Mouth reflection, or diffraction, I still don't know how to call it the most appropriate way (it's always a mix of both, I guess).
The termination can't be much more abrupt than that.
- Now the question is whether we could tell the difference in a listening test. We tend to pretend we can, but do we?
It should subjectively distort like crazy at least at high sound levels. As it's not really that much resonating, the effect on coloring the sound may not be that big. The imaging can be a bit blurred, who knows.
You know what, I'm actually tempted to try it some day, just to hear it (or not).
Last edited:
A question: With a high-sensitivity source, such as a compression driver in a horn, to what degree can sound waves incoming from the outside environment into the horn, reaching the diaphragm of the driver (can be the reproduced sound in a room itself), modulate its electrical impedance? Does anyone know? Is there a sensible SPL at which this could become significant?
When measuring electrical impedance of a driver in a horn, it's typically pretty sensitive to the outside noise, and I always think about this. My guess would be that with a passive crossover this could modulate its transfer function (in wild ways).
When measuring electrical impedance of a driver in a horn, it's typically pretty sensitive to the outside noise, and I always think about this. My guess would be that with a passive crossover this could modulate its transfer function (in wild ways).
Last edited:
I don't know the answer to the question, but if the reflection of sound in the horn causes problems, then external sounds of the same intensity should too. On the other hand, the intensity of reflected sound and noise from the room is usually much less than the pressure near the mouth of the waveguide.
By the way, any DIY attempts for the correction of horn reflections with FIR filters like Fulcrum does?
By the way, any DIY attempts for the correction of horn reflections with FIR filters like Fulcrum does?
I think there have been many DIY attempts (some successful..) incorporating an impulse response inversion to provide the corrections.
The level of granularity used in determining what may be (and how it can be) corrected takes a lot of experience.
Dave Gunness has been at it for over two decades, and mastered it before most DIY were even aware of what FIR was, or could do.
https://www.diyaudio.com/community/threads/preconditioning-fir-filters-how-do-you-make-them.366934/
Although I have not done double-blind testing of "corrected" (EAW called it "Gunness Focussing" before Dave left and they changed the name to EAW Focussing) I could definitely hear a positive difference from "before" Focussing (a system I worked with around 2000) and the same "after" Focussing around 2006.
And the Mackie speakers which had rather so-so horns and drivers which employed "Gunness Focussing" sounded (and measured) far better than prior.
Art
If the electrical impedance of a driver was modulated, it's frequency response would also be modulated, changed.
Multiple entry horns mid or bass drivers can produce SPL greater than 160dB in the throat region of the horn, yet when all the drivers are driven with pink noise the frequency response of the HF driver remains unchanged, the same as at lower levels.
I would not think any "sensible" outside environment SPL levels would be near those SPL levels.
When measuring electrical impedance of a driver it is not controlled (dampened) by a low impedance source like an amplifier provides.
A series capacitor can eliminate amplifier damping, outside SPL can modulate the diaphragm.
After shattering dozens of beryllium diaphragm TAD HF drivers actively crossed at ~5kHz (which should have limited excursion) Dave Rat found the "protection" capacitors in series could let them wildly shake at their resonant frequency (hammering the phase plug) while testing the PA with pink noise while the HF was muted.
Removing the series capacitors eliminated that problem, the only failures after were the result of thermal failure.
At high drive levels, the rising temperature of a compression driver's voice coil will increase it's electrical impedance, which can have wild effects on its frequency response using a passive crossover.
Art
I remember hearing about that happening. Wasn't the solution to include at least a tiny inductor across the driver? Not enough mh to form a second order filter against the series cap but just enough to brake movement of the diaphragm from sound in the room? Also didn't a few thou$and worth of drivers get damaged before realizing wth was going on?
I suppose the horn in its reversed view will be a sound sponge... it will like to absorb everything and give back nothing so the membrane in the end will be exposed to a lot of sound energy. I mean, thats how backward horn loading of e.g, nautilus speakers work.. absorb the back wave of the driver- no?
I suppose if they had shorted the CD input terminals during the noise tests, the drivers would have survived - an inductor does about the same thing - a decreasing short as frequency goes up...
The inductor should be between the cap and the driver or the amp will se a DC short - right!?
//
I suppose if they had shorted the CD input terminals during the noise tests, the drivers would have survived - an inductor does about the same thing - a decreasing short as frequency goes up...
The inductor should be between the cap and the driver or the amp will se a DC short - right!?
//
We're talking about distortion, and the frequency response, as commonly measured, is not enough to reveal these phenomena, I'm afraid.
It's certainly as interesting topic and thanks for all the ideas. I guess we seldom think of a loudspeaker in reverse, i.e. as a microphone, although it must be happening all the time. Maybe it's just negligible in the end, I don't know. It would seem to me that horns with compression drivers would be the most sensitive.
- I think if this was real, the acoustic nonlinear distortion (whatever its form) measured with steady signal would depend on the room location at which the source was placed. Does this happen?
Last edited:
The very first measurement of my A520G2 I took a while ago, just to check everything is up and running.
It's the small 1" Peerless with the T520-25-STD-1 adapter:
The task now is find another available 1" driver that would perform like this. Any candidates?
It's the small 1" Peerless with the T520-25-STD-1 adapter:
The task now is find another available 1" driver that would perform like this. Any candidates?
Last edited: