Hello ScottG,
The multiresolution wavelet IS essentially the same as CSD. Jean-Michel has shown a plot of comparison.
Only that the output in CSD primarily depends on your mouse hand, whereas wavelet outputs allways the same standard plot.
- Elias
The multiresolution wavelet IS essentially the same as CSD. Jean-Michel has shown a plot of comparison.
Only that the output in CSD primarily depends on your mouse hand, whereas wavelet outputs allways the same standard plot.
- Elias
Good horn
That happens to be a very good horn. Size obviously matters.Hi,
It took a while but here's some wavelets of the JBL2431H in RCA MI9594 horn.
Where is the honk in this horn?
In this big horn throat-mouth-throat distance is about 1.5m and thus it takes about 4.3ms for a mouth reflection. But there is nothing much seen at 4.3ms??
It might be needed to measure this horn outside without room reflections to separate the honk from the room? Have the snow melted yet?
- Elias
Im sure it is but what does its wavelet show that this doesn't?
An externally hosted image should be here but it was not working when we last tested it.
Im not trying to argue which one is good, Im just trying to figure out what is really different between the two wavelets.
From my earlier post left out the other reason horn honk
I assume a horn compression driver is used with whatever horn. In almost all cases the resonant frequency of the horn compression driver is below the low frequency cutoff of the horn. If the compression driver has a Qts of 1 or greater when installed on the horn it will ring at the installed resonant frequency with any stimulating energy applied no matter the frequency. This is also honk and cannot be cured except with a lower Q compression driver.
I assume a horn compression driver is used with whatever horn. In almost all cases the resonant frequency of the horn compression driver is below the low frequency cutoff of the horn. If the compression driver has a Qts of 1 or greater when installed on the horn it will ring at the installed resonant frequency with any stimulating energy applied no matter the frequency. This is also honk and cannot be cured except with a lower Q compression driver.
I haven't read this whole 12 pages, but wanted to point to
"Spruce Moose: A Slightly Bent Horn"AES E-Library: Spruce Moose: A Slightly Bent Horn
"The Sound of Midrange Horns for Studio Monitors"AES E-Library: The Sound of Midrange Horns for Studio Monitors
and "Reflecting on Echoes and the Cepstrum"
Reflecting on Echoes and the Cepstrum
may be of interest. To quote from the 3rd piece:
"What is this power cepstrum stuff all about? In simple terms, the power cepstrum is a measure of the periodic wiggliness of a frequency response plot."
To grossly oversimplify, wiggliness in the frequency response, and I believe also spikiness in the impedance, are a (bad pun warning!) reflection of the acoustic wave not propagating smoothly from the diaphragm out to the listening space. Bends in the horn, abrupt transitions at the mouth, resonances in the phase plug and horn...all these things cause bad sound. I think of it as a transmission line/transformer that's not smoothly matching from input to output-stuff kicks back towards the source and in this case makes the sound less pleasant.
Don't know if that's what "horn honk" is, but hope the reading is interesting. Gedlee has concluded there's an optimum waveguide shape and termination, and my understanding would support that. In a more real-world example, we took an angled rectangular horn and changed it to a sliced ellipse type of shape, with a radiused edge. Voila! Just like I hoped, smoother response and nicer sound. So gratifying when vague theory-based ideas work in actuality
"Spruce Moose: A Slightly Bent Horn"AES E-Library: Spruce Moose: A Slightly Bent Horn
"The Sound of Midrange Horns for Studio Monitors"AES E-Library: The Sound of Midrange Horns for Studio Monitors
and "Reflecting on Echoes and the Cepstrum"
Reflecting on Echoes and the Cepstrum
may be of interest. To quote from the 3rd piece:
"What is this power cepstrum stuff all about? In simple terms, the power cepstrum is a measure of the periodic wiggliness of a frequency response plot."
To grossly oversimplify, wiggliness in the frequency response, and I believe also spikiness in the impedance, are a (bad pun warning!) reflection of the acoustic wave not propagating smoothly from the diaphragm out to the listening space. Bends in the horn, abrupt transitions at the mouth, resonances in the phase plug and horn...all these things cause bad sound. I think of it as a transmission line/transformer that's not smoothly matching from input to output-stuff kicks back towards the source and in this case makes the sound less pleasant.
Don't know if that's what "horn honk" is, but hope the reading is interesting. Gedlee has concluded there's an optimum waveguide shape and termination, and my understanding would support that. In a more real-world example, we took an angled rectangular horn and changed it to a sliced ellipse type of shape, with a radiused edge. Voila! Just like I hoped, smoother response and nicer sound. So gratifying when vague theory-based ideas work in actuality
Hello,
Do you know any compression driver having such a large Qts value.
IMHO most compression drivers possess a Qts <0.2
Best regards from Paris, France
Jean-Michel Le Cléac'h
Do you know any compression driver having such a large Qts value.
IMHO most compression drivers possess a Qts <0.2
Best regards from Paris, France
Jean-Michel Le Cléac'h
Hello Soongsc,
We have to be careful in our interpretation of ETC.
On such an ETC curve it is very IMHO difficult to separate what is due to the group delay rise when acoustic reactance dominates on acoustic resistance from what is due to reflections or other stored energy release processus.
IMHO, better to keep using spectrograms, CSD or wavelets graphs... onto which it is more easy to separate what is due to non stored energy processus (like GD rise due to reactance) and stored energy process (like reflections, resonances...)
Best regards from Paris, France,
Jean-Michel Le Cléac'h
We have to be careful in our interpretation of ETC.
On such an ETC curve it is very IMHO difficult to separate what is due to the group delay rise when acoustic reactance dominates on acoustic resistance from what is due to reflections or other stored energy release processus.
IMHO, better to keep using spectrograms, CSD or wavelets graphs... onto which it is more easy to separate what is due to non stored energy processus (like GD rise due to reactance) and stored energy process (like reflections, resonances...)
Best regards from Paris, France,
Jean-Michel Le Cléac'h
Hello,
I used to wrote under Matlab a spectrogram routine which gives very similar results to the continuous wavelets analysis illustrated by Elias.
It is based on the use of a gaussian envelop pulse having a constant number of periods (for what it seems theorically this pulse cannot be accepted as a wavelet though...).
You'll find my routine as an attached txt file that you can easily paste and run into Matlab.
Best regards from Paris, France
Jean-Michel Le Cléac'h
I used to wrote under Matlab a spectrogram routine which gives very similar results to the continuous wavelets analysis illustrated by Elias.
It is based on the use of a gaussian envelop pulse having a constant number of periods (for what it seems theorically this pulse cannot be accepted as a wavelet though...).
You'll find my routine as an attached txt file that you can easily paste and run into Matlab.
Best regards from Paris, France
Jean-Michel Le Cléac'h
I think GD, reflection, resonance are the same thing, a problem of transmission of energy i.e a reactance effect in acoustic domain.
Hello Thend,
The way you are considering the rise of the GD due to the unavoidable reactance of any horn and the reflections from rapid flare change or mouth toward throat which are themselves avoidable is a bit sterile considering the goal of that thread.
IMHO better to separate what is a normal behaviour of the horn (the increase of the acoustic reactance/resistance ratio near the cut-off) and what is a undesired behaviour (the reflections, ....).
Best regards from Paris, France
Jean-Michel Le Cléac'h
The way you are considering the rise of the GD due to the unavoidable reactance of any horn and the reflections from rapid flare change or mouth toward throat which are themselves avoidable is a bit sterile considering the goal of that thread.
IMHO better to separate what is a normal behaviour of the horn (the increase of the acoustic reactance/resistance ratio near the cut-off) and what is a undesired behaviour (the reflections, ....).
Best regards from Paris, France
Jean-Michel Le Cléac'h
Hello Elias,
If you look at a view side of the horn:
http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=5869&stc=1&d=1110049751
You can see that 20% of its total length (32" = 81.28cm)
is a kind of adaptor the length of which is around 17cm.
So there is at a distance from mouth roughly equals to 65cm a visible change in the profile. This means that the difference of path that leads to an interference between the direct wave from throat and the reflected wave from mouth is 1.3meter and this means that we should look for a reflection around 3.8ms. What we see on the wavelets graph are reflections between 3.8 and 4ms so we can admit that there is probably some reflections induced honkyness regarding the RCA horn.
Best regards from Paris, France
Jean-Michel Le Cléac'h
If you look at a view side of the horn:
http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=5869&stc=1&d=1110049751
You can see that 20% of its total length (32" = 81.28cm)
is a kind of adaptor the length of which is around 17cm.
So there is at a distance from mouth roughly equals to 65cm a visible change in the profile. This means that the difference of path that leads to an interference between the direct wave from throat and the reflected wave from mouth is 1.3meter and this means that we should look for a reflection around 3.8ms. What we see on the wavelets graph are reflections between 3.8 and 4ms so we can admit that there is probably some reflections induced honkyness regarding the RCA horn.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Except if we consider the cut-off horn frequency as a resonant frequency of horn, more or less absorb by driver and shape of horn itself...
What I think.
Hello Thend,
In that case we should differentiate that kind of resonance due to the shape of the horn from resonances due to its dimensions (e.g. : distance from throat to mouth...).
But I really think this is a dead end and it only adds complexity to the problem we are studying.
Best regarsd from Paris, France
Jean-Michel Le Cléac'h
In that case we should differentiate that kind of resonance due to the shape of the horn from resonances due to its dimensions (e.g. : distance from throat to mouth...).
But I really think this is a dead end and it only adds complexity to the problem we are studying.
Best regarsd from Paris, France
Jean-Michel Le Cléac'h
I really think it's exactly the oposite, it's more simple on this way because instead of consider differents phenomenon separatly, I consider basis effects for explain differents phenomenon.
It's an inductive approach (contrary to deductive), but it's mine.
Hello Thend,
Some of us introduced the idea that different type of honkiness exist in horns. If we mix all together the different sources of reflections, group delay..., should it still be possible to study separately the different type of honkiness? That's doubtful.
I must confess that, even if I find your method to characterize reflections inside a horn theorically interesting and promising, I failed until now to see any definitive proof of some practical usefulness...
What you should do IMHO, is to take some more time to develop your method and to show "real life" results on different horns in order that some comparison could be done.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Some of us introduced the idea that different type of honkiness exist in horns. If we mix all together the different sources of reflections, group delay..., should it still be possible to study separately the different type of honkiness? That's doubtful.
I must confess that, even if I find your method to characterize reflections inside a horn theorically interesting and promising, I failed until now to see any definitive proof of some practical usefulness...
What you should do IMHO, is to take some more time to develop your method and to show "real life" results on different horns in order that some comparison could be done.
Best regards from Paris, France
Jean-Michel Le Cléac'h
You are right of course but please note it when the compression driver is installed on the horn and not the unistalled Q. Yes I have seen about 2 that were that high installed.
Step change at compression driver throat interface
Yet another reasons horns honk is the conical flare rate of the compression driver interfaces at the more elliptical flare rate in the throat of the horn. Now I know this is really a small change but it is down there in the part of the horn where things are very touchy. Any step change is a change in propagation Z which can cause a reflection and make the horn honk. I use a thin matching device there which re-emits the signal from the compression driver into the throat. This has completely removed honk from horns in testing. A soft gasket with a reduced hole size from that of the horn throat. Nothing but good has come from this in practice as far as I can measure.
This may not work on all horns but has worked on the horns I use as in earlier post.
Yet another reasons horns honk is the conical flare rate of the compression driver interfaces at the more elliptical flare rate in the throat of the horn. Now I know this is really a small change but it is down there in the part of the horn where things are very touchy. Any step change is a change in propagation Z which can cause a reflection and make the horn honk. I use a thin matching device there which re-emits the signal from the compression driver into the throat. This has completely removed honk from horns in testing. A soft gasket with a reduced hole size from that of the horn throat. Nothing but good has come from this in practice as far as I can measure.
This may not work on all horns but has worked on the horns I use as in earlier post.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.