Without PWT testing can only surmise it's due to the dramatic difference in (hornless) polar response combined with the 18 Sound's more throat 'choking' WG.
That's because most commercial horns have SERIOUS flaws and/or shortcomings, AND they are rarely matched properly (b/c of discontinuity of flare rate, diffraction-inducing discontinuitites in profile, etc.).
Having said that, a compression driver operated with a competently designed horn that is also PROPERLY MATCHED to its characteristics (phase plug, internal throat flare rate, etc.) will ALWAYS sound objectively much better (i.e., lower distortion, less diffraction, etc.) than the same driver "hornless".
Alas, not even this is a guarantee of a proper match. Most PA horns are designed for constant directivity above all, with little regard to other important factors that co-determine overall sound "quality".
If one is dead set on using a compression driver without a horn, I would suggest that the least bad way of doing it is to cover the driver exit (the side intended to feed into a horn) with a layer of acoustic foam, then flip the driver, remove the back cap, and use it as a large dome mid-tweeter.
A reputable recording studio somewhere in Scandinavia once built a set of large soffit-mounted recording & mastering monitors with large-format JBL drivers used that way.
[EDIT]: It was PUK studios, in Denmark. You can see the 4" JBL domes (flipped compression drivers) in the picture, they are the seemingly tiny silver disks in that blue "wall of sound". For scale, the side sub-woofers are 80cm Fostex FW800 units (four of them in total!).
A reputable recording studio somewhere in Scandinavia once built a set of large soffit-mounted recording & mastering monitors with large-format JBL drivers used that way.
[EDIT]: It was PUK studios, in Denmark. You can see the 4" JBL domes (flipped compression drivers) in the picture, they are the seemingly tiny silver disks in that blue "wall of sound". For scale, the side sub-woofers are 80cm Fostex FW800 units (four of them in total!).
Last edited:
A bit more detail on the PUK studios monitoring system in this PDF.
"For the mid range we tried a lot of horns. They had to be very big which gave a lot of reflections and phase cancellation problems. Finally we decide not to use horns, but rather a large dome tweeter. A normal JBL loudspeaker has a 4 in dome tweeter with a titanium membrane which works very well."
That 4" dome was that of the JBL 2445J compression driver.
The design engineer also published an AES paper with even more detail.
"For the mid range we tried a lot of horns. They had to be very big which gave a lot of reflections and phase cancellation problems. Finally we decide not to use horns, but rather a large dome tweeter. A normal JBL loudspeaker has a 4 in dome tweeter with a titanium membrane which works very well."
That 4" dome was that of the JBL 2445J compression driver.
The design engineer also published an AES paper with even more detail.
Last edited:
The Eighteen Sound ND3ST on the XR1464C horn has about 6dB less high frequency than the Beyma TPL200 pleated diaphragm driver/horn.
The ND3ST without a horn likely has a rising high frequency response.
Such a large difference in on-axis frequency response alone ("something lost") could easily explain the "by ear" preference of you and your friends .
JonBocani, the original poster of this thread wrote that he had sold the "hornless" Radian 950PBbe a while before the time he posted this:
https://www.diyaudio.com/community/...200hz-no-audible-difference-whatsover.353554/
Try the same A/B test with the response equalized to the same target and you may also find your preference shifts.
Art
Thank you for all the comments!
I would tend to rule out that my preference is due to differences in response above 10,000 Herz: I even did the test, just to rule out this hypothesis, of cutting the drivers to 10,000 and using a supertweeter (bw 48db active). Also (perhaps I forgot to mention) in all the tests or equalized the system to have a flat response up to 20,000.
In the meantime, I did some more tests taking some friends along. I would tend to rule out the presence of any particular diffraction problems or HOMs.
Further listening would seem to point to more pleasantness of the wider dispersion of the driver without a horn and a slight but still recognizable and annoying nasal or “pipe” sound that cannot be avoided by using this horn.
I have noticed that in the case of one-inch drivers used from 2 kHz this phenomenon does not occur: they sound significantly better with the horn than without. In further confirmation of this phenomenon it would seem that even here by moving the crossover point around 2000 the driver begins to sound just as good with the horn. Perhaps the depth of the horn could cause the “pipe” sound. Mind you, it is not very nasal, but it is nasal enough to bother people like yours truly who like this kind of sound little.
I would tend to rule out that my preference is due to differences in response above 10,000 Herz: I even did the test, just to rule out this hypothesis, of cutting the drivers to 10,000 and using a supertweeter (bw 48db active). Also (perhaps I forgot to mention) in all the tests or equalized the system to have a flat response up to 20,000.
In the meantime, I did some more tests taking some friends along. I would tend to rule out the presence of any particular diffraction problems or HOMs.
Further listening would seem to point to more pleasantness of the wider dispersion of the driver without a horn and a slight but still recognizable and annoying nasal or “pipe” sound that cannot be avoided by using this horn.
I have noticed that in the case of one-inch drivers used from 2 kHz this phenomenon does not occur: they sound significantly better with the horn than without. In further confirmation of this phenomenon it would seem that even here by moving the crossover point around 2000 the driver begins to sound just as good with the horn. Perhaps the depth of the horn could cause the “pipe” sound. Mind you, it is not very nasal, but it is nasal enough to bother people like yours truly who like this kind of sound little.
Plane Wave Tube, historically the standard for testing CDs. As I noted, the dramatic difference of the driver's on/off axis polar response combined with a WG 'choking'/'constricting' it to an 80 x 60 deg. over an acoustically short distance compared to a much slower expanding flare such as expo.
The differences are very subtle and perhaps difficult to perceive in these videos recorded with a cell phone, nevertheless I try to send you clips.The hope is that you too can get an idea. I must admit that listening back to the clips, especially if you are not using headphones, the differences are really minimal.
The sound is a little bright because the EQ is absolutely flat and in close listening like this the sound is a little too unbalanced on the treble. Also keep in mind that the first song has perceptible sibilance with any type of system.
And on a first listen with the horn it would seem to sound more "in your face," especially in terms of vocals. Likewise it would seem to perceive more fullness in the lower part of produced by the horn. All this seems to me more than logical and also appreciable.
What is more difficult to perceive is the part related to the greater dispersion of the driver without the horn, which in live listening is particularly perceptible and pleasant, also the ”pipe” sound would seem to be less perceptible and unpleasant in the clip But if you listen carefully to the voice it is there, very subtle but there.
ps: the woofer is a jbl 2226j, the crossover points are 1000 and 10000 48db butterworth.
the horn is this: https://www.eighteensound.it/en/products/horn/1-4/0/XT1464
The sound is a little bright because the EQ is absolutely flat and in close listening like this the sound is a little too unbalanced on the treble. Also keep in mind that the first song has perceptible sibilance with any type of system.
And on a first listen with the horn it would seem to sound more "in your face," especially in terms of vocals. Likewise it would seem to perceive more fullness in the lower part of produced by the horn. All this seems to me more than logical and also appreciable.
What is more difficult to perceive is the part related to the greater dispersion of the driver without the horn, which in live listening is particularly perceptible and pleasant, also the ”pipe” sound would seem to be less perceptible and unpleasant in the clip But if you listen carefully to the voice it is there, very subtle but there.
ps: the woofer is a jbl 2226j, the crossover points are 1000 and 10000 48db butterworth.
the horn is this: https://www.eighteensound.it/en/products/horn/1-4/0/XT1464
Last edited:
Spot on, you can't fool 'Mother Nature' in that we either 'dance to her tune' or we pay the price in various distorted ways, leaving our 'job' to finding the best way to 'fool' ourselves while otherwise meeting the needs of the app. 
a compression driver used with the back of the dome is no longer a compression driver for me or i'm wrong ? and if you in the normal way without a horn your horn is just the conical throat very short horn but still a horn..
It is in the sense that it's now a BR tuned way above its Fs or even more so if its throat/vent is plugged. Correct.
edit: Note that a compression horn is composed of an under damped driver (Qts' > ~0.403) coupled to an over damped (heavy air mass 'plug') vent that when properly designed as a system with be 'critically' damped over the driver's designed BW.
edit: Note that a compression horn is composed of an under damped driver (Qts' > ~0.403) coupled to an over damped (heavy air mass 'plug') vent that when properly designed as a system with be 'critically' damped over the driver's designed BW.
I don’t understant what you mean.Spot on, you can't fool 'Mother Nature' in that we either 'dance to her tune' or we pay the price in various distorted ways, leaving our 'job' to finding the best way to 'fool' ourselves while otherwise meeting the needs of the app.
Although a little extra distortion without using the horn is there, my measurements show that at these listening volumes the difference in thd and imd is negligible. If there is a price to pay and of course there will be because, as you say you can't fool nature, that price at these listening levels is not particularly big. I, however, would like to think in scientific terms and not stay exclusively on dogmatic arguments. The horn+driver system was developed for different uses and for those uses extract the most out of the system. Here, however, we are talking about completely different uses, and different sound levels and needs in terms of long-distance sound projection that are completely antithetical to my setup.
It follows that although no one doubts that these systems were developed to work together and that the person who designed them is a person of great ability, nonstandard use in situations different from those for which driver+horn were designed may not be heretical.
I would suggest thinking with as much freedom of mind as possible, while remaining firmly planted on the ground in a scientific approach. No one would dream of denying that a fig plant serves to produce figs, nevertheless if I cut a stick of it I can also use it to help me walk.
Greets! First Time that I appreciate ( also because the photos are Indeed movies! Surprise! Surprise!) A recording showing some [dramatic] ...
A request for my personal curiosity?
A cilynder truncated diagonally..yes, a Karlon [ tweeter] !?
A request for my personal curiosity?
A cilynder truncated diagonally..yes, a Karlon [ tweeter] !?
Beyma cp21 used as supertweeter to exclude the >10000hz radiation out of the equation. also excluding the differences in the radiation pattern above 10000hz the phenomenon is still audible.