thend said:
We are not free to "redefine" diffraction, it is well defined, and your idea of it is not diffraction, its reflection. Diffraction is when a wave bends around an edge or a radius and a diffracted wave is created - there CAN BE a reflection IF this wave is sent back in the direction from whence it came, but that it is not a requirement and does not, in general, occur.
Hello,
If you don't mind and send it to me by email, I could put it on my local webspace (http://ibtk.de) and provide the link here promptly.
timo.kirschke (at) ibtk.de
B.T.W. ARTA can do a lot for you.
Cheers, Timo
editI was too slow...
If you don't mind and send it to me by email, I could put it on my local webspace (http://ibtk.de) and provide the link here promptly.
timo.kirschke (at) ibtk.de
B.T.W. ARTA can do a lot for you.
Cheers, Timo
editI was too slow...
tiki said:Hello,
ARTA can do a lot for you
Cheers, Timo
I prefer a higher resolution plot like www.gedlee.com/downloads/Nathan10_hor.pdf
This has me curious as well. I was under the impression that diffraction, say at a baffle edge, in a sense appeared as a signal source that then radiated in all directions, including back towards the origin. Intensity would be greatly diminished on that axis, but it would occur nevertheless.
Dave
Earl Geddes reply on “How Horns Work” by Bjorn Kolbrek:
http://www.kinotechnik.edis.at/pages/diyaudio/geddes/Horn_Theory_reply.pdf
enjoy!
Michael
http://www.kinotechnik.edis.at/pages/diyaudio/geddes/Horn_Theory_reply.pdf
enjoy!
Michael
For me, the reflexion and diffraction is the same thing (I know, with me it's often the same thing), insofar as it's a question of wavelength and geometry.
Seen of compression, wich is not back is transmited (resistiv acoustic impedance like an infinite plane wave tube) no reflection no interaction in horn.
Wich is back, it's reflexion(s) (reactive acoustic impedance) the sound is not completely transmitted and we have interaction of waves.
So, diffraction with phase addition or cancelation, depending of wavelength and geometry of the source of reflexion (typically a discontinuity like the tip of a tube).
Seen of compression, wich is not back is transmited (resistiv acoustic impedance like an infinite plane wave tube) no reflection no interaction in horn.
Wich is back, it's reflexion(s) (reactive acoustic impedance) the sound is not completely transmitted and we have interaction of waves.
So, diffraction with phase addition or cancelation, depending of wavelength and geometry of the source of reflexion (typically a discontinuity like the tip of a tube).
dlr said:
We are geting detailed now! You are exactly correct, a "new" source must be created to meet the boundary conditions of any change in the boundary. But is this "reflection"? To me its not. BUT, when one does the math, the same boundary conditions would apply and both reflection and diffraction will result from the calculations. But its usually possible to seperate the solutions into different terms and in acoustics it is common to define the diffraction field and the reflection field seperately, although this isn't always easy to do.
Take a sphere for example. When a sound wave hits it there is both reflection and diffraction generated, but we generally call the reflection that part of the wave that goes backwards and diffraction as that part of the wave that continues to go forward albeit along different directions.
So at one level - pure mathematics - diffraction and reflection are one and the same thing because they are both part of the same solution. But they act differently and in the literature they are discussed seperately. In the "general" literature the term "scattering" is often used, which does encompass both reflection and diffraction.
happily we have solved the issue of secondary point source / reflection / diffraction in the semantic domain by sustitution through scattering.
Now - if I may go back one or two steps
- don't you agree that *scattering* is happening along each and every contour that does not provide equal area for the wave front on its way (and even then when we account for the effect of surface friction) as well as at the mouth horn of course?
- don't you agree that HOM is crated by whatever reflections back into the throat (to the membrane) *and* its nonlinear interaction with the driver - as is summarised by Bjorn?
- why then isn't it a good idea to measure impedance and analyse it for a "lumped HOM number"?
Michael
Now - if I may go back one or two steps
- don't you agree that *scattering* is happening along each and every contour that does not provide equal area for the wave front on its way (and even then when we account for the effect of surface friction) as well as at the mouth horn of course?
- don't you agree that HOM is crated by whatever reflections back into the throat (to the membrane) *and* its nonlinear interaction with the driver - as is summarised by Bjorn?
- why then isn't it a good idea to measure impedance and analyse it for a "lumped HOM number"?
Michael
No I don't agree. Its not about area its about the wall slope. Waves can expand without diffraction as long as they are allowed to expand properly.mige0 said:
mige0 said:
No I don't agree. In an OS waveguide the mathematics do not show a diffracted wave back to the source for an infinite device, but there are HOM. In a finite OS from the mouth, sure - thats why there is foam.
mige0 said:
Because it won't show the HOM.
gedlee said:
But not in reality - point source excluded as not real...
Also "breathing" diaphragms excluded as not real - needed for (non real) infinte conical horns...
So in the end, it *is* all about contour and diffraction – no?
gedlee said:
So HOM then for you isn't non-linear distortion of the arriving acoustic wave at a certain point in space but the disturbance of the wave front in its geometry?
gedlee said:
What about measuring distortion at several SPL levels - say 90dB 100dB 110dB – at the same position in space – as you mention in your paper that HOM increases with SPL other than "usual" (non-linear) distortion - or was it only the masking you referred to?
I guess I have to move over to your thread, Earl for further questions. Did you alrady outline your thoughts about quantifying HOM somewhere ?
Michael
mige0 said:
As I say in the paper, it is all about the wall contours and how this contour affects diffraction.
mige0 said:
Yes, there is nothing nonlinear about a HOM. They are perefectly linear.
mige0 said:
You are confusing the effect with its perception. The effect is perfectly linear, but its perception is not.
mige0 said:
I think that would be appropriate, I agree. I think that HOM are discussed in some detail there, yes.
I have read all three of these. It seems that Bjorn Kolbrek is trying to provide an overview of various designs, whereas Earl Geddes talking about the math associated with it.mige0 said:Earl Geddes reply on “How Horns Work?by Bjorn Kolbrek:
http://www.kinotechnik.edis.at/pages/diyaudio/geddes/Horn_Theory_reply.pdf
enjoy!
Michael
soongsc said:
Bjorn and I agreed that what he was talking about and what I was talking about were, in general, not the same things (except in a few places where he was talking about waveguides, which is where he got it wrong and the reason for my letter). I believe that this highlights the fact that horns and waveguides are not the same thing.
To summarize, the AudioXpress papers on “How Horns Work” by Bjorn Kolbrek are at:
Part 1 (PDF)
Part 2 (PDF)
Earl Geddes' reply to the AudioXpress articles
The topic is continued in How is HOM Measured? and the most recent postings at Geddes on Waveguides.
Bjorn Kolbrek's most recent work (over the last two years) involves extensive BEM simulation of the fields inside and outside the horn, and has not been published. The BEM simulations were used to select the profile for the AH425 Azurahorn, for example, which is optimized for an Altec or Great Plains Audio 288 compression driver.
Bjorn is on a well-deserved vacation for the next several weeks, and I sincerely hope that when he's ready, he publishes the work I've seen in private communications. Most of all, I want to thank Bjorn, Michael (mige0), JMMLC and Dr. Geddes for all they've done over the last several years, and for sharing it here with the readers of DIYAudio. Although we may have different goals, there is much that can be shared with the DIYAudio community.
Part 1 (PDF)
Part 2 (PDF)
Earl Geddes' reply to the AudioXpress articles
The topic is continued in How is HOM Measured? and the most recent postings at Geddes on Waveguides.
Bjorn Kolbrek's most recent work (over the last two years) involves extensive BEM simulation of the fields inside and outside the horn, and has not been published. The BEM simulations were used to select the profile for the AH425 Azurahorn, for example, which is optimized for an Altec or Great Plains Audio 288 compression driver.
Bjorn is on a well-deserved vacation for the next several weeks, and I sincerely hope that when he's ready, he publishes the work I've seen in private communications. Most of all, I want to thank Bjorn, Michael (mige0), JMMLC and Dr. Geddes for all they've done over the last several years, and for sharing it here with the readers of DIYAudio. Although we may have different goals, there is much that can be shared with the DIYAudio community.
Hi
thought there may be some interest in "the latest" regarding horn contour.
Below my measurements of how sonngsc's horn contour turns out in my "cylinder wave" double horn.
As far as I have understood soongsc's approach was to merge the ideas of Earl and JeanMichel and mixing in some own thoughts.
I then (slightly) adapted his outcome to fit my needs for the dual horn.
Voila – enjoy !
The traces were gained by referencing to the 0 deg measurements in ARTA – overlay whatever XO slope and you get the picture.
Below the corresponding measurements (roughly same overall 2D dual horn size / same measurenment
conditions / 1m mic distance / 90 dB) with the "pure" LeCleach contour posted here earlier to compare directly.
We see that we definitely could come one step closer to John Kreskovsky's min phase idea over some radiating area.
How does it sound and bled over with the OB directivity?
Frankly - no idea ! - as the second horn is work in progress.
Anyway – my congratulations soongsc, for this outstanding work !
soongsc's contour can be found :
http://www.diyaudio.com/forums/showthread.php?postid=1851334#post1851334
soongsc's BEM simus for that contour can be found :
http://www.diyaudio.com/forums/showthread.php?postid=1851004#post1851004
Michael
thought there may be some interest in "the latest" regarding horn contour.
Below my measurements of how sonngsc's horn contour turns out in my "cylinder wave" double horn.
As far as I have understood soongsc's approach was to merge the ideas of Earl and JeanMichel and mixing in some own thoughts.
I then (slightly) adapted his outcome to fit my needs for the dual horn.
Voila – enjoy !
The traces were gained by referencing to the 0 deg measurements in ARTA – overlay whatever XO slope and you get the picture.
Below the corresponding measurements (roughly same overall 2D dual horn size / same measurenment
conditions / 1m mic distance / 90 dB) with the "pure" LeCleach contour posted here earlier to compare directly.
An externally hosted image should be here but it was not working when we last tested it.
We see that we definitely could come one step closer to John Kreskovsky's min phase idea over some radiating area.
How does it sound and bled over with the OB directivity?
Frankly - no idea ! - as the second horn is work in progress.
Anyway – my congratulations soongsc, for this outstanding work !
soongsc's contour can be found :
http://www.diyaudio.com/forums/showthread.php?postid=1851334#post1851334
soongsc's BEM simus for that contour can be found :
http://www.diyaudio.com/forums/showthread.php?postid=1851004#post1851004
Michael