What is the basis to say that a traditional horn expansion would do well with HOMs? I would expect the opposite. With the exception of the conical all other traditional expansions have continuously varying side wall curvatures. IOW the second derivative of the curve's profile is non zero which means that a wave traveling along the length of the horn will diffract.
In a conical any diffraction will be due to the abrupt transition at the throat only (and unavoidably at the mouth) and that's what the OS profile - or expansion if you want to call it, is there to do: transition with the minimum of diffraction and thus minimum HOM generation. It seems that the OS waveguide is the best so far in both pattern control and HOMs.
If you care to study that has already been presented throughout this thread, probably there will be better understanding. If you have any data related with your studies to show, it's much appreciated. I'm here mainly because lots of discussions inspired me to investigate this type of design, and after reading through the thread, I feel that such design concept can be further improved, and I share some of my findings which may not be agreeable with everyone, so I'm not out to convince people. Anyone that wants more discussion will really have to show their own data first.
HOMs are generated by diffraction or reflection waves, I'm sure Earl had mentioned this earlier in the thread. The throat controls diffraction and the side walls control expansion/reflection. I am not going to spectulate what the foams does, but if there is data showing the curves without the foam vs with the foam, the answers should be clear. If you can persuade Earl to show such data or anyone willing to conduct measurements and present them from the kits they work on. Which would be the best way to end any arguments BTW. Anone doing such tests, please use MLS signals.musical noise said:
I have no data to share...only conjecture and questions...
The foam's stated purpose is absorption, what is the result of that? Are there no artifacts from the absorption process?...or do the results sum to 0 across the spectrum? I think the data soongsc requests would shed some light on that.
The foam's stated purpose is absorption, what is the result of that? Are there no artifacts from the absorption process?...or do the results sum to 0 across the spectrum? I think the data soongsc requests would shed some light on that.
Soongsc is guessing, as usual, with no data to support his guesses. Musical Noise has it correct.
I have shown what Soongsc has asked before, probably more than once. And in fact this data still exists on my web site. A highly detailed discussion of the foam has been available for years at
http://www.gedlee.com/downloads/Sound Quality Improvements in Compression Driver Systems.pdf
The problem here is that Soongsc either does not remember this data or never read it in the first place. I try and be patient with him, but it gets very difficult at times. He has not shown one shread of evidence to support anything that he claims.
What exactly is it that will satisfy "I think the data soongsc requests would shed some light on that." that has not already been shown or evaluated? I don't think that Soongsc knows what he wants to see and usually when data is shown that he doesn't like he just responds with "well I hear it" or "I don't accept that" or "there has to be a better way", all without a single justification for his position.
Soongsc just seems to ignore everything that I say (so I ignore him) and rambles on and on about my not posting data etc. etc. when this has all been done before.
I have shown what Soongsc has asked before, probably more than once. And in fact this data still exists on my web site. A highly detailed discussion of the foam has been available for years at
http://www.gedlee.com/downloads/Sound Quality Improvements in Compression Driver Systems.pdf
The problem here is that Soongsc either does not remember this data or never read it in the first place. I try and be patient with him, but it gets very difficult at times. He has not shown one shread of evidence to support anything that he claims.
What exactly is it that will satisfy "I think the data soongsc requests would shed some light on that." that has not already been shown or evaluated? I don't think that Soongsc knows what he wants to see and usually when data is shown that he doesn't like he just responds with "well I hear it" or "I don't accept that" or "there has to be a better way", all without a single justification for his position.
Soongsc just seems to ignore everything that I say (so I ignore him) and rambles on and on about my not posting data etc. etc. when this has all been done before.
This is my understanding of what the foam does.
As we can see, if we go just by the OS concept, that actual wave front cannot be spherical at all frequencies. Once we rely totally one the throat diffraction too much, the higher frequency HOMs will occur. How the driver and wave guide throat match, and how they match with the guides will effect how significan the HOMs are.
This post contains a 90deg OS wave guide sim.
http://www.diyaudio.com/forums/showthread.php?postid=1769282#post1769282
There is a later sim that involves eh LeCleach expansion calculation method, thus posted in that thread. But when Jean-Michel pointed out a similarity between one of my sims here and a LeCleach calculated expansion, it pointed out what I had in mind for better control of directivity.
An externally hosted image should be here but it was not working when we last tested it.
As we can see, if we go just by the OS concept, that actual wave front cannot be spherical at all frequencies. Once we rely totally one the throat diffraction too much, the higher frequency HOMs will occur. How the driver and wave guide throat match, and how they match with the guides will effect how significan the HOMs are.
This post contains a 90deg OS wave guide sim.
http://www.diyaudio.com/forums/showthread.php?postid=1769282#post1769282
There is a later sim that involves eh LeCleach expansion calculation method, thus posted in that thread. But when Jean-Michel pointed out a similarity between one of my sims here and a LeCleach calculated expansion, it pointed out what I had in mind for better control of directivity.
Here is where I started to take the challenge. Thanks Earl.
gedlee said:Keep trying! If my experince is any indication, in another five or ten years you'll be making great sounding waveguides>
soongsc said:Sounds encouraging, I'm good at cutting schedules.
Russell Dawkins said:
"Pride goeth before a fall", they say.
Hope I'm doing okay so far.soongsc said:
"Never let pride stand in the way of thy goal", they say.
As usual what you are saying makes little sense. "Lines closer together" means that the directivity is wider not "more uniform" unless by "uniform" you mean less directional. An omni-directional source would thus be completely "uniform", but so what? It's uniformity with frequency that matters not angle - I try and get the line as far apart as possible because this is called "directional". CD not not mean "uniform in angle, it means uniform in frequency".
soongsc said:
Hope I'm doing okay so far.
I would say that you are grabing at straws. You haven't grasped the fundamentals but are seeking to refine the details. You should read ALL that is currently KNOWN before striking out to develop the new. This is fundamental to all research and you seem to want to try a short cut by not learning the basics and just doing the hypothesizing. This is virtually always a failure.
Lines closer together means directivity is wider if they don't intersect with each other. Once they start intersecting and crossing each other back and forth, then it means the wave front is not uniform and varies depending on frequency.gedlee said:
This BEM software is commonly available. Since you have already posted your waveguide geometry data points, if you can provide geometry data from the diaphragm to the throat, I'm sure many would be glad to run the sim.
I wonder how the light bulb was invented.gedlee said:
Alright, I now hear you on the "no loss of resolution" thing. But my other question remains:
"Also, does the foam attenuate the low end of the WG+CD much? And if so, does it greatly affect the added sensitivity (and thus the ability to use a lower XO point)?"
...and is quite pertinant for any speaker designer wishing to ever consider purchase a license to use or make your foam plugs.
"Also, does the foam attenuate the low end of the WG+CD much? And if so, does it greatly affect the added sensitivity (and thus the ability to use a lower XO point)?"
...and is quite pertinant for any speaker designer wishing to ever consider purchase a license to use or make your foam plugs.
gedlee said:
and if I remember my highschool honors thesis correctly, Edison was decently educated (given science of that time period) and well read on the topic. Known for spending much of his time reading the works of others, and learning from others accomplishments before developing his own.
Hello musical noise
What you said (quited) contradicts what Earl used to say:
the HOMs are more absorbed through the foam plug than the direct ways as the HOMs will travel a longuer path...
Also when replying to the question about the modelisation of the foam Earl indicated that the BEM should should not be modelised as an acoustically resistive material but as a reactive material (complex impedance). So there is some good reason to think that a simple reduction of the SPL is not the only modification due to the foam.
I would like to study pulse responses taken, with and without the foam plug, at different angles from the axis and different distances... but that's a huge work of mesurement.
Best regards from Paris, France
Jean-Michel Le Cléac'h
What you said (quited) contradicts what Earl used to say:
the HOMs are more absorbed through the foam plug than the direct ways as the HOMs will travel a longuer path...
Also when replying to the question about the modelisation of the foam Earl indicated that the BEM should should not be modelised as an acoustically resistive material but as a reactive material (complex impedance). So there is some good reason to think that a simple reduction of the SPL is not the only modification due to the foam.
I would like to study pulse responses taken, with and without the foam plug, at different angles from the axis and different distances... but that's a huge work of mesurement.
Best regards from Paris, France
Jean-Michel Le Cléac'h
musical noise said:
Jmmlc said:
Jean-Michel incorrect again.
What I said was that the wave velocity had to be complex (which means real and imaginary) NOT that the impedance had to be complex (which can also be real AND imaginary). When something is complex it need not be purely reactive, it can have a resistive part and in fact the resistive part can be the major part. And what if the foam does have a reactive part? So what? Its obvious from the data that that its not doing anything resonant.
Why would you care about off axis impulse responses? You never do off axis measurements (or at least never show any) why all of a sudden the interest in off-axis?
And musical noise said nothing that contradicts what I have said. Where do you get this stuff?