I have the CBT36K for my home theater and the problem with that is the tweeters. The crossover (dsp active) is at 1kHz which is too low for the 1/2 inch tweeter array. I hear sibilance when I listen to voices on the system. The question with the CBT35K is at what frequency do you lose horizontal coverage from the midranges. I am now trying out the JBL CBT 70J-1 for the home theater sitting on top of some 18 inch subwoofers (Parts Express kits). I am using a miniDSP Flex (balanced) for fine tuning of the 70J-1 and the subwoofers response.
-
Welcome to ASR. 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
- Thread starter RickS
- Start date
The only upcoming CBT from us (K&O Tech) is the one seen in the post below and it's not ready yet.
Constant Beamwidth Transducer (CBT) Speakers
I am curious about which direction represents the apparent source direction for the CBT speaker. When you consider that there is an acoustic 'mirror image' of this speaker below the floor level, it seems to me that the apparent midpoint/source is where the speaker touches the floor. Surely...
www.audiosciencereview.com
Regards,
Bjorn Omholt
It will be interesting to hear your take on the CBT 70J-1, once you think you've formed a solid opinion on it.
Yeah, 1kHz sounds too low for 1/2 tweeters.......I don't care how many are in the array. And the Legendre shading has to both reducing the total potential summation, and shifting the load to the lower drives in the line. Just one of the CBT tradeoffs, huh.
The problem with the lines I made using 24 "full-range" 3.5" TC-9 drivers, was with the two ends of the spectrum.
The drivers are too small to do any real bottom end below 100Hz, and too large to work as a line above above 2kHz. Common sense engineering results as you'd expect really.
But what a beautiful midrange they have!
It's amazing to me, once a person gets deeper into speaker design, how much you can tell about a speaker by simply knowing the driver sizes and crossover points.
So much information there....
Tis something I think the ASR speaker reviews are missing....driver sizes and numbers should be front page news imho.
Newman
Major Contributor
- Joined
- Jan 6, 2017
- Messages
- 4,976
- Likes
- 6,973
All too common to think a speaker's midrange is exceptional when the highs and lows are somewhat missing. Psychoacoustics at work.
Y~BNA
Member
Hi y'all, breakin into this thread a bit late, just having joined recently.
I see interesting discussions on issues i identified as being crucial for an impending project i'm drifting toward.
On shading/tapering, in amplitude as well as time domain.
Also on (assumptions of) cylindrical wave form, that according to various sources are only true subject to specific frequency, distance and other circumstances (due to source size, shape, and other factors).
Posted below a very informative white paper (IMO, tx to Dan Twomey on DiyAudio) by one James Griffin, dealing with al the caveats of living room line arrays (as opposed to Ureda, Toole, Button, etc. whose important work mainly concentrated on PA and other larger scale situations).
Found a somewhat sobering paper by Merlijn van Veen, putting the 2Pi instead of 4Pi theoretical assumptions in practical perspective.
Siggy (RIP) also posted some encouraging observations about line arrays in living rooms on his unmatchedly insightful website www.linkwitzlab.com, maintained by his son in law, but he unfortunately passed away before exploring his early insights any further.
Am now researching open baffle dipole line array possibilities, further complicating the horizontal integration also mentioned in this thread. Because i never want to build a boxed speaker again (for now, neva say...).
For that horizontal integration, point source two- or three ways would be a potential solution, but not in an OB dipole setup. And so forth.
So i thought it might be fun to seek assistance in having sleepless nights over these issues. I already have some, anyone wanna have some too, please?
I see interesting discussions on issues i identified as being crucial for an impending project i'm drifting toward.
On shading/tapering, in amplitude as well as time domain.
Also on (assumptions of) cylindrical wave form, that according to various sources are only true subject to specific frequency, distance and other circumstances (due to source size, shape, and other factors).
Posted below a very informative white paper (IMO, tx to Dan Twomey on DiyAudio) by one James Griffin, dealing with al the caveats of living room line arrays (as opposed to Ureda, Toole, Button, etc. whose important work mainly concentrated on PA and other larger scale situations).
Found a somewhat sobering paper by Merlijn van Veen, putting the 2Pi instead of 4Pi theoretical assumptions in practical perspective.
Siggy (RIP) also posted some encouraging observations about line arrays in living rooms on his unmatchedly insightful website www.linkwitzlab.com, maintained by his son in law, but he unfortunately passed away before exploring his early insights any further.
Am now researching open baffle dipole line array possibilities, further complicating the horizontal integration also mentioned in this thread. Because i never want to build a boxed speaker again (for now, neva say...).
For that horizontal integration, point source two- or three ways would be a potential solution, but not in an OB dipole setup. And so forth.
So i thought it might be fun to seek assistance in having sleepless nights over these issues. I already have some, anyone wanna have some too, please?
Last edited:
Newman
Major Contributor
- Joined
- Jan 6, 2017
- Messages
- 4,976
- Likes
- 6,973
The CBT line source is completely different to the typical line arrays that Griffin etc. So, much of what you might pick up from those sources are only tangential to the CBT.
Have you had a chance to look into the CBT itself?
cheers (and welcome!)
It would be wise to remember that all of these designs involve what is basically FIR filter design, of an unusual, but not terribly tricky, sort.
Keith_W
Master Contributor
you can do that with an FIR and also then add room correction if needed
You want different gains and shading at different frequencies, but you're missing the point that an array is identical to a short FIR filter, with the pattern being the dual of the frequency response.
So, the array itself is a kind of FIR, and you can use FIR's to shape the shading and delay of each element.
It's a big 2D filter design problem.
Y~BNA
Member
Hi Newman, tx!
I have not studied the CBT's in depth, but they been around a while and studying the general principles it becomes clear Keele had many insights into the factors at play, some of which long predated white papers later published on said factors.
So yes, i think they're very interesting, but i mainly grasp the amplitude shading and temporal shading that at play in the curving, which in OB (one of my main fields of interest) should both be obtained strictly in the DSP domain. Although on general principle i prefer to start of with a (physically) solid acoustical design, for dipole purposes physical curving is obviously not in the toolbox.
Frequency shading, which could be accomplished in OB with DSP as well, might also be at play, but seems hard to pin down. HF looses energy to air absorption sooner the higher the Hz, so frequency shading with distance already IS a natural factor, and would seem to be increased by the curve, but i don't know if this is a contributing or diminishing factor in the resulting wavefield the CBT's are famous for.
The curving does, from gut feeling, almost certainly play a role in mitigating the temporal smear inherent to unfiltered or uncurved line arrays, as demonstrated by the comments observing a point source projected BEHIND the CBT's. Whether the shading and the curving can mimic each other is therefore subject to my keen interest!
Pfff, long...
In short, i see CBT and LA in general are not to be compared one on one, but the curving, shading, tapering, etc. is going on in both, and for reasons that are obviously associated.
CBT was also an effort at finding the holy grail of Constant Directivity (Even Power Response, Equal Polar Pattern, etc) that drove and drives all the greats in audio.
My search combines OB with LA..
Cheers!
Y~BNA
Member
Yes definitely. Although a solid acoustical approach should always be the starting point (no amount of filtering can correct for bad design), we're pretty lucky to have DSP in general, and FIR in particular, around in our present time. To be applied sparingly and subtly, but to be exploited for the finer details of filtering definitely.
Cheers!
Last edited:
Y~BNA
Member
Perhaps its hailed phase linear properties? Although those would seem more applicable to regular line arrays and their time smearing. Keele's design seems to not suffer from that as much (although i never actually heard them, to my regret).
Also lobing and comb filtering, but i personally would prefer to address that by driver size, shape and placement/distancing, before applying filtering.
So yes, good question.
Last edited:
Keith_W
Master Contributor
Hmm, I am still struggling to understand what you mean. I think you are suggesting that we understand the array as akin to a short FIR filter, with each driver as a "tap" in a "delay line". Just as the impulse of a linphase filter is located in the middle of the tap length, the driver with the least delay and zero shading is located in the centre of the array. And you are not suggesting that FIR is required to realize the full potential of a CBT.
Have I understood you correctly?
gberchin
Active Member
- Joined
- May 11, 2016
- Messages
- 8
- Likes
- 69
If you haven't checked out my Dad's new website, please do. I've expanded the CBT page and included all 18 of his papers on CBTs, plus the original Navy papers. One can also browse all of his 83 Audio Magazine reviews (full magazines)
dbkeele.com
www.dbkeele.com
Kevin Keele
Constant Beamwidth Transducers (CBT Loudspeakers)
Top Benefits of Constant Beamwidth Transducer Loudspeakers invented by D.B. (Don Keele)
dbkeele.com
www.dbkeele.com
Kevin Keele
Y~BNA
Member
Wow, mr. Kevin Keele! Tx for joining, and more tx for making me aware of the treasures of knowledge no longer hidden to me/us, and for your efforts working on that and sharing it with us mortals. I feel really delighted and honored you entered this thread and shared this. Gonna check it out NOW!
Cheers.
Cheers.
I'm saying that it IS an FIR (the drivers themselves plus the driver radiation patterns, of course) and that an FIR (with perhaps some response adjustments for the driver), applied to each, to provide proper shading ACROSS FREQUENCY, is a good way to go. It does require amp per driver, of course.
For instance:
Newman
Major Contributor
- Joined
- Jan 6, 2017
- Messages
- 4,976
- Likes
- 6,973
Thanks to you for maintaining the online site and papers, much appreciated.If you haven't checked out my Dad's new website, please do. I've expanded the CBT page and included all 18 of his papers on CBTs, plus the original Navy papers. One can also browse all of his 83 Audio Magazine reviews (full magazines)
Constant Beamwidth Transducers (CBT Loudspeakers)
Top Benefits of Constant Beamwidth Transducer Loudspeakers invented by D.B. (Don Keele)
www.dbkeele.com
Kevin Keele
Newman
Major Contributor
- Joined
- Jan 6, 2017
- Messages
- 4,976
- Likes
- 6,973
In terms of the origins of the concept, DB Keele noted right at the start that it followed the release of classified (for 30 years before public release) military technology into underwater* sonar CBT, which he then applied to loudspeakers for sound reproduction. So, in a sense, it is new loudspeaker technology, not a deeper understanding of classical line source loudspeaker tech.
*hence the mention of the navy in Kevin Keele's post above.
Last edited: