• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Constant Beamwidth Transducer (CBT) Speakers

Looks like I may have bought a future collector's item as the remaining holdout (CBT24 kit) at Parts Express has gone NLA. This pretty much leaves the JBL Pro versions as the closest cousin left on the market.

@Bjorn can you update on your plans please?
 
Winkleswizard said:
Not sure how many members have CBT speakers, but I know a few have shown interest. Since there have been scattered ongoing discussions, am creating this thread to give it a home. If you are wondering what CBTs are, they are the work of former JBL and Audio Artistry engineer Don Keele. They offer some major benefits (more uniform sound distribution and better room integration) that are (IMO) way beyond traditional monopole speakers.

Here is an example pic and Keele’s website: http://www.xlrtechs.com/dbkeele.com/CBT.php
Click to expand...

I'm currently using them with a pair of Rythmik L22s, but I had posted this to Audio Asylum last year (https://www.audioasylum.com/cgi/vt.mpl?f=speakers&m=363550): "I also have a pair, which I set up with the MiniDSP SHD (I also bought the 2x4 HD software plug-in and copied over the recommended EQ settings to the SHD plug-in) and eventually supplemented with an Infinity Cascade Model 15 subwoofer that I had purchased on closeout almost ten years ago but only hardly used, which I positioned between the speakers. I am currently using a Nord NCore MP amplifier but haven't pushed the power very much at this point.

They are pulled up a bit close to the wall behind them, since they're flanking the fireplace in our family room, but they work really well in that area because of their reduced sound pressure fall-off with distance up to about 10-12 feet, as well as their extremely wide dispersion characteristics. The near-uniform frequency response at different heights also suits our family members' differing heights (including a five- and eight-year-old).

I had originally intended to use a pair of subwoofers, one behind each speaker, but the single subwoofer between the speakers works quite well, especially since I set up the electronics in a bookshelf several feet to the side of and behind the left speaker.

I haven't gotten around to measuring with REW and using Dirac yet, but frankly, they're so enjoyable for the time being that I'm in no hurry."

Here's an old pic, have changed some stuff around since then.

thumbnail_IMG_2580.jpg
 
I use DSP + individual amplification to run my home made CBT. My standards aren't audiophile they are...sounds ok.

I've used this arrangement to convert the CBT between 5, 4 and 3 banks of speaker groups -- just change the wiring. It is set up for 4 banks in the photograph. It also lets me switch between Legendre, Cosine, and Chebychev shading schemes. As others have pointed out, shading is the decrease in speaker volume level (dB SPL) as you go up the array. Switching between Legendre, Cosine, and Chebychev shading functions influences the horizontal directivity.

It also happens to be a flat CBT. Each bank is time delayed to approximate a physical arc. I can switch it back and forth between an array with no time delay or shading and a CBT array.

In my opinion, the CBT (delay + shading) sounds better than a regular flat array. Audiophiles might describe it as "the music opens up." But I'm not an audiophile and I have a hard time understanding their language so I might not have communicated that correctly. All the CBT performance descriptions you hear from other members are true, e.g. the speaker volume level remains constant as you walk up to the speaker.

I can't say any more because this is a top secret project.

Just kidding. It's not a top secret project it's just that I've reached the limits of my understanding. I wish people would DIY these so they could improve them so I could copy them and make mine better. Although, it easily meets my standard of "sounds ok." And it was gawdawful easy to build: it's just wires screwed into terminals. And inexpensive: those amps were $1.50 each from China, $18 per each CBT array. Granted, those cheap amps aren't struggling to runs watts through passive crossovers and low frequency sub woofers. Neither are they working alone to fill a room with sound, they're team players.

Thanks to the flexibility of the design I have been able to try different delays and shading schemes. Keele's shading and delays work as he claims.
 

Attachments

  • IMG_20200906_124824046.jpg
    IMG_20200906_124824046.jpg
    867 KB · Views: 407
BradleyPNW said:
I use DSP + individual amplification to run my home made CBT. My standards aren't audiophile they are...sounds ok.

I've used this arrangement to convert the CBT between 5, 4 and 3 banks of speaker groups -- just change the wiring. It is set up for 4 banks in the photograph. It also lets me switch between Legendre, Cosine, and Chebychev shading schemes. As others have pointed out, shading is the decrease in speaker volume level (dB SPL) as you go up the array. Switching between Legendre, Cosine, and Chebychev shading functions influences the horizontal directivity.

It also happens to be a flat CBT. Each bank is time delayed to approximate a physical arc. I can switch it back and forth between an array with no time delay or shading and a CBT array.

In my opinion, the CBT (delay + shading) sounds better than a regular flat array. Audiophiles might describe it as "the music opens up." But I'm not an audiophile and I have a hard time understanding their language so I might not have communicated that correctly. All the CBT performance descriptions you hear from other members are true, e.g. the speaker volume level remains constant as you walk up to the speaker.

I can't say any more because this is a top secret project.

Just kidding. It's not a top secret project it's just that I've reached the limits of my understanding. I wish people would DIY these so they could improve them so I could copy them and make mine better. Although, it easily meets my standard of "sounds ok." And it was gawdawful easy to build: it's just wires screwed into terminals. And inexpensive: those amps were $1.50 each from China, $18 per each CBT array. Granted, those cheap amps aren't struggling to runs watts through passive crossovers and low frequency sub woofers. Neither are they working alone to fill a room with sound, they're team players.

Thanks to the flexibility of the design I have been able to try different delays and shading schemes. Keele's shading and delays work as he claims.
Click to expand...

That looks pretty damn cool. What drivers are you using?
 
You might also look into using an optimization program to design a shading and filtering scheme. This has been known to work really well, but it is a touch more complex, and requires quite a bit of DSP.
 
maverickronin said:
That looks pretty damn cool. What drivers are you using?
Click to expand...

This is going to get a little weird. I experimented with exciters instead of cones. It's a tall thin dipole panel array. I wanted to see if the characteristics of panels reinforced the characteristics of the CBT.

The panels generate a strange looking frequency response graph in REW because they vibrate. When you smooth the graph it looks like the off axis is closely aligned with the on axis in high frequencies. However, I don't know if that's real because REW is being asked to smooth chaos. According to my ear, it's real; they sound the way I expected them to sound and do what I wanted them to do. These are raw measurements without equalization. That's why there's a large hump at 2.1kHz. I have to use DSP equalization (gain, PEQ) to get a flat frequency response.

In contrast, Jim Griffin, Ph.D. built a traditional physically curved CBT24 in 2017 using cone speakers. He used SB Acoustics SB65 drivers. His array uses 3 banks with passive shading rather than DSP. I think he worked that part out with Don Keele. At the very least, Don Keele supplied consulting. You can attach his arrays to a regular amplifier and they will work. He's very happy with those drivers and the end result. The fact that he's happy matters because he has expertise in line arrays, for example https://audioroundtable.com/misc/nflawp.pdf

Jim's design is a sure thing if you want to be certain of success.

IMG_20200812_101849302.jpgIMG_20200812_102231266.jpgLegendre - raw.JPGLegendre - smoothed.JPG
 
Last edited:
Scgorg said:
Is there any target price? I've been seriously considering CBTs, but there are so few commercial ones, and I do not have the time to build one myself currently.
Click to expand...
Everything is relative but it will not be a cheap CBT. The manufacture cost of this CBT is quite high. On the other hand, compared to high-end speakers in the market and in regards to how it measures and the complex cabinet construction I would say the cost will be very low.

I can't give an accurate price now but a rough target price is 55 000 - 65 000 NOK excl. tax. Which is $6150 - $7266 with today's currency. Shipping cost needs to be added.

It's possible that we may be interested in some conceivers/early adopters who will get a discount by pre-ordering the speaker. Sort of like a few kickstarters. But it's bit early to start with that now.

Take note that this speaker will require a external crossover unit. So that will come additionally. The crossover will also function as the DAC and preamplifier and will have very low distortion. 4-channels amplification is needed to run the speakers and can be chosen by the customer. So it's really a complete system and where subwoofer(s) can be added if desired and signal aligned with the DSP unit (not automatically though). But it will work great without subwoofer. The woofers we are using is of very high quality and can go low with the number.
 
Bjorn said:
Everything is relative but it will not be a cheap CBT. The manufacture cost of this CBT is quite high. On the other hand, compared to high-end speakers in the market and in regards to how it measures and the complex cabinet construction I would say the cost will be very low.

I can't give an accurate price now but a rough target price is 55 000 - 65 000 NOK excl. tax. Which is $6150 - $7266 with today's currency. Shipping cost needs to be added.

It's possible that we may be interested in some conceivers/early adopters who will get a discount by pre-ordering the speaker. Sort of like a few kickstarters. But it's bit early to start with that now.

Take note that this speaker will require a external crossover unit. So that will come additionally. The crossover will also function as the DAC and preamplifier and will have very low distortion. 4-channels amplification is needed to run the speakers and can be chosen by the customer. So it's really a complete system and where subwoofer(s) can be added if desired and signal aligned with the DSP unit (not automatically though). But it will work great without subwoofer. The woofers we are using is of very high quality and can go low with the number.
Click to expand...
Exciting stuff! The price sounds reasonable enough to me at least. You're located in Bergen, correct? Whenever the speaker releases I'd love to take the drive for a demo.
 
They will be demoed in Bergen and later at hifi shows.

Looking at some of the prices of high-end speakers now, I guess it's an absolutely bargain even if the cost ends up being quite a bit higher.
Kef Blade $32k: https://www.amazon.com/KEF-Blade-Snow-White-Pair/dp/B009K5C8S8
Revel Ultima Salon2 $22k: https://www.musicdirect.com/speakers/Revel-Salon2-Tower-Speakers

One of the reasons we can sell as such low cost is because of direct sales. Price will go up if we start to sell through retailers.
 
BradleyPNW said:
This is going to get a little weird. I experimented with exciters instead of cones. It's a tall thin dipole panel array. I wanted to see if the characteristics of panels reinforced the characteristics of the CBT.
Click to expand...

That's even cooler. :cool:
 
Was just perusing the latest edition of Voice Coil and the editor called out that Don Keele has recently been recognized for his contributions to constant directivity (work on horns from 50 years ago). His work on CBTs was also highlighted. When Don was asked, he lamented how long the industry seems to take to adopt new technology. Here is a excerpt from the article...

In a recent exchange I had with Don Keele, regarding the recent TECnology Hall of Fame announcement, he highlighted this same enthusiasm for the concept, which he sees as a natural evolution of that original Constant Directivity work that was now inducted. "For the last 20 years, my current research topic is CBT technology, which I pioneered (along with the US military in underwater sound in the late 1970s). I have written over 17 papers on the topic and given many presentations. This topic far out shadows my work on horns from 50 years ago (I’m 80 now!)."

More importantly, Keele regrets that, like so many important loudspeaker innovations, that work remains to be explored to its full potential. "I believe, my CBT work on constant directivity/coverage loudspeaker line arrays is truly revolutionary and it’s too bad that not many manufacturers haven’t picked up on the topic! It’s a shame that it hasn’t taken hold in the industry!"
 
You must log in or register to reply here.

Similar threads

Y
Concert hall acoustics links and excerpts
Replies
17
Views
4K
Smaestro
S
Back
Top Bottom