I'm planning on buying a pair of CBT36 speakers and I'm designing a power amp for it. I've read that they are not that sensitive but precise info is vague. I'm trying to gauge what the peak wattage of the power amplifier I'm going to need to be for these speakers. Does anyone have any info on the sensitivity of the CBT36 speakers?
http://www.daytonaudio.com/media/resources/301-980-epique-cbt36-assembly-manual.pdf
Also see Fig. 32.
14.1.12.4. Sensitivity (2.83 Vrms/1m):
Note: the raw sensitivity (no crossover or EQ) of the CBT36 is frequency dependent.
It is roughly flat from 80 to 300 Hz and then rolls off at 3 dB/octave (10 dB/decade) up to 20 kHz.
See later section “CBT36 Power Rolloff” in Appendix 2 for further explanation of this rolloff.
For more details see Fig. 26 in this section.
Here are some approximate sensitivity numbers at different frequencies:
80 to 300 Hz: 94 dB
800 Hz: 89 dB
8 kHz: 79 dB
Also see Fig. 32.
14.1.12.4. Sensitivity (2.83 Vrms/1m):
Note: the raw sensitivity (no crossover or EQ) of the CBT36 is frequency dependent.
It is roughly flat from 80 to 300 Hz and then rolls off at 3 dB/octave (10 dB/decade) up to 20 kHz.
See later section “CBT36 Power Rolloff” in Appendix 2 for further explanation of this rolloff.
For more details see Fig. 26 in this section.
Here are some approximate sensitivity numbers at different frequencies:
80 to 300 Hz: 94 dB
800 Hz: 89 dB
8 kHz: 79 dB
hellokitty123,
I've heard Don Keele's prototype CBT36 at Parts Express events a few years ago but that space was a very large room. The public event did not offer me time to move around and have a long listening session but they sounded impressive for a first time non-critical listen.
More recent in the past 6 months I have completed my modified CBT24 (think of the Don's CBT24 (Dayton Audio Epique arrays) but with extensive mods but a 4 ohms SB Acoustics SB65 driver vs. the Epique 16 ohms unit. You can read about my efforts at:
My New Line Array--It's a Modified CBT24
I find that my CBT arrays (should be true for all of Keele's CBT arrays and even straight line arrays) produce virtually constant front to back volume within my 23 x 32 feet listening room. Think of a typical 3 dB per doubling of the distance vs. the usual 6 dB for a point source. Furthermore, with the CBTs you have a broader sound space within the room as the curved and weighted CBT array yields excellent polar plots. Keele has drawings of his CBT polar plots in his vast writings and you observe broad eye shaped plots with pinched sides.
Bottom line on CBT sensitivity is that they have significantly less sound fall off in-room so extreme sound levels aren't necessarily needed.
Other listeners have mentioned that the CBT36 weak point is the performance of its small tweeter. You might also consider the CBT24 as an alternative design in your thoughts.
Jim
I've heard Don Keele's prototype CBT36 at Parts Express events a few years ago but that space was a very large room. The public event did not offer me time to move around and have a long listening session but they sounded impressive for a first time non-critical listen.
More recent in the past 6 months I have completed my modified CBT24 (think of the Don's CBT24 (Dayton Audio Epique arrays) but with extensive mods but a 4 ohms SB Acoustics SB65 driver vs. the Epique 16 ohms unit. You can read about my efforts at:
My New Line Array--It's a Modified CBT24
I find that my CBT arrays (should be true for all of Keele's CBT arrays and even straight line arrays) produce virtually constant front to back volume within my 23 x 32 feet listening room. Think of a typical 3 dB per doubling of the distance vs. the usual 6 dB for a point source. Furthermore, with the CBTs you have a broader sound space within the room as the curved and weighted CBT array yields excellent polar plots. Keele has drawings of his CBT polar plots in his vast writings and you observe broad eye shaped plots with pinched sides.
Bottom line on CBT sensitivity is that they have significantly less sound fall off in-room so extreme sound levels aren't necessarily needed.
Other listeners have mentioned that the CBT36 weak point is the performance of its small tweeter. You might also consider the CBT24 as an alternative design in your thoughts.
Jim
Last edited:
Hmmmmm. Interesting. I'm surprised the tweeter is considered the weak point considering how there are so many of them in the array. Is there a solution to this tweeter deficiency? What's the practical difference between the CBT36 and the CBT24? The CBT36 seems better. Can the array be configured to be at a lower impedance than 4 ohms, I.E. more parallelism within the array? I want to keep my power amp design in full class A, the lower the voltage headroom I require the better. I recall Nelson pass saying at some point that current source amplifiers are really good for driving full range drivers. Does this apply here?
Last edited:
As I write this, I'm listening to a CBT + waveguide array.
The fundamental defect with arrays is that the high frequencies are limited by a single unit.
Yes, you read that right.
When the frequencies become smaller than the smallest driver, even with a million drivers, the maximum output is limited by the output of a single driver.
The reason that this happens is because the wavelengths are so short. For instance, 13.5kzh is an inch long. That means that if you want two tweeters to constructively sum, they need to be within a fraction of an inch. Obviously, that's impossible, so arrays exhibit a pattern of extreme comb filtering at high frequencies, and as you add additional drivers, it doesn't make much difference, because the problem is simply geometric.
The good news is that most loudspeakers are limited by displacement, and you'd be surprised how loud an array can get, due to to all the drivers summing constructively at low frequency.
But, yeah, high frequencies are tricky.
The fundamental defect with arrays is that the high frequencies are limited by a single unit.
Yes, you read that right.
When the frequencies become smaller than the smallest driver, even with a million drivers, the maximum output is limited by the output of a single driver.
The reason that this happens is because the wavelengths are so short. For instance, 13.5kzh is an inch long. That means that if you want two tweeters to constructively sum, they need to be within a fraction of an inch. Obviously, that's impossible, so arrays exhibit a pattern of extreme comb filtering at high frequencies, and as you add additional drivers, it doesn't make much difference, because the problem is simply geometric.
The good news is that most loudspeakers are limited by displacement, and you'd be surprised how loud an array can get, due to to all the drivers summing constructively at low frequency.
But, yeah, high frequencies are tricky.
Last edited:
Hmmm, that sounds like a significant issue. This will be my first pair of "high end" speakers, such an issue leaves me wondering if I should be looking at another solution. I do like my treble to be high quality.
I wonder if I should make a full range ribbon or something. Does CBT apply to ribbons?
I wonder if I should make a full range ribbon or something. Does CBT apply to ribbons?
A practically sized CBT array with ribbons does not work. Ribbons are too large straight pieces that cannot follow the curved cabinet. For pro audio arrays, where curvature is lower, it can work.
Figure 35 of the same paper shows polars at 12.5 and 16 kHz for the 0.7" driver prototype. Behaviour at (and below) 12.5 kHz is fine, at 16 kHz it is losing pattern control.
A CBT36 has similarly sized high frequency drivers as the 0.7" drivers of the paper, so pattern control will be similar.
Page 12, figure 22 of Keele (2003-10 AES Preprint) - CBT Paper4 shows that the effect is quite pronounced on his large driver prototype with 2.25" / 57 mm drivers and happens around 6-7 kHz. It does not show up on his measurement of the smaller driver prototype with 0.7" / 17.8 mm drivers, for which ~20 kHz would be expected, based on the previous result and dimensions. His measurements do not fully go up to 20 kHz though. Most people do not hear well above 15 - 20 kHz so this suggests that comb filtering is not a problem in this example.
Figure 35 of the same paper shows polars at 12.5 and 16 kHz for the 0.7" driver prototype. Behaviour at (and below) 12.5 kHz is fine, at 16 kHz it is losing pattern control.
A CBT36 has similarly sized high frequency drivers as the 0.7" drivers of the paper, so pattern control will be similar.
So what amps are you using with CBT36? I am forced to use 4x400W and I know it is an overkill but this is what I have available at the moment. I want to switch to hypex modules and wonder what modules would be proper?
If you manage to find 0.7" ribbons then indeed that should work. I do not know whether such small ones exist, though larger ones might even work if their radiating surface covers > 80% of their center-center spacings, if I remember correctly.
I'm starting to lean toward the possibility of the CBT24
Quoted sensitivity here
How does it scale as you add on wattage? How loud would 16W be here?
Quoted sensitivity here
So 2.83Vrms into 4 ohms is 4 watts and double that is about 16W.
How does it scale as you add on wattage? How loud would 16W be here?
As noted on page one of this thread, those numbers are the midrange efficiency of the CBT array. As you go higher and higher in frequency with a CBT, the efficiency gets lower and lower. The specs are on the first page of this thread.
I think you made a small error, P = U^2 / R with P = power, V = voltage and R (nominal) resistance which would make 2.83V in 4 ohms to be 2 watts.
To find the gain in decibels when using more power, use SPL2 = SPL1 + 10 * 10log (P2 / P1) where 10log is de logarithm with base 10, SPL2 the sound pressure level at power level P2 and SPL1 the sound pressure level at power level P1.
Ah right, I did miscalculate. I was running numbers for a lot of different things I must have got something mixed up.
I ran your equation and came up with 129. That doesn't seem right though. I must he off on something.
I ran your equation and came up with 129. That doesn't seem right though. I must he off on something.
Do you think I could get away with a 10W amp in a 200 square foot room using the CBT24 or CBT36?
The CBT36 requires bi-amplification, with it's sensitivity it could reach about 101 dB at 1 meter with 10 watts from 80 to 300 Hz, and about 86 dB at 8 KHz.
The CBT 24 could hit around 97dB mid band with 10 watts. SPL drops about 5 dB at 3 meters.
You could "get away" with that level, but would be leaving over 10 dB of the dynamic range available unused, they could easily sound more than "twice as loud" and still be very clean with 100 watts available.
That said, there is nobody forcing you to use amps with the headroom that would allow live music levels in your 200 square foot room ;^).
The CBT 24 could hit around 97dB mid band with 10 watts. SPL drops about 5 dB at 3 meters.
You could "get away" with that level, but would be leaving over 10 dB of the dynamic range available unused, they could easily sound more than "twice as loud" and still be very clean with 100 watts available.
That said, there is nobody forcing you to use amps with the headroom that would allow live music levels in your 200 square foot room ;^).
Last edited:
Oh okay. Thanks for your answer. I'm clearly not competent enough @ math to complete the equation above. 😱
Guys this thread is quite old now but I thought to add my experience with CBT36 powered by hypex fusion amps fa252. They play incredibly loud but you can clip the amps when not using a separate subwoofer. It is loud enough for me though..
More important in relation to the tweeter, I increased the cross-over frequency to 3khz because I had audible (and measurable) odd order harmonic distortion between 1000-2000 hz. When I increased the cross-over frequency to 3khz these harmonic distortions were gone! The sound quality improved A LOT!! I shared my findings with Don Keele, but he has yet to reply.
See measurements attached.
More important in relation to the tweeter, I increased the cross-over frequency to 3khz because I had audible (and measurable) odd order harmonic distortion between 1000-2000 hz. When I increased the cross-over frequency to 3khz these harmonic distortions were gone! The sound quality improved A LOT!! I shared my findings with Don Keele, but he has yet to reply.
See measurements attached.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.