Hi,
The
PRINCIPLE OF THE DECWARE PHASE GUIDE NOV 2003 by Steve Deckert
http://www.decware.com/newsite/main...per94.htm&intro
Does not go directly to the article.
speakers>drivers>DFR-8> read about the phase guide here
You may suffer from deja-vu in a section of the article ............
/sreten.
The
PRINCIPLE OF THE DECWARE PHASE GUIDE NOV 2003 by Steve Deckert
http://www.decware.com/newsite/main...per94.htm&intro
Does not go directly to the article.
speakers>drivers>DFR-8> read about the phase guide here
You may suffer from deja-vu in a section of the article ............
/sreten.
Hi John K.
I have been communicating that your test for the baffle surface pattern was only 5 thousandths high.
To me 5 thou is not a block, and certainly not high enough to reflect a transient, as per your results.
A figure still in my mind from reading early on about EnABL and the block pattern was 1.5mil., which I read (past tense now unless I am informed otherwise) as 1.5mm ~= 1/16". I must now conclude that this meant as 1.5/1000"
I did ask Bud for clarification above in Post#241, so maybe I have made a mistake and my challenge to the relevence of your surface transient was erroneous on my part.
This does not mean I accept your 'No' answer to my last question related to coincidentally arising pressures/velocities at different frequencies. It is just that I cannot answer this one myself, and there has been no proof.
Thus I still cannot accept your transient test was sufficient to prove that an EnABLE tape pattern has no audible effect at any change of contour once a tuned pressure/velocity gradient has developed which is not directly related to the driven source.
Surely white noise is the best method for observing diffraction induced distortion, our ears being notably sensitive to fractional resonant effects and their modification - as indeed they are to a purely circular baffle. With spectral observation of white noise a critically short distance/time limitation need not then be imposed before dimensionally related peaks and troughs can develop.
Hi Bud,
My Post#241 question still stands. Is there a recommended height for EnABL 'blocks' at the baffle edge, or did I make a mistake based upon my reading of the descriptions such that here too it just a 5/1000" tape pattern ?
Cheers ........... Graham.
I have been communicating that your test for the baffle surface pattern was only 5 thousandths high.
To me 5 thou is not a block, and certainly not high enough to reflect a transient, as per your results.
A figure still in my mind from reading early on about EnABL and the block pattern was 1.5mil., which I read (past tense now unless I am informed otherwise) as 1.5mm ~= 1/16". I must now conclude that this meant as 1.5/1000"
I did ask Bud for clarification above in Post#241, so maybe I have made a mistake and my challenge to the relevence of your surface transient was erroneous on my part.
This does not mean I accept your 'No' answer to my last question related to coincidentally arising pressures/velocities at different frequencies. It is just that I cannot answer this one myself, and there has been no proof.
Thus I still cannot accept your transient test was sufficient to prove that an EnABLE tape pattern has no audible effect at any change of contour once a tuned pressure/velocity gradient has developed which is not directly related to the driven source.
Surely white noise is the best method for observing diffraction induced distortion, our ears being notably sensitive to fractional resonant effects and their modification - as indeed they are to a purely circular baffle. With spectral observation of white noise a critically short distance/time limitation need not then be imposed before dimensionally related peaks and troughs can develop.
Hi Bud,
My Post#241 question still stands. Is there a recommended height for EnABL 'blocks' at the baffle edge, or did I make a mistake based upon my reading of the descriptions such that here too it just a 5/1000" tape pattern ?
Cheers ........... Graham.
Attachments
First my number of 0.0035" being representative of an enable pattern on a driver was based on one of Bud's posts where he mentioned the height.
Second, a mil is a common using of measure equal to 1/1000". 1.5 mils is, as you note, 0.0015".
Third, it is your prerogative as to what to believe and what not to believe. I presented the measurement as it was. You can make you own, or interpret mine it as you like. It makes no difference to me. I see no need to defend objective data. It is what it is.
Fourth the term “block” seems to be used loosely here, more in a 2-dimensional sense, defining the length and width of the patch. The topic of height didn't come up until people other than Bud started using things like tooth paste and tape to apply the patterns. AFAIK, the basic intension was always to use the model paints Bud indicated to apply patches. When the patches start getting significantly thicker and thicker I question whether it would qualify as an enable patch in the original sprit of the topic. Of course, if it improves the sound (or not) I suspect surrounding you listening position with bricks lain out in an enable like pattern might be claimed to be EnABLe.
Fifth, according to Alex in Oz he used duct tape and aluminum tape for baffles and ports. The aluminum tape I have in house is just over 0.002" thick. My duct tape is about 0.003".
Second, a mil is a common using of measure equal to 1/1000". 1.5 mils is, as you note, 0.0015".
Third, it is your prerogative as to what to believe and what not to believe. I presented the measurement as it was. You can make you own, or interpret mine it as you like. It makes no difference to me. I see no need to defend objective data. It is what it is.
Fourth the term “block” seems to be used loosely here, more in a 2-dimensional sense, defining the length and width of the patch. The topic of height didn't come up until people other than Bud started using things like tooth paste and tape to apply the patterns. AFAIK, the basic intension was always to use the model paints Bud indicated to apply patches. When the patches start getting significantly thicker and thicker I question whether it would qualify as an enable patch in the original sprit of the topic. Of course, if it improves the sound (or not) I suspect surrounding you listening position with bricks lain out in an enable like pattern might be claimed to be EnABLe.
Fifth, according to Alex in Oz he used duct tape and aluminum tape for baffles and ports. The aluminum tape I have in house is just over 0.002" thick. My duct tape is about 0.003".
sreten said:Hi,
The
PRINCIPLE OF THE DECWARE PHASE GUIDE NOV 2003 by Steve Deckert
http://www.decware.com/newsite/main...per94.htm&intro
Does not go directly to the article.
speakers>drivers>DFR-8> read about the phase guide here
You may suffer from deja-vu in a section of the article ............
Not the right link but I knew how to get there. That is funny. Thanks for pointing to it. It sure was a Yoggie moment when. I also liked this quote, "The results are too close to be measured,..."
Steve has significantly "upgraded" 
the format and content of his site during the past year or so, and there are still a few dead-end links
for some interesting reading, try this most recent page regarding his customized drivers
http://www.decware.com/newsite/DFR8.htm
and to inject even more glorious Tube sound in your full range system:
http://www.decware.com/newsite/DFR6.htm
the format and content of his site during the past year or so, and there are still a few dead-end linksfor some interesting reading, try this most recent page regarding his customized drivers
http://www.decware.com/newsite/DFR8.htm
and to inject even more glorious Tube sound in your full range system:
http://www.decware.com/newsite/DFR6.htm
Graham Maynard.
5 mils, or 0.005" is rather taller than I have measured the Poly S paint, when applied to a sheet of mylar, where it cannot sink into the material. On the mylar 0.0015" to 0.002" is typical. On a paper cone the overall height above the cone is still about 0.0015" with usually an unknown amount of sink into the cone material. For large diameter bass drivers, molded out of soft fiber, with very little binding agent or fill, a sink depth of 0.007" is not an unrealistic expectation. For filled paper cone material, as found in all Lowther and Fostex drivers I have dealt with, sink is not likely to be more than 0.002".
As for the diffraction at baffle edge, I have never listened with an artificial test signal. As with all of my subjective findings, music is always the source. I use particular music, in a particular sequence, to learn about a drivers in use performance, with each of these sequential steps used to inform me about initial and subsequent treatment steps. I begin with a recording of a single piano, with considerable decay time and very few chords. This eventually progresses to the Hallelujah Chorus.
I intend to continue in this practice until some reliable cues are found that allow me to relate the changes in test data to what I rely upon currently. I would prefer to use artificial test signals and an objective test suite. My own efforts to find even repeatable data, from the same driver, treated or untreated, much less differences between them, that were reliable, is what has kept me from using objective testing. And this before any attempt tp correlate objective data with the results obtained by using the subjective tests I employ.
Thanks for continuing to move the investigation along John. I am sure we will find out what Alex is doing to obtain his results and it will be something that can be tested. It will be interesting to see what those eventual tests results look like, when compared to the distant from baffle edge tests you have already presented.
Bud
5 mils, or 0.005" is rather taller than I have measured the Poly S paint, when applied to a sheet of mylar, where it cannot sink into the material. On the mylar 0.0015" to 0.002" is typical. On a paper cone the overall height above the cone is still about 0.0015" with usually an unknown amount of sink into the cone material. For large diameter bass drivers, molded out of soft fiber, with very little binding agent or fill, a sink depth of 0.007" is not an unrealistic expectation. For filled paper cone material, as found in all Lowther and Fostex drivers I have dealt with, sink is not likely to be more than 0.002".
As for the diffraction at baffle edge, I have never listened with an artificial test signal. As with all of my subjective findings, music is always the source. I use particular music, in a particular sequence, to learn about a drivers in use performance, with each of these sequential steps used to inform me about initial and subsequent treatment steps. I begin with a recording of a single piano, with considerable decay time and very few chords. This eventually progresses to the Hallelujah Chorus.
I intend to continue in this practice until some reliable cues are found that allow me to relate the changes in test data to what I rely upon currently. I would prefer to use artificial test signals and an objective test suite. My own efforts to find even repeatable data, from the same driver, treated or untreated, much less differences between them, that were reliable, is what has kept me from using objective testing. And this before any attempt tp correlate objective data with the results obtained by using the subjective tests I employ.
Thanks for continuing to move the investigation along John. I am sure we will find out what Alex is doing to obtain his results and it will be something that can be tested. It will be interesting to see what those eventual tests results look like, when compared to the distant from baffle edge tests you have already presented.
Bud
method of acceptance
In construction, a method of acceptance for something made by craftsmen is the mock-up. An example of what the customer, architect or owner describes is made for approval. If it meets with the approval of the buyer, fine. If not, then adjustments are made and another round of approval continues until an example of what is expected is available on the site where the work will be produced.
This is in most cases a visual approval process. Sure,...color, texture & dimensional attributes come into play. Rarely, in my experience, are measuring devices brought to bear on the suitability of the work provided by the contractor.
Millions of dollars are traded on this basis every year. Those who use the method find it to be satisfactory to their needs.
Just a thought...
In construction, a method of acceptance for something made by craftsmen is the mock-up. An example of what the customer, architect or owner describes is made for approval. If it meets with the approval of the buyer, fine. If not, then adjustments are made and another round of approval continues until an example of what is expected is available on the site where the work will be produced.
This is in most cases a visual approval process. Sure,...color, texture & dimensional attributes come into play. Rarely, in my experience, are measuring devices brought to bear on the suitability of the work provided by the contractor.
Millions of dollars are traded on this basis every year. Those who use the method find it to be satisfactory to their needs.
Just a thought...
john k... said:
G'day john k,
I actually use aluminium kitchen foil and double-sided tape.
I don't know if this makes any difference whatsoever, but here is what I do for foil blocks.
Making Aluminium Blocks
Parts list:
- Heavy Duty Aluminium Kitchen Foil
- Double sided tape - Sellotape "Acid free" (12mm) - thin stuff not the padded kind
- Cutting knife or blade
- Cutting Mat
- Ruler
Instructions
1. Place a length of foil shiny side down on a hard flat surface
2. Stick strips of double sided tape onto the foil - leave about 1.5 cm on either side of each strip
3. Press down on firmly on the back of the double sided tape to get a good bond with the foil
4. Carefully tear the foil along the edge of the tape - keep tension on the tape with one hand and you will get a clean edge.
5. You now have foil strips you can cut
6. Peel off the backing and stick the foil strips onto the cutting mat
7. Cut the strip into blocks of the required size
8. Pick the blocks off the cutting mat using the blade and apply.
FWIW, I use 48mm duct tape and based on some numbers provided by auplater in another post, I believe the thickness is probably closer to 0.007".
Cheers,
Alex
BudP said:
On a baffle, there is nothing more one can do to "move the investigation along". The contradictory facts, both empirical and mathematical, keep piling up. What John has presented is conclusive, face it Bud. I expected this from day one, but I expected a bit more rationality once hard data was presented. There is not one, single piece of hard data to support anything other than added mass on a driver. All you have left is platitudes. You are in an unbelievable, absolute state of denial. What anyone hears on a baffle or a port is placebo.
Dave
Thanks for the replies confirming dimensions, my descriptive model had been wrong.
No matter - I must continue.
Diffraction is a wave effect and as thus can only be fully observed only via continuous wave excitation. It acts at an edge - not before an edge - and any newly developed waveform generated at (due to) the edge can be modified by surface shapes at and bounding the edge.
Our hearing sensitivity determines which components of any edge diffraction will be most noticeable within reproduction, so the straight line plot determination presented by John K in Post#367 will have different significance when applied to music reproduction, for it will need to have Fletcher-Munchen superimpositions in order to represent which frequency components are significant to our hearing, plots which will then of course change with reproduced level.
Baffle edge diffraction modifies radiation at frequencies in relation to baffle size and shape, starting at a frequency corresponding to the largest baffle dimension, and extending to beyond human hearing; ie. having time period effects shorter than those represented by the first distance between the driver and the test rings in John K's transient test.
Any cabinet shape checks I have made regarding edge diffraction have shown this to be most audible when listening to white noise, from a position which is in line with the baffle edge, at right angles to the plane of the loudspeaker baffle, and generally from no closer to the baffle than its width dimension. I found that being directly in front of the driver masked edge changes - though this probably be due to human hearing characteristics.
Thus I have restated my case that John K's impulse test is not valid for examining the effect of EnABL at a baffle edge and upon music reproduction because edge excitation at all frequencies has not been continuously applied over sufficient time period to excite the full diffraction range of baffle/edge tuning effects.
(Have not enough time to reason more of this at the moment.)
Cheers ........... Graham.
No matter - I must continue.
Diffraction is a wave effect and as thus can only be fully observed only via continuous wave excitation. It acts at an edge - not before an edge - and any newly developed waveform generated at (due to) the edge can be modified by surface shapes at and bounding the edge.
Our hearing sensitivity determines which components of any edge diffraction will be most noticeable within reproduction, so the straight line plot determination presented by John K in Post#367 will have different significance when applied to music reproduction, for it will need to have Fletcher-Munchen superimpositions in order to represent which frequency components are significant to our hearing, plots which will then of course change with reproduced level.
Baffle edge diffraction modifies radiation at frequencies in relation to baffle size and shape, starting at a frequency corresponding to the largest baffle dimension, and extending to beyond human hearing; ie. having time period effects shorter than those represented by the first distance between the driver and the test rings in John K's transient test.
Any cabinet shape checks I have made regarding edge diffraction have shown this to be most audible when listening to white noise, from a position which is in line with the baffle edge, at right angles to the plane of the loudspeaker baffle, and generally from no closer to the baffle than its width dimension. I found that being directly in front of the driver masked edge changes - though this probably be due to human hearing characteristics.
Thus I have restated my case that John K's impulse test is not valid for examining the effect of EnABL at a baffle edge and upon music reproduction because edge excitation at all frequencies has not been continuously applied over sufficient time period to excite the full diffraction range of baffle/edge tuning effects.
(Have not enough time to reason more of this at the moment.)
Cheers ........... Graham.
Hi sreten,
At least I present a case; not just say 'No' or 'not valid' or 'technobabble' etc.
An amplitude effect in time due to John's mortite ring was clearly illustrated, though this not amplified by continuous drive and tuned (pressure at the pattern wrt discontinuity) by related dimensions, nor appreciated via F-M weighting.
Cheers ........ Graham.
At least I present a case; not just say 'No' or 'not valid' or 'technobabble' etc.
An amplitude effect in time due to John's mortite ring was clearly illustrated, though this not amplified by continuous drive and tuned (pressure at the pattern wrt discontinuity) by related dimensions, nor appreciated via F-M weighting.
Cheers ........ Graham.
Not at all. How would you explain the diffraction artifacts present in frequency response measuremnts taken using an true impulse?
The plot was not for diffraction. It was for scattering by an obstruction. It tells us how much of the incident energy is scattered in all directions. The incident wave can always be analized for spectral content and the power spectrum of the scattered energy calculated. You (and apparently Bud) fail to recognize the ramifications of that plot and the discussion of what the maximum amount of incident energy on a patch is compared to the total energy radiated by a driver. So let's review. First the total incident energy on an enable pattern all around the baffle edge can never be greater that 20 Log(h/D) below the total radiated energy, D being an effective baffle diameter. In fact, when driver directionality is accounted for it will be even less than. Now, looking at my plot at 5k Hz the amount of incident energy scattered in all directions by the patch will be on the order of -60dB below the incident intensity or the total scattered energy will be (20 Log(h/D) - 60) dB below the total radiated energy, assuming a hemispherical wave front expanding over the baffle. If the Enable patch is 0.005" high and the effective D is 20" we have the scattered energy -132dB below the total radiated energy at 5k Hz.
As I said many posts ago in the old thread, it makes no difference it you drive over a pebble before driving over the edge of a 1000 ft cliff.
Again, in error. Please consult a good reference on acoustics. Not the typical loudspeaker book that discusses baffle step and diffraction as different issues. Diffraction is the result of an acoustic wave encountering an edge. The mathematics of what happens at the edge is not dependent on wave length or frequency. Wave length enters the picture only when the pressure field at some point in space is to be computed resulting is constructive and destructive interference between the original wave and the diffracted wave.
What you hear with white noise is what you hear with white noise. How can you separate what is edge diffraction from driver directionality? You ear only processes the acoustic wave incident on the ear drum. It can not distinguish how that wave form came to be. Certainly response aberrations will be more obvious since white noise is a full spectrum of random noise with equal energy per cycle, thus peaks and dips will stand out against a uniform background. But it's certainly NOT music. Additionally, the spectrum of an MLS signal is closely related to white noise in that it contains the full frequency spectrum, but not in a random manner.
You are very much like Bud. You make statements with no technical validity; dispute the facts because you don't like them. Feel free to disagree.
You know what guys, I must give it a rest.
I have tried, but I am so fed up with rebuffals based upon time isolated tests, and I am even more fed up with being told I am wrong about something which I have not actually stated, but which I have been assumed to think, or I have been told that I cannot appreciate the significance of tests to date, when I do.
We need someone willing to run tests with on-going music/waveform energisations or white noise or sine bursts, and then to check for EnABL induced differences.
There is still a lack of investigation relating to what happens to driver radiation after coincidental contour change induced diffraction/reflection (pressure change) events have been energised over several cycles, both before and after EnABL treatment.
The two Johns, DLR and Sreten would have me believe this has already been covered, but I cannot agree.
I have not been convinced that you have allowed waveform driven surface/spatial dimension tuned diffraction/reflection events to develop before conducting your comparison tests/measurements. If you have then your maybe your last impulse results could be presented in a manner which illustrates amplitude/frequency domains.
-132dB is inaudible; in human terms absolutely so.
So John, if the effects of EnABL are audible on a baffle or in a pipe where surface mass contribution can be ignored (audible even in the slightest degree), and your measurements are not wrong, then either you are not making the correct measurements or your interpretation of them is inadequate.
Time will tell.
Cheers ......... Graham.
I have tried, but I am so fed up with rebuffals based upon time isolated tests, and I am even more fed up with being told I am wrong about something which I have not actually stated, but which I have been assumed to think, or I have been told that I cannot appreciate the significance of tests to date, when I do.
We need someone willing to run tests with on-going music/waveform energisations or white noise or sine bursts, and then to check for EnABL induced differences.
There is still a lack of investigation relating to what happens to driver radiation after coincidental contour change induced diffraction/reflection (pressure change) events have been energised over several cycles, both before and after EnABL treatment.
The two Johns, DLR and Sreten would have me believe this has already been covered, but I cannot agree.
I have not been convinced that you have allowed waveform driven surface/spatial dimension tuned diffraction/reflection events to develop before conducting your comparison tests/measurements. If you have then your maybe your last impulse results could be presented in a manner which illustrates amplitude/frequency domains.
-132dB is inaudible; in human terms absolutely so.
So John, if the effects of EnABL are audible on a baffle or in a pipe where surface mass contribution can be ignored (audible even in the slightest degree), and your measurements are not wrong, then either you are not making the correct measurements or your interpretation of them is inadequate.
Time will tell.
Cheers ......... Graham.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.