MJL21193 said:
??
You seem very angry about something. Why would I save a screenshot of a novelty application on the internet? Doesn't make sense to me. I took the thing, annoyed my kids by playing that tune over and over again, then moreso when I ran the octaves test. Marveled at my fantastic hearing, looked in the mirror, flexed a couple times. Then decided that other things were more pressing. The end.
BudP said:
This alone is as you say, a bit inconclusive. It does show some areas of improvement along with a couple of areas worsened, more narrow, for the harmonic distortion. From some of what I've read, this kind of distortion measurement is limited in scope, but a good start. The test regiment as conducted by Mark K. and zaph using multi-tone tests and repeatable procedures would be what I'd recommend. It does require some dedication to do thoroughly. The procedure would need to be well documented so that others could consider duplicating the effort.
One critical issue up front is the test signal magnitude. This has serious consequences, especially when doing test affected by displacement such as for distortion. Mark K's site shows some of the efforts required. I don't see an easy way to do it adequately.
One of the difficulties is the limitations in the harmonics that can be tested. The test equipment sampling rate as well as the fundamental frequency are both issues as well.
Dave
staggerlee said:
Did you detect a certain tone in my post with your superior hearing? 😉
You claim to have done the test, achieved a big score, but failed to provide any proof. One would get to thinking that maybe you didn't do so well after all.
That and the fact that the number of people that have scored -39 on the test hasn't changed for quite a long time...😀
Attachments
Now that Ron has a sim that gives an idea of what could be happening how do we measure to verify or refute?
As i stated, a reflection of a light wave off of the cone to a sensor to an amp to an o scope.
I have used this at the Lazy B Ranch (Boeing) on some projects where the mass of a mounted sensor would affect the test.
Its not just amplitude ,as suggested, its phase coherence and little lumps (<scientific term) that show up in the trace. I have used this in the past to determine the actual mechanical response to a sine/square wave input by using a simple dual trace o scope with gates that freezes the pattern between the input and produced signals at the same time. The advantage to this test, in this application, is you would be measuring the energy/response at a given point of the cone surface. By measuring at various points an effective pattern of affected reigons/response could be established.
ron
(to me there is very little value in subjective, give me a number)
As i stated, a reflection of a light wave off of the cone to a sensor to an amp to an o scope.
I have used this at the Lazy B Ranch (Boeing) on some projects where the mass of a mounted sensor would affect the test.
Its not just amplitude ,as suggested, its phase coherence and little lumps (<scientific term) that show up in the trace. I have used this in the past to determine the actual mechanical response to a sine/square wave input by using a simple dual trace o scope with gates that freezes the pattern between the input and produced signals at the same time. The advantage to this test, in this application, is you would be measuring the energy/response at a given point of the cone surface. By measuring at various points an effective pattern of affected reigons/response could be established.
ron
(to me there is very little value in subjective, give me a number)
dlr,
Those tests, if I am to perform them, will have to wait for a much better business climate, for manufactured goods to flourish here in the US. That or a serious up tick in audio transformers. I cannot afford anything other than what I have for test gear, and as you will see in the next set of attached graphs, nothing conclusive, or particularly interesting shows up. The driver involved, the same one in the previous pdf, went on to work in an extremely revealing system, once the tweeter had also been treated. It was ok before treatment, but my short notes complain of very murky and dull sound, with a bad match to the tweeter. I never saw evidence of either sonic state in these tests.
The same untreated test on top orientation, is followed here.
Bud
Those tests, if I am to perform them, will have to wait for a much better business climate, for manufactured goods to flourish here in the US. That or a serious up tick in audio transformers. I cannot afford anything other than what I have for test gear, and as you will see in the next set of attached graphs, nothing conclusive, or particularly interesting shows up. The driver involved, the same one in the previous pdf, went on to work in an extremely revealing system, once the tweeter had also been treated. It was ok before treatment, but my short notes complain of very murky and dull sound, with a bad match to the tweeter. I never saw evidence of either sonic state in these tests.
The same untreated test on top orientation, is followed here.
Bud
Mark K's site shows some of the efforts required. I don't see an easy way to do it adequately.
See above.
ron
See above.
ron
Bud, let me see what i can set up in my lab. I have a virgin set of 166 drivers and i believe Dave has a set of EnABL drivers.
ron
ron
if I understand Ronc's recent posts about the nature of the radiation, very small (molecularly small?), parts of surface of the diaphragm will "kink' and "un-kink" and these "molecular" movements push air molecules, causing radiating sound we hear. Neat!
Its simply called "re-generated sound". In ultrasonics its reffered to as a Delta wave.
ron
Its simply called "re-generated sound". In ultrasonics its reffered to as a Delta wave.
ron
His is a tool that (as far as i can figure -- ron can correct me) is as capable of simulating a real-world event as any (except maybe some of the really advanced stuff at Lawrence Livermore & similar)
Yep! only difference is they can do it faster. A wave or energy at a point is the same, They take seconds, i take hours.
ron
Yep! only difference is they can do it faster. A wave or energy at a point is the same, They take seconds, i take hours.
ron
Hi Ronc - Visited your little piece of heaven when I was playing around with AWACS gear. One or our guys named Steve Brown went to work for you back about '97 or '98 - last seen working on the interior layouts for the 777 - wiring and so fourth. You might find this site of interest:
http://vanadium.rollins.edu/~tmoore/Bell_vibrations_1.htm
Oops! Just noticed that you are located in Texas - Steve is up at the plant in Washington.
http://vanadium.rollins.edu/~tmoore/Bell_vibrations_1.htm
Oops! Just noticed that you are located in Texas - Steve is up at the plant in Washington.
Visited your little piece of heaven when I was playing around with AWACS gear. One or our guys named Steve Brown went to work for you back about '97 or '98 - last seen working on the interior layouts for the 777 - wiring and so fourth. You might find this site of interest:
http://vanadium.rollins.edu/~tmoore...ibrations_1.htm
Oops! Just noticed that you are located in Texas - Steve is up at the plant in Washington.
I spent my time in WA state. Worked on AWACS, PHM, E3-A, E4B,ALCM and many black box projects, I think i remember Steve(minds failing , i am old). Only there as a consultant anymore. Boeing and I seperated due to a conflict in assingments. I was handling more problems than i could cope with, so they "retired " me to projects that had no impact. i left and entered the petro chemical and steel production industry. At least now i feel appreciated.
ron
http://vanadium.rollins.edu/~tmoore...ibrations_1.htm
Oops! Just noticed that you are located in Texas - Steve is up at the plant in Washington.
I spent my time in WA state. Worked on AWACS, PHM, E3-A, E4B,ALCM and many black box projects, I think i remember Steve(minds failing , i am old). Only there as a consultant anymore. Boeing and I seperated due to a conflict in assingments. I was handling more problems than i could cope with, so they "retired " me to projects that had no impact. i left and entered the petro chemical and steel production industry. At least now i feel appreciated.
ron
Bud those distortion graphs you posted are interesting. It appears that between about 150Hz and 1 kHz the treatment kept the f2 below 2% and f3 fell off more precipitously.
So there was a change affected through almost four octaves. This is an important range. Nearly all musical fundamental tones are within it. Orchestra tunes at A 440 Hz.
I notice a new lot of distortion was generated at around 100 -150 Hz but they are at the low end of the band where our hearing looses sensitivity.
Given that nearby in time sounds attach themselves in our perception to other nearby in time sounds, and that nearby in frequency loud sounds tend to mask nearby in frequency soft sounds, and that lower frequencies tend to mask higher ones more than vice versa, I'd guess this would lead to subjective improvement in the listening experience in this case.
I think also there is is also quite a change around 6 kHz but I don't suppose this speaker would be used in this range.
So there was a change affected through almost four octaves. This is an important range. Nearly all musical fundamental tones are within it. Orchestra tunes at A 440 Hz.
I notice a new lot of distortion was generated at around 100 -150 Hz but they are at the low end of the band where our hearing looses sensitivity.
Given that nearby in time sounds attach themselves in our perception to other nearby in time sounds, and that nearby in frequency loud sounds tend to mask nearby in frequency soft sounds, and that lower frequencies tend to mask higher ones more than vice versa, I'd guess this would lead to subjective improvement in the listening experience in this case.
I think also there is is also quite a change around 6 kHz but I don't suppose this speaker would be used in this range.
pinkmouse said:
Gentlemen, behave.
Sorry Al.
I think this will be my last post in this thread.
Every time I come here it feels like I've fallen down the rabbit hole, sitting at the table with the March Hare, the Dormouse And the Mad Hatter.
There's something off about this tea...
So that was you was it? And here were thinking it was that girl had shown up again, with her toothy cat! Your sure it was you then? Where is that hat of mine..... I wish you wouldn't just rush off then, we do need a fourth you know...
Bud
ronc said:
Unless I'm missing something in your suggested measurement scheme, it's not going to show anything with regard to actual FR or distortion.
Dave
FrankWW said:
It appears to be more in the way of tradoff where the harmonic distortion is concerned. The 100Hz area is a concern, unless it's only used as a true midrange. There's also quite a peak in second harmonic introduced at 3.5K. This is a fairly sensitive area. The area of most concern is the fairly high peaks introduced in both 2nd and 3rd harmonic at about 1100Hz. This is right where our ears are most sensitive.
Dave
Dave, you're right about there being trade offs with this speaker's treatment.
What makes me think the treated specimen might lead to a more satisfactory subjective experience, although not an optimum one, is the number of contiguous octaves affected. That's a lot of musical notes in a high energy part of spectrum where our sensitivity is also rising.
The untreated speaker was pretty well behaved between 1 kHz and 4500 kHz that with quieter bottom octaves of the treated one the retreat to the new spike around 3500 might not have been that subjectively noticeable. (I don't actually see much difference at 1100 and the situation seems improved in the higher area next, but sensitivity is high there).
It would be interesting to see if other speakers treated this way show a similar kind of pattern with the distortion products suppressed through a band with with lot of spikes bunched at the bottom.
http://en.wikipedia.org/wiki/Equal-loudness_contour
What makes me think the treated specimen might lead to a more satisfactory subjective experience, although not an optimum one, is the number of contiguous octaves affected. That's a lot of musical notes in a high energy part of spectrum where our sensitivity is also rising.
The untreated speaker was pretty well behaved between 1 kHz and 4500 kHz that with quieter bottom octaves of the treated one the retreat to the new spike around 3500 might not have been that subjectively noticeable. (I don't actually see much difference at 1100 and the situation seems improved in the higher area next, but sensitivity is high there).
It would be interesting to see if other speakers treated this way show a similar kind of pattern with the distortion products suppressed through a band with with lot of spikes bunched at the bottom.
http://en.wikipedia.org/wiki/Equal-loudness_contour
FrankWW said:
Keep in mind that the graphs showed only harmonic distortion. There was no separate information on IMD nor non-linear distortion, the motor being the primary culprit in the latter. Cheap drivers and most older drivers often had little in the way of motor distortion control in the design. I suspect that this one was such an example, so we have no information on that aspect.
Newer drivers with better motors should show less alteration in the distortion profile with any changes. But since IMD and motor non-linear distortion are not controllable by any crossover, changes in the frequency response of the driver would likely still alter these two other distortion characteristics a bit since any peaks/dips are "amplifiers" of any distortion products that coincide with them. That, of course, is why hard diaphragms can still have their characteristic "sound" to some extent even with steep crossovers. Displacement-induced motor distortion is not altered by the crossover.
Dave
FrankWW said:Dave, you're right about there being trade offs with this speaker's treatment.
What makes me think the treated specimen might lead to a more satisfactory subjective experience, although not an optimum one, is the number of contiguous octaves affected. That's a lot of musical notes in a high energy part of spectrum where our sensitivity is also rising.
The untreated speaker was pretty well behaved between 1 kHz and 4500 kHz that with quieter bottom octaves of the treated one the retreat to the new spike around 3500 might not have been that subjectively noticeable. (I don't actually see much difference at 1100 and the situation seems improved in the higher area next, but sensitivity is high there).
It would be interesting to see if other speakers treated this way show a similar kind of pattern with the distortion products suppressed through a band with with lot of spikes bunched at the bottom.
http://en.wikipedia.org/wiki/Equal-loudness_contour
I had found that same graph recently as well, useful info. I would note that much of the perception of changes (better/worse) in any driver mod is going to be auditioning level dependent aside from other issues such that affects of perception at any given time.
If you look at those curves, comparing 40 phons against 100 phons (since there's the big spike in change around 100Hz in the measurements), the difference in SPL required for perceived loudness at 40 phons between 100Hz and 1KHz is about 20. But at 100 phons, the difference drops to about 5. So any single individual's perception of changes to any part of a system would minimally have to be conducted in a comparison with precisely the same volume settings. Then there are other variables that are not easily controlled. And of course most tests are not conducted double-blind, so anecdotal evidence is rife with possibility of contamination. For those who may take issue with this, it's a fact well-known in any testing industry. We as DIYers are just as subject to the influences as anyone is, if not more so due to our personal involvements in the changes.
This is the primary reason that objective measurements are so useful. They remove any influences of those conducting the tests and if done properly can be duplicated by others.
Dave
- Status
- Not open for further replies.