Perhaps not "purely", but "dominantly". It should be obvious why it can't be symmetric.
That is precisely what Nate Hanson did over on another thread. It works very well.
ok please explain how air compression produces 2nd order distortion and why it's considered symmetric.
Its not symmetric, that's the point. Air nonlinearity is easiest thought of as an increase in the wave speed with the pressure. (This is not precisely correct, but correct enough for our example.) This means that the wave speeds up as the pressure goes up and slows down as the pressure falls (about the ambient point,) which is completely non-symmetric. So no matter what terms are required in the Taylor expansion, they have to be even (all odd terms are symmetric.)
Update: The wave speed change is perhaps not the best way to think about the nonlinearity in a compression driver. The wave speed analogy is more applicable to a wave moving through a nonlinear medium, but in a compression driver this nonlinearity virtually all occurs right at the diaphragm. In the compression driver the major nonlinearity results from the fact that the PV curve is not linear since PV=nRT which implies that the pressure P change will be inversely related to the volume change, i.e. P = c/V where c is a constant. For small changes in V this nonlinear form can be linearized, but for large changes it is not and it is never symmetrical.
so it's like doppler distortion but at audio speeds.
pitch increases under pressure and decreases on dilation, like the train whistle analogy, is it not the variation of harmonics the thing that changes from consonance to disonance? but unlike the train who's whistle variation in apparent pitch is dependent upon direction of travel towards or away from us it's now occurring on every 1/2 cycle(as in positive cycles produce even order effects and negative cycles produce odd order)
pitch increases under pressure and decreases on dilation, like the train whistle analogy, is it not the variation of harmonics the thing that changes from consonance to disonance? but unlike the train who's whistle variation in apparent pitch is dependent upon direction of travel towards or away from us it's now occurring on every 1/2 cycle(as in positive cycles produce even order effects and negative cycles produce odd order)
Doppler is frequency modulation and air nonlinearity is more like amplitude modulation. Both act similar, but not identical.
For musicians, vibrato is fm tremelo us am and phase shiters are time modulation which in effect causes a moving comb filter similar to wah wah. A Leslie cabinet for ecample manages to do all three via moving speaker baffles. The guys at Hammond organ and Bach way bach... knew a whole lot about physics (-:
well it's time to recap.
distortion (which is a never ending debate) is audible in horns.
experimentation with driver types is one potential area for experimentation.
horn profiles all have sonic anomalies that go with them (pick your poison)
is there any thing else that would have a potential effect on decreasing distortion in horns?
distortion (which is a never ending debate) is audible in horns.
experimentation with driver types is one potential area for experimentation.
horn profiles all have sonic anomalies that go with them (pick your poison)
is there any thing else that would have a potential effect on decreasing distortion in horns?
Again, these statements don't say much since they are so broad.
Distortion isn't always audible in horns - that it "can be" is not the point. The point is that it can also be negligible. Isn't "negligible" the goal? Then the fact that it "can be audible" is kind of pointless.
By "horns" here you must mean a horn/driver combination. That's not the way I think of the problem. The "horn" is one aspect and the "driver" is another. The two things have completely different audible aspects to them and lumping them together gets problematic.
The "horn profile ... sonic anomalies" are not all equal. There are huge differences in the distortions from the different profiles. So its not just a simple question of "picking one".
Distortion isn't always audible in horns - that it "can be" is not the point. The point is that it can also be negligible. Isn't "negligible" the goal? Then the fact that it "can be audible" is kind of pointless.
By "horns" here you must mean a horn/driver combination. That's not the way I think of the problem. The "horn" is one aspect and the "driver" is another. The two things have completely different audible aspects to them and lumping them together gets problematic.
The "horn profile ... sonic anomalies" are not all equal. There are huge differences in the distortions from the different profiles. So its not just a simple question of "picking one".
well i have not yet heard a horn/driver combination that does not exhibit sonic anomalies (i would say distortion but stating it that way is "pointless") that i would consider negligible or inaudible which is why i am currently engaged in learning as much as i can about horn design.
thus far it seems that there's no way to cheat physics and all known driver/horn combinations have their caveats so yes it does come down to what one is willing to accept/live with.
thus far it seems that there's no way to cheat physics and all known driver/horn combinations have their caveats so yes it does come down to what one is willing to accept/live with.
Meh, just haven't heard a good implementation. Although I'm sure that's not an argument that will convince you . 😛
I haven't heard a direct radiator yet that was as clear as horn loaded compression drivers. See my thread on dialog clarity. Surely there are clues there to "inaudible" distortion that maybe is not so inaudible. My take is that in decently designed horns they are doing far far far more good than harm by limiting driver excursion exponentially.
well i'm not a hater and a few horns i've heard where close to approaching "uncolored" performance.
the "negligible" artifacts that most have are all clearly audible to me.
it cheez's me off when i'm told the fact i can hear this "pointless"
the "negligible" artifacts that most have are all clearly audible to me.
it cheez's me off when i'm told the fact i can hear this "pointless"
Last edited:
That you hear distortion in horns isn't pointless, but it is wrong to believe that all horns sound that way just because the ones that you have heard are distorted. If you listen carefully I have described exactly what it takes to make these devices so that the distortion is inaudible at home listening levels. Pro applications are a different discussion.
Turk - go look at the thread from Nate Hanson. The waveguides that he made have no audible distortions at home listening levels. Nate paid close attention and has created a superb design. It can be done.
What is the highest source of distorsion : the horn shape expansion or the amplification factor of a horn ?
Does as wave-guide produce distorsion by its own yet (does it produce amplification yet when very open as an OSD for instance) ? (where a horn will produce some !)
Does as wave-guide produce distorsion by its own yet (does it produce amplification yet when very open as an OSD for instance) ? (where a horn will produce some !)
- Status
- Not open for further replies.