The reflex ws about the size of the Karlson's rear chamber and similar tuning (~50Hz(
Last edited:
There are distortions specific to our hearing and vinyl recording/playback, but neither the stylus tracking vinyl grooves or our tympanic membrane physically move far and fast enough to cause a Doppler shift, a frequency modulation pitch-change.
The lathe cuts the master disc according to the RIAA curve with the lowest frequency (~20Hz) reduced -40dB compared to the highest frequency (~20,000Hz) to reduce groove width and increase signal to noise when the inverse curve is applied at playback.
The IMD in speaker systems stems from the physics of sound reproduction:
Four times the displacement (excursion) is required for each halving of frequency to maintain the same sound pressure level.
As an example, a 5" speaker would require ~12 millimeters peak to peak movement to produce 76dB SPL at 30Hz, while at 3000Hz only ~0.0012mm would be required.
The 30Hz excursion is 10,000 times that of 3kHz for the same output level.
Art
Last edited:
It's straightforward enough - because sound wave propagation is longitudinal (as opposed to transverse), intermodulation distortion due to more than one tone being reproduced by the same cone exists (in the form of frequency modulation - Doppler - and phase modulation). Doppler relates to excursion amplitude and frequency, phase modulation to excursion.
How audible these two are is down to listening tests. And if there's significant driver non-linearity, there's amplitude-related harmonic distortion and intermodulation distortion, just to muddy the waters...
How audible these two are is down to listening tests. And if there's significant driver non-linearity, there's amplitude-related harmonic distortion and intermodulation distortion, just to muddy the waters...
AM and FM produce sidebands with different phase relationships relative to the original frequencies. However spectral analysis only looks at frequency magnitudes, not phases. That's part of what makes it hard to distinguish the two different effects in typical measurements.
Thanks to @Markw4 I've been thinking, and reading, and eating humble pie. The article by Rod Elliot is fantastic and I recommend to everyone on both sides of this argument. What impressed me is he actually measured the phenomenon. I also spent a lot of time on the Klippel site reading the papers there on testing and loudspeaker modes, good stuff!
Rod Elliot calls it phase modulation distortion and that's what It should be defined as not Doppler distortion. It may not be the best name but it's more descriptive of what is happening.
Bottom line: PMD exists, it's not Doppler, it's very difficult to measure, it's difficult to reproduce, it's probably impossible to hear on any system under normal listening levels and with a subwoofer, since it is a phase phenomenon it can possibly be removed with DSP as the original article referenced. So, if the test required an O-scope running at a sample rate of 2.5Ms/s at 12 bits resolution to measure and record relatively low frequencies, what kind of computing power is needed to correct PMD in a wide-band music situation?
You came a long way.
But still: Doppler and phase modulation are the same..
I can learn and I'm not afraid to admit I was wrong. All it takes is a good experiment with data to back it up. I would like to re-create the experiment to see for myself but I don't have the equipment. I may just rely on my ears to see if I can hear it. There is a wide separation of frequencies required to make it audible 36.8Hz to 4.65kHz -- seven octaves! Also, the woofer was cranking at 10mm of travel and had to be anchored to the floor! This is pretty extreme. It's around the lowest note played by a symphony orchestra and the high frequency is higher than any fundamental produced by any musical instrument except for keyboards and cymbals. I doubt this situation presents itself in much music. Even Elliot says, "The effect is very small (to the point of being virtually inaudible by itself), and is usually swamped (or masked if you prefer) by amplitude modulation and intermodulation distortion, so could be considered immaterial in any typical loudspeaker system." If this is so, then why do we need an electronic gizmo to remove Doppler? Then there is the issue of what the heck people are hearing. If the PMD is so low to be covered by other distortions, and required an extreme non musical experiment to create this distortion, then what are the warbles that people attest to hearing? So, I still think there is a lot of hokum going on.
I tend to think of it like this: There is physics, which is hard science (not hokum), there is Psychoacoustics (which provides some [arguably rough] statistical information about human auditory perception), there is Auditory Scene Analysis (which studies auditory perception in the presence of noise), there is general human Perceptual Science (including sensometrics), there is Auditory Neuroscience, and I don't know how many other scientific fields informing our understanding of audio-related phenomena and perception. The science is generally not hokum, but some of it is still in a state of active development.
At the other end of the spectrum is what might be terms audiophoolery, or the tendency to fool oneself or to be fooled by others.
Somewhere between the extremes are people who learn from working in audio professionally, such as very experienced mastering engineers, recording engineers, professional musicians, etc. In this category there are people who learn to hear small details most people don't notice. However, what musicians are good at noticing is not always exactly the same as what mastering engineers learn how to notice.
Also somewhere in between the extremes we have people in audio forums which may have some experience and or expertise in various aspects of the above.
--------------------------------------------------------------
Then in a more general area of study we have Cognitive Psychology, which studies is an abstract or modular sort of way how people think and believe.
--------------------------------------------------------------
Exactly where and how we can know with certainty this or that claim is hokum can be hard to say. We see all sorts of arguments in internet forums where people disagree with each other in a more or less polarized way, and where both sides are pretty much set in their beliefs. Seems to be a common theme in human nature, would be my observation.
At the other end of the spectrum is what might be terms audiophoolery, or the tendency to fool oneself or to be fooled by others.
Somewhere between the extremes are people who learn from working in audio professionally, such as very experienced mastering engineers, recording engineers, professional musicians, etc. In this category there are people who learn to hear small details most people don't notice. However, what musicians are good at noticing is not always exactly the same as what mastering engineers learn how to notice.
Also somewhere in between the extremes we have people in audio forums which may have some experience and or expertise in various aspects of the above.
--------------------------------------------------------------
Then in a more general area of study we have Cognitive Psychology, which studies is an abstract or modular sort of way how people think and believe.
--------------------------------------------------------------
Exactly where and how we can know with certainty this or that claim is hokum can be hard to say. We see all sorts of arguments in internet forums where people disagree with each other in a more or less polarized way, and where both sides are pretty much set in their beliefs. Seems to be a common theme in human nature, would be my observation.
Last edited:
I would say the above is a very important lesson about audio: Some things are difficult to measure, including some things that can be audible in certain circumstances. What can be just as difficult, or even more difficult, is to properly conduct high quality double blind listening tests. IOW, there are practical reasons some things remain controversial in audio. Controversy and or lack of measurements and or lack of formal listening tests do not necessarily indicate hokum or snake oil is involved. PMD would still exist even if Rod Elliott hadn't measured it -- maybe something useful to keep in mind going forward.
I could make a matlab simulation to create a wav file that emulates what happens?