Below – again – a closed box behaviour.
Again, with a large form factor for the box volume – actually again a pipe as before, this time equal in cross section though (not tapered).
This may show us some relevance to the beloved small and tall speakers that are en vogue today.
and animation:
http://www.kinotechnik.edis.at/pages/diyaudio/CMPB/ASAR/anim_1closed_box_2m_pipe_8000.gif
In general, no surprise here
– the ASAR pattern is just the same as for the tapered pipe before. Most obvious to see is the one octave frequency shift compared to the tapered pipe (both are of same length of roughly 2m).
For comparison this is the CSD plot (easy to see the decay in steps) :
and IR plot
the looped echoes are most easily to identify – which is not the case in most other real world measurements (where a looot of ASAR patterns plus a looot of “simple” filter patterns mud up to almost un-idetifyable traces).
It should not be any difficult to accept that a transmission line speaker is of more or less the same sonic pattern – no?
Also – at a larger scale - it should not be any difficult to accept that room modes are of more or less the same sonic pattern – no?
Also – at a smaller scale - it should not be any difficult to accept that diaphragm modes (break up) are of more or less the same sonic pattern – no?
In the (strict) technically / physically meaning, these are simple echoes between more or less massive boundaries - *not* a “resonance” .
As said - ASAR pattern / CMP behaviour is a different animal.
🙂
Michael
Again, with a large form factor for the box volume – actually again a pipe as before, this time equal in cross section though (not tapered).
This may show us some relevance to the beloved small and tall speakers that are en vogue today.
and animation:
http://www.kinotechnik.edis.at/pages/diyaudio/CMPB/ASAR/anim_1closed_box_2m_pipe_8000.gif
In general, no surprise here
– the ASAR pattern is just the same as for the tapered pipe before. Most obvious to see is the one octave frequency shift compared to the tapered pipe (both are of same length of roughly 2m).
For comparison this is the CSD plot (easy to see the decay in steps) :
and IR plot
the looped echoes are most easily to identify – which is not the case in most other real world measurements (where a looot of ASAR patterns plus a looot of “simple” filter patterns mud up to almost un-idetifyable traces).
It should not be any difficult to accept that a transmission line speaker is of more or less the same sonic pattern – no?
Also – at a larger scale - it should not be any difficult to accept that room modes are of more or less the same sonic pattern – no?
Also – at a smaller scale - it should not be any difficult to accept that diaphragm modes (break up) are of more or less the same sonic pattern – no?
In the (strict) technically / physically meaning, these are simple echoes between more or less massive boundaries - *not* a “resonance” .
As said - ASAR pattern / CMP behaviour is a different animal.
🙂
Michael
Last edited:
Below – again – a closed box behaviour.
Exactly the same large form factor volume (non tapered pipe) and measurement setup as before.
As the tapering did not help any, I thought it might be a nice try to show how an internal Helmholtz resonator (IHR) is affecting ASAR pattern.
Basically its a volume coupled to the pipe in close proximity to the chassis that is port tuned such to “absorb” specific frequencies.
and animation:
http://www.kinotechnik.edis.at/pages/diyaudio/CMPB/ASAR/anim_1closed_box_2m_pipe_IHR3_8000.gif
As we can see in comparison with the plots in the last post I have placed here a notch attenuation centered around the 270Hz ridge (the third one) -
BUT
- the ASAR *pattern* itself got not affected at all.
The notch was provided by a heavily dampened volume of roughly 2 liters, coupled by a port of 6cm diameter and 1cm length
Basically its the same picture we get as was shown at the beginning of the thread where combining of patterns has been covered. This “simple” notch filter applied by the IHR is just a little bit lower in Q, thus affecting the neighboring frequency bands slightly more.
But - the conclusion remains – adding a “simple” filter whatsoever *can not* correct for ASAR pattern.
Also – at a larger scale - it should not be any difficult to accept that room modes are more or less affected in the same - insufficient - way by such Helmholtz traps – no?
Also – at a smaller scale - it should not be any difficult to accept that diaphragm modes (break up) are more or less affected in the same - insufficient - way by such “simple” EQing measures – no?
🙂
Michael
Exactly the same large form factor volume (non tapered pipe) and measurement setup as before.
As the tapering did not help any, I thought it might be a nice try to show how an internal Helmholtz resonator (IHR) is affecting ASAR pattern.
Basically its a volume coupled to the pipe in close proximity to the chassis that is port tuned such to “absorb” specific frequencies.
and animation:
http://www.kinotechnik.edis.at/pages/diyaudio/CMPB/ASAR/anim_1closed_box_2m_pipe_IHR3_8000.gif
As we can see in comparison with the plots in the last post I have placed here a notch attenuation centered around the 270Hz ridge (the third one) -
BUT
- the ASAR *pattern* itself got not affected at all.
The notch was provided by a heavily dampened volume of roughly 2 liters, coupled by a port of 6cm diameter and 1cm length
Basically its the same picture we get as was shown at the beginning of the thread where combining of patterns has been covered. This “simple” notch filter applied by the IHR is just a little bit lower in Q, thus affecting the neighboring frequency bands slightly more.
But - the conclusion remains – adding a “simple” filter whatsoever *can not* correct for ASAR pattern.
Also – at a larger scale - it should not be any difficult to accept that room modes are more or less affected in the same - insufficient - way by such Helmholtz traps – no?
Also – at a smaller scale - it should not be any difficult to accept that diaphragm modes (break up) are more or less affected in the same - insufficient - way by such “simple” EQing measures – no?
🙂
Michael
Last edited:
Hi,
Certainly "ASAR" sounds much cooler, but "PBW" is only at the same paar with "CMP". The conclusion could be four letter acronymes are superior to three letter ones 😀
The ridge shifting is interesting looking phenomena. It's nice to have high resolution, in the early days these would have been neglected. Here is some weird looking stuff, using impulse responses from horn honk and other diyaudio threads.
Quest: Can the amplitude in the time-frequency domain make a full circle? 😀
bigger plot:
http://dl.dropbox.com/u/2400456/diyaudio/cQsw_da--ir_01.PNG
bigger plot:
http://dl.dropbox.com/u/2400456/diyaudio/cQsw_da--ir_02.PNG
bigger plot:
http://dl.dropbox.com/u/2400456/diyaudio/cQsw_da--ir_04.PNG
bigger plot:
http://dl.dropbox.com/u/2400456/diyaudio/cQsw_da--ir_05.PNG
bigger plot:
http://dl.dropbox.com/u/2400456/diyaudio/cQsw_da--ir_06.PNG
- Elias
Certainly "ASAR" sounds much cooler, but "PBW" is only at the same paar with "CMP". The conclusion could be four letter acronymes are superior to three letter ones 😀
The ridge shifting is interesting looking phenomena. It's nice to have high resolution, in the early days these would have been neglected. Here is some weird looking stuff, using impulse responses from horn honk and other diyaudio threads.
Quest: Can the amplitude in the time-frequency domain make a full circle? 😀
An externally hosted image should be here but it was not working when we last tested it.
bigger plot:
http://dl.dropbox.com/u/2400456/diyaudio/cQsw_da--ir_01.PNG
An externally hosted image should be here but it was not working when we last tested it.
bigger plot:
http://dl.dropbox.com/u/2400456/diyaudio/cQsw_da--ir_02.PNG
An externally hosted image should be here but it was not working when we last tested it.
bigger plot:
http://dl.dropbox.com/u/2400456/diyaudio/cQsw_da--ir_04.PNG
An externally hosted image should be here but it was not working when we last tested it.
bigger plot:
http://dl.dropbox.com/u/2400456/diyaudio/cQsw_da--ir_05.PNG
An externally hosted image should be here but it was not working when we last tested it.
bigger plot:
http://dl.dropbox.com/u/2400456/diyaudio/cQsw_da--ir_06.PNG
- Elias
LOL !
Really nice plots you show thanks - I'll come back to this after finishing (quite time consuming 🙄) ...
🙂
Michael
Now lets proceed to the interesting topic of ASAR patterns with dampening materials applied, dig slightly deeper and investigate “part spectrum CMP" behaviour - meaning – lets take a look at ASAR patterns where the delayed spectrum is not a 100% copy of the original one (as never is with dampening materials involved).
good ol' sheep's wool thats been used as dampening material:
Below – again – a closed box measurement.
Exactly the same large form factor volume (non tapered pipe) and measurement setup as before – no IHR this time - but – now stuffed with some dampening material instead.
animation thereof :
http://www.kinotechnik.edis.at/page..._1closed_box_2m_pipe_damp_very_loose_8000.gif
a single sine burst:
Lets compare that to a simulation - an ideal behaviour combined with a “part spectrum CMP” behaviour – in this case a 4 times loop that undergoes low pass filtering each time.
IR :
ASAR pattern
:
animation thereof :
http://www.kinotechnik.edis.at/pages/diyaudio/CMPB/ASAR/anim_1ir_part_CMP_6db_12ms_4multi_8000.gif
one „slice“ - a single sine burst (rectified and in log scale):
As a first observation we clearly see the effect of “ridge splitting” happening.
This is caused by phase rotation and hence different frequencies face either destructive or constructive interference.
Another consequence of those phase rotation is that the ridges don't get higher and higher (as is the case with “congruent CMP” behaviour).
Of course, also the deep nulls get smoothed out step by step, because 100% destructive interference no longer can happen.
So the smoothing effect is supported by ridge splitting / phase rotation as well as by mere SPL attenuation.
All in all those effects form the perceived and undoubtedly specific sonic pattern of “dampening materials” in the vast majority of applications – be it stuffing boxes, transmission line speakers or horns / wave guides / diffraction alignment device or whatever - whereas in a more strict sense the sonic pattern of the dampening material *itself* is the variation of dampening efficiency over its applied length related to frequency.
As a second observation that is neither caused by ASAR pattern nor by dampening materials - but nevertheless may provide a “perception boost” due to repetition and thus elongation of time window where this effect happens prominently - we see “fading amplitude modulation”.
Of course this is only seen in the measurement analysis (and not in the simulation analysis) as the simulation is lacking any high pass filter .
That's causing the visual appearance that the ridges seem to be “chipped out of a block”. Visually this is kind of the inverse of what's been presented in post #19 – though its visual appearance in PBW analysis here is rather a “gorge shifting” and – even more important to notice - now the *whole* time/ frequency matrix is affected.
What has a nice “curtain like” appearance visually, translates to quite some amplitude wiggles for a single sine burst though.
The underlying mechanism is due to the specialty of “signal start” / “signal stop” and the specific conditions of band pass limits plus single or looped echo.
This is not (merely) related to the presence of dampening materials but as it shows up so prominently, its been covered right here.
If one takes the effort to align dampening, delay time and high pass filter involved, a pretty smooth amplitude modulation may be achievable as well as pretty cruchy and weird looking and sounding AM effects that may or may not be perceived as pleasing or distorting.
To sum up:
there are two pronounced effects with respect to the specific sonic pattern of dampening materials (in combination with ASAR patterns)
a) ridge splitting
b) smoothing of ridges and gorges that happens step by step
There is more to the story that possibly could be shown regarding “dampening materials”, but this should do for now.
Again we see that those patterns are kind of “overdetermined” as they spread out into time / frequency matrix in a very specific manner.
Below – as it nicely fits the topic – the simulation of a “congruent CMP” behaviour :
IR :
ASAR pattern
animation thereof :
http://www.kinotechnik.edis.at/page...im_1ir_congruent_CMP_6db_12ms_4multi_8000.gif
one „slice“ - a single sine burst (rectified and in log scale):
As is easily seen:
By mere SPL attenuation only – meaning without the effect of phase rotation / ridge splitting – the sonic pattern of dampening materials would be quite a different story.
Actually this last “congruent CMP” behaviour simulation is of the same type as the "looped echoes" that were presented in post #6 – just with a LP limited signal this time
🙂
Michael
good ol' sheep's wool thats been used as dampening material:
Below – again – a closed box measurement.
Exactly the same large form factor volume (non tapered pipe) and measurement setup as before – no IHR this time - but – now stuffed with some dampening material instead.
animation thereof :
http://www.kinotechnik.edis.at/page..._1closed_box_2m_pipe_damp_very_loose_8000.gif
a single sine burst:
Lets compare that to a simulation - an ideal behaviour combined with a “part spectrum CMP” behaviour – in this case a 4 times loop that undergoes low pass filtering each time.
IR :
ASAR pattern
:
animation thereof :
http://www.kinotechnik.edis.at/pages/diyaudio/CMPB/ASAR/anim_1ir_part_CMP_6db_12ms_4multi_8000.gif
one „slice“ - a single sine burst (rectified and in log scale):
As a first observation we clearly see the effect of “ridge splitting” happening.
This is caused by phase rotation and hence different frequencies face either destructive or constructive interference.
Another consequence of those phase rotation is that the ridges don't get higher and higher (as is the case with “congruent CMP” behaviour).
Of course, also the deep nulls get smoothed out step by step, because 100% destructive interference no longer can happen.
So the smoothing effect is supported by ridge splitting / phase rotation as well as by mere SPL attenuation.
All in all those effects form the perceived and undoubtedly specific sonic pattern of “dampening materials” in the vast majority of applications – be it stuffing boxes, transmission line speakers or horns / wave guides / diffraction alignment device or whatever - whereas in a more strict sense the sonic pattern of the dampening material *itself* is the variation of dampening efficiency over its applied length related to frequency.
As a second observation that is neither caused by ASAR pattern nor by dampening materials - but nevertheless may provide a “perception boost” due to repetition and thus elongation of time window where this effect happens prominently - we see “fading amplitude modulation”.
Of course this is only seen in the measurement analysis (and not in the simulation analysis) as the simulation is lacking any high pass filter .
That's causing the visual appearance that the ridges seem to be “chipped out of a block”. Visually this is kind of the inverse of what's been presented in post #19 – though its visual appearance in PBW analysis here is rather a “gorge shifting” and – even more important to notice - now the *whole* time/ frequency matrix is affected.
What has a nice “curtain like” appearance visually, translates to quite some amplitude wiggles for a single sine burst though.
The underlying mechanism is due to the specialty of “signal start” / “signal stop” and the specific conditions of band pass limits plus single or looped echo.
This is not (merely) related to the presence of dampening materials but as it shows up so prominently, its been covered right here.
If one takes the effort to align dampening, delay time and high pass filter involved, a pretty smooth amplitude modulation may be achievable as well as pretty cruchy and weird looking and sounding AM effects that may or may not be perceived as pleasing or distorting.
To sum up:
there are two pronounced effects with respect to the specific sonic pattern of dampening materials (in combination with ASAR patterns)
a) ridge splitting
b) smoothing of ridges and gorges that happens step by step
There is more to the story that possibly could be shown regarding “dampening materials”, but this should do for now.
Again we see that those patterns are kind of “overdetermined” as they spread out into time / frequency matrix in a very specific manner.
Below – as it nicely fits the topic – the simulation of a “congruent CMP” behaviour :
IR :
ASAR pattern
animation thereof :
http://www.kinotechnik.edis.at/page...im_1ir_congruent_CMP_6db_12ms_4multi_8000.gif
one „slice“ - a single sine burst (rectified and in log scale):
As is easily seen:
By mere SPL attenuation only – meaning without the effect of phase rotation / ridge splitting – the sonic pattern of dampening materials would be quite a different story.
Actually this last “congruent CMP” behaviour simulation is of the same type as the "looped echoes" that were presented in post #6 – just with a LP limited signal this time
🙂
Michael
Last edited:
Given the fact that “science” in audio didn't even help to keep apples from oranges with respect to resonance versus echo (delay time effects versus “simple” filter effects) in most audio guys heads, I think it would be pretty worthless to call even louder for “bringing back science to audio”.
Instead – some patience and passion to trace down simple effects of physics to a level of *understanding* works just fine.
What has been covered as ASAR relevant topics here is quite a bow:
- the distinction between delay time effects and “simple” filter effects has been made
- the sonic patterns of “apples and oranges” have been presented by visualisation via PBW analysis and compared to each other
- its been shown how those patterns combine and a brief discourse to energy behaviour involved was made
- the relevance of CMP concept with regard to closed box, open baffle, pipes, cone brake up, room modes etc. was presented
- its been shown how dampening material affects ASAR patterns
- some fruitless measures that claim to correct ASAR patterns have been investigated
With respect to audio perception the speculation has been presented that our ear brain system may mainly process filter functions in the ear (splitting arriving sound spectrum into time/ frequency clues) and processes kind of pattern comparison (matrix look up table) via brain work.
Some further misnomer in common audio speak (“cavity resonance”) has been exposed “en passant” by investigating and discussing underlying physics of those mechanisms. As shown, this effects are actually strongly linked to CMP concept and by no means to “simple” filter theory.
Not covered here was the topic of “distortion” related to CMP behaviour and also not covered was CMP behaviour effects in electronic signal processing (and some other nearby topics).
Also not further covered here were the consequences out of how ASAR pattern can – or even more so – can not be corrected, due to the fact, that there is no “interaction” between “simple filter patterns” versus “ASAR patterns” versus “dampening patterns” in a strict sense.
This work has focused on “the basics” of time domain point of view only.
At the end of the day, if we look at all those fancy patterns introduced – its no wonder that music listening can be such thrilling.
I mean - one can only be impressed that with all that patterns to chose from, we are still able to pick “the right ones”, those that enjoy us with the sensation of image and stage, rhythm and tone.
Besides that - this thread actually wasn't merely about ASAR patterns.
Its been even more fun sharing *development* of a stringent concept ( in this case: with respect to time domain effects in audio).
I'm really grateful to all people having provided input and platform over the years.
Keep swingin'
🙂
Michael
Instead – some patience and passion to trace down simple effects of physics to a level of *understanding* works just fine.
What has been covered as ASAR relevant topics here is quite a bow:
- the distinction between delay time effects and “simple” filter effects has been made
- the sonic patterns of “apples and oranges” have been presented by visualisation via PBW analysis and compared to each other
- its been shown how those patterns combine and a brief discourse to energy behaviour involved was made
- the relevance of CMP concept with regard to closed box, open baffle, pipes, cone brake up, room modes etc. was presented
- its been shown how dampening material affects ASAR patterns
- some fruitless measures that claim to correct ASAR patterns have been investigated
With respect to audio perception the speculation has been presented that our ear brain system may mainly process filter functions in the ear (splitting arriving sound spectrum into time/ frequency clues) and processes kind of pattern comparison (matrix look up table) via brain work.
Some further misnomer in common audio speak (“cavity resonance”) has been exposed “en passant” by investigating and discussing underlying physics of those mechanisms. As shown, this effects are actually strongly linked to CMP concept and by no means to “simple” filter theory.
Not covered here was the topic of “distortion” related to CMP behaviour and also not covered was CMP behaviour effects in electronic signal processing (and some other nearby topics).
Also not further covered here were the consequences out of how ASAR pattern can – or even more so – can not be corrected, due to the fact, that there is no “interaction” between “simple filter patterns” versus “ASAR patterns” versus “dampening patterns” in a strict sense.
This work has focused on “the basics” of time domain point of view only.
At the end of the day, if we look at all those fancy patterns introduced – its no wonder that music listening can be such thrilling.
I mean - one can only be impressed that with all that patterns to chose from, we are still able to pick “the right ones”, those that enjoy us with the sensation of image and stage, rhythm and tone.
Besides that - this thread actually wasn't merely about ASAR patterns.
Its been even more fun sharing *development* of a stringent concept ( in this case: with respect to time domain effects in audio).
I'm really grateful to all people having provided input and platform over the years.
Keep swingin'
🙂
Michael
Those are pretty interesting time/ frequency patterns you show here.
Though I don't have as much time for audio as I'd like, couldn't you possibly share the IR files (first plot is the most pronounced as far as I can see)?
Any ideas where this circles and part circles stem from ? pretty high for XO ? coupling ?
Michael
Last edited:
- Status
- Not open for further replies.