I was thinking of what MAY happen inside the compression chamber ,but since
waves are going in circles and find their way outside , maybe I'm wrong
Maybe the drawbacks are behind the diaphragm
Ok ...too many uncertainties ! For sure , a CD cannot work properly without
the associated horn . Direct radiation tweeters , dome or cone or ring , would
benefit of a WG if the design requires it .
My latest 'designs'
all include a waveguidethat's why I'm here
to clarify, i understand and support the use of waveguides/horns to control dispersion. It just doesnt work in all listening environs. Keyser has my ideal listening space, mind and many others here in the UK are more like 20 to 40 sqm. That makes an awful lot of difference. Indeed Collums' is a read, but were his tests aligned to suit a US or UK listening room? Back on topic- one downside to waveguiding, as with horns, is multiple obtuse flare reflections, and thats before you consider throat reflections. Whether or not that is audible depends greatly on the listener and room characteristic, etc.
Last edited:
I was thinking of what MAY happen inside the compression chamber ,but since
waves are going in circles and find their way outside , maybe I'm wrong
Waves "going in circles"? A compression driver probably works slightly different
I do believe that 45° toe-in with 90° directivity is not desirable because of the strong contralateral reflection.
Do you have a clearer version of the horn contours image ? I've seen this article a few times before but I really struggle to identify which coloured line matches which label, (some look almost the same) and I'm interested in getting a better idea of which profile is which, especially the OS profile.
In most rooms that reflection is normally delayed by over 15ms, and thus perceived differently from an early reflection.
This is a complex subject and somewhat hard to generalize about. I have designed a number of CD horn type products including the orriginal JBL CD monitors, the 4430 and 4435. I have also put CD flares on a lot of dome tweeters, either straight conical flares with a break point to a second slope, or conical flares with a broad radius termination. These were all empirically derived but tended to work pretty well.
There are technical improvements that you can expect from any good CD device: The unit is less likely to see cabinet edges and will have smoother response due to to that. I've done in-wall coaxials with that approach that were dramatically better in terms of reflections from the cone behind. Certainly, with constant directivity response off axis can better track the axial response.
A flare on a tweeter can make the tweeter directivity more uniform and a better match to the woofer's directivity on the other side of the crossover. Power handling will go up nearly in proportion with the d.i. gain. This can be welcome and will also allow a slightly lower crossover point which can have knock on improvements to vertical directional issues.
If we are talking true horns with compression drivers, then I would certainly go with a CD horn rather than any other horn type. The polars are more uniform and the d.i. is more uniform so the frequency response on and off axis will generally be signifcantly better than any previous nonCD horn types.
Still, for domestic use I am not convinced that going from good dome drivers, or even domes with added flares, to horn compression drivers is all positives. Output capability is really good but response smoothness tends to be less so.
Most people assume that the flatter power resonse that you might achieve is the biggest benifit of CD units, but I don't think there is any evidence to back that up. First off, a CD horn does not ensure a CD system. Every system with CD drivers that I have seen has the usual slowly climbing directivity to some mid frequency where is can be more or less uniformly maintained by the upper range CD unit. It is not CD for 9 or 10 Octaves, it is CD only over the horns range. Secondly, many horns that claim to be CD are not truly constant. The only units that I have seen with true constant directivity are the JBL types, or similar, with a fairly narrow diffraction slot feeding a conical flare that includes a "midrange narrowing" secondary flare. Units with constantly changing side wall angles can have very smooth directivity but it will not be constant in directivity. Using such a driver with a typical woofer is likely to have rising directivity from the bottom to the top of the audio spectrum.
Another important issue will be what is the appropriate directivity for a CD system? Generally anything with a CD horn will have considerably higher directivity than any cone or dome system. Some people will enjoy the greater clarity that higher directivity gives, due to a general reduction in the rooms reverberent field. Others will find the sound more sterile with less involvement of the acoustics of the room. There are no absolutes in that area but we should remember that 2 channel reproduction is necessarily full of compomise.
My personal opinions are that very high directivity is not desirable. That 90 x 40 as a minimum should be a fair target. That any particular power response is far less important than achieving the best possible direct field response both on axis and over a wide potential listening window. To me that is the best attribute of CD done well, that, due to high polar uniformity, you have the potential to achieve very smooth and flat response at any likely listening position.
David S.
There are technical improvements that you can expect from any good CD device: The unit is less likely to see cabinet edges and will have smoother response due to to that. I've done in-wall coaxials with that approach that were dramatically better in terms of reflections from the cone behind. Certainly, with constant directivity response off axis can better track the axial response.
A flare on a tweeter can make the tweeter directivity more uniform and a better match to the woofer's directivity on the other side of the crossover. Power handling will go up nearly in proportion with the d.i. gain. This can be welcome and will also allow a slightly lower crossover point which can have knock on improvements to vertical directional issues.
If we are talking true horns with compression drivers, then I would certainly go with a CD horn rather than any other horn type. The polars are more uniform and the d.i. is more uniform so the frequency response on and off axis will generally be signifcantly better than any previous nonCD horn types.
Still, for domestic use I am not convinced that going from good dome drivers, or even domes with added flares, to horn compression drivers is all positives. Output capability is really good but response smoothness tends to be less so.
Most people assume that the flatter power resonse that you might achieve is the biggest benifit of CD units, but I don't think there is any evidence to back that up. First off, a CD horn does not ensure a CD system. Every system with CD drivers that I have seen has the usual slowly climbing directivity to some mid frequency where is can be more or less uniformly maintained by the upper range CD unit. It is not CD for 9 or 10 Octaves, it is CD only over the horns range. Secondly, many horns that claim to be CD are not truly constant. The only units that I have seen with true constant directivity are the JBL types, or similar, with a fairly narrow diffraction slot feeding a conical flare that includes a "midrange narrowing" secondary flare. Units with constantly changing side wall angles can have very smooth directivity but it will not be constant in directivity. Using such a driver with a typical woofer is likely to have rising directivity from the bottom to the top of the audio spectrum.
Another important issue will be what is the appropriate directivity for a CD system? Generally anything with a CD horn will have considerably higher directivity than any cone or dome system. Some people will enjoy the greater clarity that higher directivity gives, due to a general reduction in the rooms reverberent field. Others will find the sound more sterile with less involvement of the acoustics of the room. There are no absolutes in that area but we should remember that 2 channel reproduction is necessarily full of compomise.
My personal opinions are that very high directivity is not desirable. That 90 x 40 as a minimum should be a fair target. That any particular power response is far less important than achieving the best possible direct field response both on axis and over a wide potential listening window. To me that is the best attribute of CD done well, that, due to high polar uniformity, you have the potential to achieve very smooth and flat response at any likely listening position.
David S.
I did and it doesn't work for me.
Define "most rooms". In my room (4,47x10,63m) the contralateral reflection is delayed by 11ms.
In my opinion, the common 15ms/10dB rule is a gross oversimplification. There simply are no conclusive studies showing how reflection level and delay influence our perception of (phantom) sources with 2 speaker stereo.
to complicate my situation in my 5x4 plus 2x3m L shaped living area also has 45deg ceilings reaching down to 1.5m height on opposite walls. Add to that cement MDF floor, and double dry wall with green glue walls, and i think it would be VERY difficult indeed to predict the early reflection period and absorption coefficient.
True, you would need to measure. Even in perfectly rectangular rooms there can be very loud second or even third order reflections. But then again, which reflections are good and which ones are bad? How to measure? How to map those measurements to perception? We know a little bit about how (single) reflections change perceived spaciousness and size. Unfortunately not enough to know what particular speaker design is "best".
I haven't seen those particular curves (response as opposed to polar) but I assume so.
A cone or dome can be a perfect 2nd order high-pass but any horn will have to show some amount of the periodic ripple of a finite length device.
I know you can get around it in system design, but it is a factor.
David
It is pretty smooth, but the curves show an on-axis hole in the response caused by finite aperture diffraction. So the response in the listening window (0-30 degrees) is not as smooth as at larger angles.
True, but as bwaslo already said, the design axis is not 0°. Anyway, the question is if the response of "good dome drivers, or even domes with added flares" are smoother and/or more constant directivity. The data I've seen so far indicates that the opposite is true.
Why? Are you drawing the line that the average enthusiast will not want to use more than 6 components in their crossover?
I would rather draw the line at things that can't be repaired by a crossover, such as diffraction, and reflections and a good horn is no slouch with these.
A good recording will have all the ambience needed, the room would only fix it when it aint broke (or make it worse). On the other hand, many recordings are broken. For what it's worth though, I like the way a low reflection environment can help reveal fine detail in so many recordings that sound so bad you'd hardly believe there was detail in them, and that combined with some forced EQ a lot of these can become listenable.
Isn't achieving the best possible on-axis response over a wide area reasonably analogous with a power response?
I have calculated power for axial slices in order to verify where the power is, and I found that in my case the bulk of it is within 20 degrees either side of my chosen listening axis. 20 degrees one way will have a falling top end, and 20 degrees the other way will be falling at the lower end. All other plots were down.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.