins]

actually Bag End gets down to 8 Hz flat with an 18 in a box 24X24X20 with an EQ of course. The particular clever trick with the system is that the sub operates entirely below the resonant frequency of the system, by design.

They also claim that the improved phase accuracy afforded by having response to 1 octave below what's needed is very desirable, once heard.

Unfortunately, I don't get the feeling that much is happening there at the moment - their website seems fairly unresponsive:

http://www.bagend.com/

[Kevin Haskins]

actually Bag End gets down to 8 Hz flat with an 18 in a box 24X24X20 with an EQ of course. The particular clever trick with the system is that the sub operates entirely below the resonant frequency of the system, by design.

Unfortunately, I don't get the feeling that much is happening there at the moment - their website seems fairly unresponsive:

http://www.bagend.com/

I am very skeptical.   I've never understood why they would want to operate under the resonant frequency.   The entire idea of a resonant system is that you can use it to limit driver excursion/distortion and extend the cutoff.  You also have some power handling benefits.

By using a driver under the resonant frequency it would be required to have MASSIVE capability.   More than is possible with even the most extreme 18" drivers.    Remember.. you need about four times the displacement for a given output as you go down each octave.   In other words, to get the same output @ 10Hz as you would at 20Hz would require four times the number of drivers (in equal sized enclosures, not the same enclosure).   Also... Hoffmans Iron Law is at work.   The box size vs. bandwidth vs. power requirements makes their claim of "flat @ 8Hz" a rather ridiculous claim.   Flat up to what SPL level and at what percent distortion?   My guess is that Bag End is also being pretty generous with the room gain calculations to get anywhere near that kind of silliness.