auplater,
Are you sure you want to know this? It is not a neat and tidy study of a statistically derived, and thereby recognized need, for an obvious improvement. There will not be any clear lines of investigation, leading in a systematic fashion, from carefully researched and developed points of of support, for the overall argument.
i will do it if you wish, but you will be appalled.
Bud
Are you sure you want to know this? It is not a neat and tidy study of a statistically derived, and thereby recognized need, for an obvious improvement. There will not be any clear lines of investigation, leading in a systematic fashion, from carefully researched and developed points of of support, for the overall argument.
i will do it if you wish, but you will be appalled.
Bud
Unless I'm missing something in your suggested measurement scheme, it's not going to show anything with regard to actual FR or distortion.
It shows distortion of the energy levels of the cone. If you know how to read a trace. Distortion is an easy thing to spot.
I have yet to see any singular test show all parameters. I use at work all the avaliable tests i believe are necessary and then accumulate data and stand back and do trend analysis.
The test method i put forth has and is being used by NASA, API, ASME,ASTM and NATO, so please call them up and tell them that it does not work based on your subjective opinion.
ron
It shows distortion of the energy levels of the cone. If you know how to read a trace. Distortion is an easy thing to spot.
I have yet to see any singular test show all parameters. I use at work all the avaliable tests i believe are necessary and then accumulate data and stand back and do trend analysis.
The test method i put forth has and is being used by NASA, API, ASME,ASTM and NATO, so please call them up and tell them that it does not work based on your subjective opinion.
ron
Graham Maynard said:
I'm sure we all recognize that, but there's a lot of certainty expressed by those without it.
But it is not reliably repeatable and has no precision. It won't show what is desired to be discerned.
First, saying "if it and its near neighbouring sines are reproduced with a different group delay" is a complete misunderstanding of what group delay is. The phase at any frequency has no "group delay". Second, we most certainly CAN tell hos accurately music waveforms will be reproduced during music. One more time, since you try to claim otherwise, the impulse response tells more about what any driver will do under a complex waveform than can ANY music, bar none. Repeating inaccuracies is, unfortunately, becoming a regularity. I'm sorry to be blunt, but there seems to be no other way to make it more clear. Your claim is patently false, motor non-linear distortion excepted.
And you know this how?
ronc said:
How is the distortion represented? Is it shown in the acoustic domain as distortion in the acoustic output or is it a distortion of the mechanical vibrational modes? Can it predict the distortion imposed in the acoustic domain? Can it be translated into the acoustic domain in some way that is predictive of the acoustic impact, since that is the domain of interest in a driver? This is not being cynical, I'm curious to know the details.
Dave
Re: Re: Re: Re: "Do you hear what I hear?"
At the same time the focus included details of the mechanism that came into question as well as data said to support the hypothesis that was contradictory. The described mechanism was used to support the claimed efficacy, so the focus narrowed to that. It does seem that most would rather not question, but would rather accept everything at face value, whatever it is.
Dave
planet10 said:
At the same time the focus included details of the mechanism that came into question as well as data said to support the hypothesis that was contradictory. The described mechanism was used to support the claimed efficacy, so the focus narrowed to that. It does seem that most would rather not question, but would rather accept everything at face value, whatever it is.
Dave
Well, I'm not sure we're playing and singing from the the same page of music, or even the same piece!
Subjective reporting can be too highly valued, but it also can be undervalued.
If qualitative terms can be reliably correlated to "objective phenomena" over a reasonable number of responses, then most of the time, one can rely on peoples' use of the terms.
For instance, musical instrument makers will tell you, and I think correctly, that when folk talk about an instrument having a dark tone or timbre then it's harmonics tend to be a bit louder in the low frequencies, and conversely if it has a bright sound then the harmonic structure is a bit louder in high frequencies. Furthermore, they can even correlate such things with patterns of psychoacoustic
excitation.
The URL for the quote above is I think too long to go in DIYaudio's url device but if you go to the page for the url below and look for "A PSYCHOACOUSTIC HEARING TEST" at the bottom, you can find the rest of the article. This site is fabulously interesting because it has made a successful effort to connect objective correlates to subjective auditory qualitative terms.
http://www.schleske.de/index.php?la...eigenbauer/en_akustik3schall4musikdarst.shtml
The page you do pull up actually has a demonstration of the psychoacoustic differences between different individual instruments and thus the reasons for the agreed upon differences in their sound.
http://www.schleske.de/06geigenbauer/en_akustik3schall4musikdarst.shtml
We ideally don't want a loudspeaker to have the characteristics of a musical instrument - it should not have a characteristic sound - and in its production of the signal put into it, it should not hide from our hearing some part of that signal. Our experience tells that a good number of speakers do have these faults.
I don't think it's very constructive to dismiss either measurement of things like distortion, or to dismiss subjective judgment and it's terms. If for instance, someone tells us, "It's as if the sound is unveiled," then if we're being rigorous, we note he's reporting that he can hear something now he wasn't hearing before. Even if we're technophiles the meaning of words is important. To veil is to hide.
There is no point in mocking him for being poetical - we should pursue the matter: What is description of the before and after sound? What was done to unveil? What was the reasoning (theory) behind the activity? What physical things were done? Can we measure their physical effects? Can we connect those effects to pychoacoustic (and therefor physiological) activity? If we can do that, then we have to decide if he's misinterpreting what he heard: he might be, if he's running the tweeter really hot and he's getting "detailed sound", creating a more interesting veil.
It's not useful to dismiss measurements because they don't always seem to correlate to what we hear - we have to find a correlation or our speaker making remains in the realm of handicraft, or if the furniture's nice enough, craftsmanship, but we won't really know what we're doing. The example of the German guys at the url above is proof - they know what they're doing to a far greater extent than most speaker makers, and they have at least as many variables to deal with.
I think it would be useful to go back to a description of the before and after sound: the denominator was 'fricative hash'; more before treatment; less of it afterward. The phrase was "less fricative hash."
Fricatives have unharmonic content. Unharmonic content in this case is white noise caused by released compressed air.
By their very nature speaker diaphragms have no mechanism for storing air to decompress.
So what can a speaker diaphragm do to produce such a sound when it's not in the original signal?
That's the question to ask first. Because that's actual phenomena what the guy described. I don't think it makes any difference if we're seventh grade science students or Nobel prize winners, that's where we have to start.
The next question I think would be, how can such a sound veil other sounds?
That's before we get to examining what he did, his reasoning, etc.
Subjective reporting can be too highly valued, but it also can be undervalued.
If qualitative terms can be reliably correlated to "objective phenomena" over a reasonable number of responses, then most of the time, one can rely on peoples' use of the terms.
For instance, musical instrument makers will tell you, and I think correctly, that when folk talk about an instrument having a dark tone or timbre then it's harmonics tend to be a bit louder in the low frequencies, and conversely if it has a bright sound then the harmonic structure is a bit louder in high frequencies. Furthermore, they can even correlate such things with patterns of psychoacoustic
excitation.
The URL for the quote above is I think too long to go in DIYaudio's url device but if you go to the page for the url below and look for "A PSYCHOACOUSTIC HEARING TEST" at the bottom, you can find the rest of the article. This site is fabulously interesting because it has made a successful effort to connect objective correlates to subjective auditory qualitative terms.
http://www.schleske.de/index.php?la...eigenbauer/en_akustik3schall4musikdarst.shtml
The page you do pull up actually has a demonstration of the psychoacoustic differences between different individual instruments and thus the reasons for the agreed upon differences in their sound.
http://www.schleske.de/06geigenbauer/en_akustik3schall4musikdarst.shtml
We ideally don't want a loudspeaker to have the characteristics of a musical instrument - it should not have a characteristic sound - and in its production of the signal put into it, it should not hide from our hearing some part of that signal. Our experience tells that a good number of speakers do have these faults.
I don't think it's very constructive to dismiss either measurement of things like distortion, or to dismiss subjective judgment and it's terms. If for instance, someone tells us, "It's as if the sound is unveiled," then if we're being rigorous, we note he's reporting that he can hear something now he wasn't hearing before. Even if we're technophiles the meaning of words is important. To veil is to hide.
There is no point in mocking him for being poetical - we should pursue the matter: What is description of the before and after sound? What was done to unveil? What was the reasoning (theory) behind the activity? What physical things were done? Can we measure their physical effects? Can we connect those effects to pychoacoustic (and therefor physiological) activity? If we can do that, then we have to decide if he's misinterpreting what he heard: he might be, if he's running the tweeter really hot and he's getting "detailed sound", creating a more interesting veil.
It's not useful to dismiss measurements because they don't always seem to correlate to what we hear - we have to find a correlation or our speaker making remains in the realm of handicraft, or if the furniture's nice enough, craftsmanship, but we won't really know what we're doing. The example of the German guys at the url above is proof - they know what they're doing to a far greater extent than most speaker makers, and they have at least as many variables to deal with.
I think it would be useful to go back to a description of the before and after sound: the denominator was 'fricative hash'; more before treatment; less of it afterward. The phrase was "less fricative hash."
Fricatives have unharmonic content. Unharmonic content in this case is white noise caused by released compressed air.
By their very nature speaker diaphragms have no mechanism for storing air to decompress.
So what can a speaker diaphragm do to produce such a sound when it's not in the original signal?
That's the question to ask first. Because that's actual phenomena what the guy described. I don't think it makes any difference if we're seventh grade science students or Nobel prize winners, that's where we have to start.
The next question I think would be, how can such a sound veil other sounds?
That's before we get to examining what he did, his reasoning, etc.
dlr,
Here are some more distortion plots. This and a number of follow on PDF's (due to size constraints for posting documents) show a Vifa 5 inch mid/bass driver from the same time period. A waterfall is included, but I think LAUD from that era used something other than a pulse for signal. I will have to dig to find the instruction manual, but this was from the DOS based test suite.
Again, every bit as uninformative as to what was causing the distinctly audible change, as the previous, abbreviated tests I posted. I suspect you can begin to see where my distress was coming from, when we first wrangled about the usefulness of piston model based tests.
I do believe there are more squirreled away, but so far the computer they are on is unresponsive. It's only been turned off for 6 years now, so, like, what's the problem, huh? I don't think any of those tests are any more informative. This is not to say that more modern test suite's cannot find a difference. Just that, as I said earlier, my experience has been an unsuccessful one.
Bud
Here are some more distortion plots. This and a number of follow on PDF's (due to size constraints for posting documents) show a Vifa 5 inch mid/bass driver from the same time period. A waterfall is included, but I think LAUD from that era used something other than a pulse for signal. I will have to dig to find the instruction manual, but this was from the DOS based test suite.
Again, every bit as uninformative as to what was causing the distinctly audible change, as the previous, abbreviated tests I posted. I suspect you can begin to see where my distress was coming from, when we first wrangled about the usefulness of piston model based tests.
I do believe there are more squirreled away, but so far the computer they are on is unresponsive. It's only been turned off for 6 years now, so, like, what's the problem, huh? I don't think any of those tests are any more informative. This is not to say that more modern test suite's cannot find a difference. Just that, as I said earlier, my experience has been an unsuccessful one.
Bud
And the second set of curves is the treated driver, as it was in the first set of tests. This is true of every page of the PDF.
Bud
Bud
Nice work on the overlays David. On 1st look raises some questions - but then one of the key overall questions is are we looking in the right place with the right tools? Nah- let's not go around the block on that one - again.
Hi Dave (dlr).
You keep widening your argument beyond what I (and I believe others too with their posts) intend or think is reasonable, and in particular you keep failing to cover the aspect I have been trying to highlight - that of dynamic transverse air motions over the cone which develop in music time, with respect to reactive driver/air impedances which also change in time, after initial e-m transduction, but which can be directly related to the modification/veiling of on-going reproduction.
Therefore I shall not be responding to you any more within this thread.
Cheers ........ Graham.
You keep widening your argument beyond what I (and I believe others too with their posts) intend or think is reasonable, and in particular you keep failing to cover the aspect I have been trying to highlight - that of dynamic transverse air motions over the cone which develop in music time, with respect to reactive driver/air impedances which also change in time, after initial e-m transduction, but which can be directly related to the modification/veiling of on-going reproduction.
Therefore I shall not be responding to you any more within this thread.
Cheers ........ Graham.
I've done some searching and it appears that no-one has tried the process on some mid-low priced headphones. This is something that could give you a lot of bang for your buck.
I'm talking headphones with big foam pads around your ears here, not earphones.
Any suggestions for a model to try? something under US$100 with paper cones and reasonably flat freq response.
It would be a very cheap way for people to try this process out.
I'm talking headphones with big foam pads around your ears here, not earphones.
Any suggestions for a model to try? something under US$100 with paper cones and reasonably flat freq response.
It would be a very cheap way for people to try this process out.
BudP said:
Really? Uninformative? You are correct in that the data doesn't support an esoteric theory of damping standing waves, or altering the launch of an acoustic wave from the driver, or that standing waves in the cone are damped. However, they do clearly show the resonances associated with the excitation of cone modes and the resulting standing waves . And clearly show differences, audible differences, which are consistent with added mass changing the resonant modes of the cone.
I'm looking at the overlaid measurements posted by Planet10. These plots of amplitude and phase, and the CSD sure slow that the difference in FR response is very obvious and pronounced at higher frequencies where the drive is in bending wave mode or radiation. Everything here is consistent with the cause-effect relationship being one of added mass. I would also say, based on this data, it would be difficult to consider the red response as an improvement. The response is clearly degradated in the 1K Hz - 1.5k Hz region with what would be a audible notch at about 1.3k Hz. The behavior of the driver above 3k Hz indicates that this is not a full range driver but still what is observed is a shift in the resonance frequencies associated with cone breakup along with changes in Q of the resonances. The phase changes are what would be expected assuming minimum phase. We see little change in phase in regions where there is no significant change in response with the phase collapsing back similar behavior where the response is more of less unaltered. The CSD plots are what would be expected as well. You will notice that the null at 1.3k Hz in the treated driver (red) shows extended decay time of a resonance in that I would expect would be in the pass band for a system using this driver. Other than that, at higher frequencies there are expected differences, but nothing indicating significant improvement or further degradation of the response. All that can be concluded above 2k is that it's different. If this was a full range driver with raised response above 3K Hz (more or less flat) the high frequency differences would be audible as a change in timbre. Subjectively that could be labeled anything. But given a decent crossover I doubt anything above 2 k would result in audible differences, good or bad, for this driver. Same with distortion. There is really nothing very conclusive here over most of the band other that that they are somewhat different. Regions of higher distortion remain higher distortion, and vise versa. The major exception is again at 1.3k Hz where there is the obvious and significant increase in distortion form a level of about 0.2% to 5%.
Everything seen here is consistent with the primary audible differences being a direct result of differences in frequency response, and secondary differences being the result of the degradation of the distortion in the 1.3k Hz region. All consistent with added mass resulting is changing the modal behavior of the cone.
This particular set of data should give one pause before applying the treatment. The evidence here clearly shows degradation of the driver response with respect to overall linearity, particularly in what would be considered the useful band pass. In other cases it could show an improvement. Of course, this is an objective observation assuming that deviation form linearity is bad. Subjectively these differences could be judged any way the listener likes. This is a clear example that if you are going to tweak you drivers, make sure the tweak can be undone.
Yes, it can be argue that the evidence isn't conclusive as to the source of the change, but it is consistent with added mass, thus supportive of that prepossession, and suggest that before other more esoteric explanations are offered this prepossession should be investigated. As you know, all things considered, the simplest explanation tends to be the correct one. Or as Judge Judy says, "If it doesn't make sense it's probably not true."
The problem I see is that you seem to claim that the audible change due to something that changes the way that the acoustic wave launched from the cones. This would require that you eliminate change in cove vibration as a source. That means you either have to A) demonstrate that the treated cone vibrates exactly as an untreated cone, or B) that the differences in response of the treated cone can not be accounted for by observed differences in the vibration of the cone. Since a treated cone does have added mass distributed in an inhomogeneous manner and which may also alter the damping and compliance of the cone, there can be no question that the cone will vibrate differently. So you won’t be able to show A. But then you are caught is a catch 22. Since if the cone vibrates differently any acoustic wave launched by the code will necessarily be different that that of an untreated cone. So all you can potentially show is that the differences observed in B can not all be accounted for by the differences in vibration. Assuming that is the case this still leaves you in the position that what is heard or measured is the result of two inseparable causes.
- Status
- Not open for further replies.