In Pursuit of a 20-20k Dipole Loudspeaker

I thought I would kick off 2019 with this thread about dipole loudspeaker design and construction. Specifically, how to create a loudspeaker having a dipole radiation pattern across as wide a frequency range as possible, when it might make sense to abandon this goal in favor of some other criteria, and the justification for employing the "full range dipole" concept.

One area that has been neglected (in my opinion) is what is happening to the rear of the loudspeaker. In an open baffle or dipole speaker, there is sound emanating from the rear that ultimately reaches the listener via the room and this is just as important as what is coming out of the front of the loudspeaker. So when I talk about a "full range dipole" loudspeaker I mean to emphasize that the front and rear radiation pattern should be as similar as possible over all frequencies.

I have been working towards these goals for a few years now and have built and dismantled a few test systems to try out some different approaches. I hope to share what I have learned, why I prefer this type of loudspeaker, and when it might work well or not work well for audio reproduction. I hope that others will chime in with their own experiences in this area so that we can all learn from each other.

I thought I could put down some thoughts into a few posts, but ended up creating a pdf file to kick off the topic (see attached). I would love to read about what systems you might be developing along these same lines and get your feedback about my approach to such a system.
 

Attachments

  • In Pursuit of the 20-20k dipole.pdf
    701.6 KB · Views: 1,764
Last edited:
You make it sound as if the old BR box is "correct" and the OB is the problem.*

It really depends on which aberrations you want to live with. Fans of OBs prefer their ambient natural sound any day, whatever the speculative textbook curves (based on speculative rooms with speculative drapery, etc) may say.

While textbook thinking may lead some folks to agonize over low freq cancellation, it is better to think about the reality of bouncing waves and incoherent phases and addressing the rear wave, not fretting over it.

Long pipe to sequester rear wave

Nobody more sophisticated than Linkwitz was (and music lover too) and he turned out
to truly endorse the sound of OBs.

B.
* OBs, like with electrostatics, ribbon tweeters, and motional feedback, don't "compute" for commercial manufacturing and that force twerks our perception (aka "availability heuristic")
 
Last edited:
Administrator
Joined 2004
Paid Member
Hi Ben,
AnOB speaker is much less expensive to ship than any other speaker design. That is a very strong pull in favour of commercial profits.

Personally, I haven't heard an OB I liked yet. I like variations on speaker boxes, like sealed, bass reflex, horn and transmission line.

-Chris
 
I love dipole sound for acoustic instruments and vocal - folk and classical recordings usually have nice sound with minimal studio trickery. Jazz and pop from 1950-60's sound good too!

I skipped the dipole radiation below 200Hz like my ideal origin Gradient 1.3. Planar drivers in narrow baffle help a lot above 2kHz! Check the link in my signature!

Front- and backside radiation (mirrored left-right), other normalized
 

Attachments

  • ainogneo83 vx out polar 10¤steps 12ms.jpg
    ainogneo83 vx out polar 10¤steps 12ms.jpg
    5.7 KB · Views: 3,295
  • ainogneo83 vx out polar normal 10¤steps 12ms.jpg
    ainogneo83 vx out polar normal 10¤steps 12ms.jpg
    5.6 KB · Views: 3,300
thanks for posting your findings Charlie!

Reading from your and Juhazi's experience I think I will stick to the B&G NEO10s for the 2nd system as well, actually only exchanging the woofers. You measured several drivers and apparently none is really as good as a planar.

What I am curious about and willing to try is dividing the bass (below say 400Hz) over two channels: the 4x WS25E up to 200Hz and a single 8" or 10" from 200Hz up to 600 (or higher): this woud reduce lobing in the bass and relieve the B&G NEO10 from bass duties.

Best, Erik
 
It's not a bad idea to add another driver between the woofer and the Neo10. I am facing this exact same dilemma in one of my own planned builds, which will use the Dayton AMTPRO-4. My measurements of that driver in free air (attached) show that it has a passband that starts just below 1kHz, but do to the distortion signature I will cross over to it above 1kHz, e.g. 1.2k-1.6kHz. A large dipole woofer will barely get there, so I will probably use a moving coil midwoofer between 500Hz and 1.5kHz, and then use some spare 18" pro drivers below 500Hz. Doing this will work, but it adds another band into the mix since I don't expect the 18's to go low enough, although I might try to use some baffle on them to help that out.

Your Neo10 is a little easier to work with. It has an advertised free-air frequency response that extends down to 200Hz or so IIRC. But the response is drooping below 1kHz. Since you should be able to use a nude 8" to 1kHz without any issues, you might consider bumping your crossover from the 8 to the Neo10 to 700-800Hz to use the planar driver where it has more sensitivity (output). For example of the response from an 8" nude in free air, see attached measurement of a Dayton RS225.
 

Attachments

  • Dayton AMTPRO-4 0-11-22-33-44-55-67-78-90deg 16in front.PNG
    Dayton AMTPRO-4 0-11-22-33-44-55-67-78-90deg 16in front.PNG
    41.3 KB · Views: 3,214
  • Dayton RS225S-8 0-11-22-33-44-55-67-78-90deg 70cm front.PNG
    Dayton RS225S-8 0-11-22-33-44-55-67-78-90deg 70cm front.PNG
    50 KB · Views: 3,191
Directivity and other measurements of the B&G Neo10 can be found in the thread on my 'Totem of Tone' dipole system.

Yes, I have read over your project. It's a great build and you got pretty much everything right. Nice job! You also figured out (via measurements) the problem(s) that arise when you start to add wings to a driver in an effort to reduce the low frequency dipole losses.

I think the post below taken from that thread captures a lot about why I am also striving for dipole and CD behavior over as much of the audio range as possible:

So, was it worth the effort? Yes, definitely yes! I’ve been listening to this system for over 6 months now and I haven’t yet detected anything problematic, as far as I’m concerned it is a blameless loudspeaker. The nearly constant horizontal directivity and perfect dipole symmetry (at least above 400Hz) results in an exceptionally focused and stable stereo image. Because the system is so narrow at the top it also does not stand in the way of it’s own back-wave reflection, I think this is very important for a dipole. A wide baffle will cast a shadow so to speak which will make it possible for your auditory system to (more easily) detect its position.

Yes, the system is horribly inefficient but the amplifiers and power supplies are all very efficient so the total power consumption in rest is just above 100W which is not bad for a system with a peak power of well over 3kW. The enormous peak power capability does provides horn like dynamics and ease of presentation. The operating temperature is somewhere between 30 and 40 degrees Celsius depending on output power. The system is very, very quiet, you have to press your ear against the Neo10 to be able to hear a faint noise. No hum, no clicks, no pop or crackle. The system is also free of any mechanical noise (humming or whistling).

So where’s the catch? Well, it won’t do an orchestra at full concert level despite the enormous amount of power available. The maximum SPL I measured on the my listening couch (3m distance, with acceptable distortion) was about 103dB which is more than enough for me and it also is enough for SMPTE RP 200, 83dBSPL average at the listening position with 20dB headroom (83dBSPL = -20dBFS). The limited vertical dispersion requires a minimum listening distance of 3m for proper integration of low,mid and high frequencies and sufficient high frequency dispersion. The system is certainly not child or pet proof and I would expect the WAF to be close to 0. And then there’s the cost, this particular implementation with the AT drivers and Ncore amps is very expensive but even with other amps and drivers this kind of system will never be cheap.

I noticed that you mentioned that the system is inefficient. I am trying to avoid this problem in my own systems by (A) using high sensitivity drivers such as pro audio woofers when possible and (B) using drivers close to their dipole peaks so that they are used where they are most efficient. To do this I need to construct a 3-way dipole plus a separate subwoofer for frequencies less than 100Hz. My 2017 nude 3-way test system was plenty loud (although I did not measure the SPL level), so I think I am on the right track.

I would not say the "WAF is near zero" as you claim. There is only so much WAF for these kind of systems, and they are DIY after all. I am a terrible wood worker and lack both tools and talent, but I can do basic cutting and routing. Luckily I don't need to do much of that sort of thing, but this leaves me with some basic wooden "frame" and drivers dangling by wires. I would say this provides much less WAF that you own effort!
 
The OB efficiency topic is always interesting.... And we are talking below 80 Hz. What sets final efficiency is based on what frequency, effective cancelation path distance, and to some extent, the woofer or woofers used.

Before we design anything, we need to set the woofer panel dimensions. After that, how low do we want or need to go before roll off begins. And how loud to we need to play where the roll off begins. The laws of physics take over and we have an answer. We may not like the answer, but we will have one. Passive, dsp, whatever it does not mater. The key element of the three above, is the path difference and that is more or less set by the baffle dimensions.

So, Charlie, what are your maximum baffle dimensions. Face & wing depth?
 
The OB efficiency topic is always interesting.... And we are talking below 80 Hz. What sets final efficiency is based on what frequency, effective cancelation path distance, and to some extent, the woofer or woofers used.

Before we design anything, we need to set the woofer panel dimensions. After that, how low do we want or need to go before roll off begins. And how loud to we need to play where the roll off begins. The laws of physics take over and we have an answer. We may not like the answer, but we will have one. Passive, dsp, whatever it does not mater. The key element of the three above, is the path difference and that is more or less set by the baffle dimensions.

So, Charlie, what are your maximum baffle dimensions. Face & wing depth?

That's a good approach. I have a sligthly different, but related one. It's explained HERE under "Load That Bass" but I will repeat the essence of it now.

I use an H-frame subwoofer. I would like this to go as "low" as possible and this in turn means I want the front-to-back pathlength to be as long as possible. But how long is "too long"? In an H-frame, a resonance forms in both front and back tunnel. It's a 1/4 wave resonance, and it forms a null. Given where I know the woofer (the driver in the band above the H-frame subwoofer) can play (down) to in frequency, I choose the length of the H-frame so that the response is just starting to turn down into the null at the frequency where I want to cross over to the woofer. Then I can use that as part of the H-frame low-pass function. Make the H-frame any deeper/longer and the null would move down into the H-frame's desired passband. This also gives me as much air loading as possible on the H-frame's driver and as I mentioned it helps to raise Qts and lower Fs.

So that's the basic approach I use. I end up with H-frames that are, front to back, around 24-36". I use an 18" woofer in the H-frame and the height and width is also around 24". This keeps the aspect ratio of the tunnels around 2:1 length to width, which helps to keep the Q of the 1/4-wavelength resonance down to a reasonable value. This kind of structure is getting a bit large, so I use one central H-frame and then I can use the top for my amp(s) and other gear. It's like a big table. You can put a granite top on it if you want it to look fancy - some weight to the H-frame is not a bad idea to counteract the force from the cone movement.
 
Last edited:
OK... Would you consider a 1% (92 dbw) a usable half space efficiency target assuming a 23"-25" net baffle width and a max effective overall depth (path difference) of approximate 15"-16"?

Also will assume 107 db clean output at or above 28-30 Hz half space per speaker would be low enough and loud enough to qualify as a full range residential speaker? At 32 - 33 Hz we can produce 1 acoustic watt per speaker, (109 db/mtr forward radiation) again in residential half space. At 20 Hz a pair can still reach 100 db. Those last 12 Hz (from 32 Hz to 20 Hz) are a real output killer! This I have built before.
 
The situation is not as cut and dry as it once seemed. Even at very low frequencies, dipole bass is still palpable as long as the dipole source can deliver the desired SPL target. As John Busch mentions above, getting below 30Hz or so is a real challenge...

Let me as you this: does Linkwitz's ultimate loudspeaker, the LX521, have a dipole source in the lowest band? That should tell you something. He was working on the Thor many years before the LX521 came along.