Sure its the same "old" argument - as its yours.
FR is a steady state characteristic - I agree on - but in addition to that it *also* tells us in a *completely determined way* how this final static spectral distribution gets developed along time line.
This is *not* the case with CMP systems -
- that makes CMP unique and FR flawed in the presence of CMP behaviour
Michael
It has noting to do with MP or multiple sources. It is simply changing the motion of the VC so that it absorbs the energy of the reflected wave rather than continue to reflect it back into the cone. It is a TL termination problem. It is not unlike standing on a diving board or trampoline and bouncing up and down and then absorbing the recoil by flexing you knees to stop the bouncing. It is also the same concept used in the double bass array to eliminate axial modes.
John,
Whether Michael has not done any further research, just want to promote a name, or just has nothing else to do, I don't know. But I just don't see anything new on the table to discuss because he's just interested in promoting a "term". This is a very typical way sales people promote products when you give certain technology a name to copyright protect it. So I avoid even quoting the "term" now.
Basically, any reflection, diffraction, can be treated like an individual source for certain types of analysis purposes. The only thing that I have not seen done is to decompose a signal back to these individual components so that the true source and magnitude of influence is fully quantified and verified by certain improvements of a design. Now, for the physical phenomena, if we send a wave through a solid device, the vibration wave will push the air as it travels through the solid, so the sound wave is as though it's coming from a series of contiguous sources that radiates sound one by one. There are some full range drivers that operate like this. What happens is that the actual sound image focus gets slightly smeared like diffraction smears image focus. Some people like the soft sound it produces.
The ZDL will have to keep these type of diffraction phenomena to minimum in order for it to work like I think Joachim would like to accomplish.
Whether Michael has not done any further research, just want to promote a name, or just has nothing else to do, I don't know. But I just don't see anything new on the table to discuss because he's just interested in promoting a "term". This is a very typical way sales people promote products when you give certain technology a name to copyright protect it. So I avoid even quoting the "term" now.
Basically, any reflection, diffraction, can be treated like an individual source for certain types of analysis purposes. The only thing that I have not seen done is to decompose a signal back to these individual components so that the true source and magnitude of influence is fully quantified and verified by certain improvements of a design. Now, for the physical phenomena, if we send a wave through a solid device, the vibration wave will push the air as it travels through the solid, so the sound wave is as though it's coming from a series of contiguous sources that radiates sound one by one. There are some full range drivers that operate like this. What happens is that the actual sound image focus gets slightly smeared like diffraction smears image focus. Some people like the soft sound it produces.
The ZDL will have to keep these type of diffraction phenomena to minimum in order for it to work like I think Joachim would like to accomplish.
Last edited:
I still don't get it. My (limited) understanding: You have a system that is minimum phase, say the front side of a loudspeaker. Then you add in a delayed copy of the same signal. The summed signal has certain characteristics, such as a change in spectral content after a certain time. You label this change in spectral content Consecutive Minimum Phase distortion (or CMP framing). Correct?
If so, all you seem to be saying is that adding in a delayed copy of a minimum phase signal can make the system non-minimum phase. One consequence of that again is that it may no longer be possible to EQ the signal back to the original shape.
Perhaps this link discusses the same concept: Minimum Phase
If so, all you seem to be saying is that adding in a delayed copy of a minimum phase signal can make the system non-minimum phase. One consequence of that again is that it may no longer be possible to EQ the signal back to the original shape.
Perhaps this link discusses the same concept: Minimum Phase
Bending wave speakers are a different animal. Drivers like the Manger and Ohm A, F, for example. The Ohm F and A were an interesting concept where the angle of the cone was set to correct for the difference in propagation speed in air and in the cone with result that a coherent sound field was radiated, They are designed so as to (hopefully) not have reflections form the surround. In a sense the Quad 63 was a design to mimic the behavior of a bending wave source where the propagation speed was controlled by the segmented diaphragm and an electronic delay line.
Anyway, let's tak SY's advice and try to keep this thread on topic.
I got me a pair of the ox tweeters and I'll be trying to keep Joachim honest. 🙂 Boy they are cute.
Possibly – but you would miss that pronounced CMP effects are also related to (created by) cone brake up and to the kind of encloser you might use (isobaric, transmission line, TH ...)
CMP describes a „systemic pattern“ – as said...
😉
I'd say you got part of the point.
🙂
Its not a mere re-labaleing – CMP is a concept that ebraces what's not been considered to belong together until now plus it points out some specific behaviour not having been pin pointed to my knowledge yet.
Hence, nothing „new“ at all – but – this specific point of view to put things into context actually seems to be new.
Possibly look here too:
http://www.diyaudio.com/forums/multi-way/178350-zdl-21.html#post2481415
yeah, basically
There was a loooong discussion about that (starting out at Lynn's thread quite some tome ago) and it's not ever finished yet, as you can see here:
http://www.diyaudio.com/forums/multi-way/178350-zdl-22.html#post2483520
My take on that is to leave it to others what they finally would like to embrace into „MP“ or not.
Regardless of to what conclusion math heads finally may come – the limitations CMP systems are facing beyond MP behaviour are clearly brought to the point.
Those are limitations that are due to two MP systems being overlaid consecutively (delayed).
As for EQing :
In theory such CMP systems *can* be 100% corrected – as shown.
Real world EQing - beyond limitations of MP – is a different matter though. (So, we still do not know whose ranch should now belong to whom 😀)
Also - that makes CMP concept interesting to understand in the presence of nowadays cheaply available advanced EQing possibilities – be it hardware or software based
Michael
There you write:
"Thats what you can read in my paper in case you haven't got the point:
in theory *you can* correct for CMP behaviour
in practice correction is limited for CMP systems (beyond the limits of correcting pure min phase system)"
As I already wrote, you have no solution in sight for your "CMP" problem, but want to discuss it.
I am well aware of that "beating" problem. It's an "old hat". You simply have to limit the frequencies involved to below the condition, where beating starts. Anything else is simply inferior design. No need to have a new name for that.
If you ever feel, that your communication efforts do not reach the intended receivers - just read my conclusion a second time. 😉
If a sender can`t make his message understood by the receiver, it is always the fault of the sender. If it's not the fault to be not understandable, it is the fault to even send it. Period.
The interference of two delayed waves does show in the FR graph as peaks and valleys (however small). One should know whether to correct that with changes in the geometry or by EQing the SPL level. It was well known before "CMP" that many timing problems can not be equalised by changing volume levels.
Thank you for lending me your attention 🙂
Rudolf
OK, then I understand what class of systems we are discussing. It would probably include loudspeakers with multiple point sources (multi-way, dipoles, etc), any loudspeaker in a reflective room, and it gets rather complicated with a multi-way speaker in a reflective room.
For systems like that, there may be frequencies and locations where multiple reflections stack up to cancel the direct signal completely, or even overpower it to a phase-inverted version of itself. Clearly, EQ'ing away something like that will be a handful. And there will be frequency bands and locations that are more well-behaved, where you can EQ the system any way you like and still get well-behaved phase response.
So what practical implications does the CMP concept have, beyond that basic understanding of how reflections add up in a room?
For systems like that, there may be frequencies and locations where multiple reflections stack up to cancel the direct signal completely, or even overpower it to a phase-inverted version of itself. Clearly, EQ'ing away something like that will be a handful. And there will be frequency bands and locations that are more well-behaved, where you can EQ the system any way you like and still get well-behaved phase response.
So what practical implications does the CMP concept have, beyond that basic understanding of how reflections add up in a room?
Would you believe that EnABL actually uses this kind of effect to achieve it's performance?😀
What about this?
http://www.diyaudio.com/forums/solid-state/135292-cmp-final-stage.html
What about this?
http://www.diyaudio.com/forums/solid-state/135292-cmp-final-stage.html
Last edited:
NOTE: This post and many after it have been moved from another thread. If the discussion is disjointed, remember that some posts (over 100) have come from another thread
___________________________________________________________________________
Hi Pano, SY, PinkMouse ... whoever
thanks for all your effort in moving roughly the last 300 posts!
########
Have to get orientation again - may take some time...
########
I guess as you post did an incredible „overnight jump“ form around #20 to #300 – at least this point possibly we do not have to discuss any further – no ?
😀
Rudolf, you are welcome anytime.
Michael
Last edited:
Replying to an old post due to catching up on a very long thread, but...
Yes, the foil is so soft that if you were to drive it in only one location as is the case with a dome tweeter, the diaphragm would be bending and thus in breakup mode across it's entire frequency range.
But it's not being driven in one place, the entire radiating surface of the foil diaphragm is the "voice coil" and is within a uniform* magnetic field. The force acting on all visible parts of the diaphragm is equal, so all parts of the foil move back and forth in perfect unison.
Even though the foil has little bending strength it is not being subject to any bending forces across it's surface due to the uniform drive, so in practical terms it is not operating in "cone breakup" at any frequency.
* Yes in a practical ribbon the magnetic field is not perfectly uniform through the volume occupied by the ribbon, but in a good design the variation is so small that it might as well be.
I would disagree that "without substantial smoothing the frequency response looks horribly non-flat" in the case of a foil Ribbon - it may be true for the other two though.
I've measured Ribbon's such as the AC G2 narrow band swept sine with no smoothing and they're exceptionally flat, with the only deviation of the G2 being a very mild and broad emphasis around 10-20Khz on axis which is due to the design of the wave guide.
There is no way the CSD of a Ribbon would be as good as it is if it was either operating in cone breakup through it's entire range, or had a horribly non-flat un-smoothed frequency response.
I haven't heard any dome tweeter at any price that can compete with even a mid range Ribbon like the AC G2 when it comes to crisp, airy, spookily realistic and natural, listening fatigue-free sound.
I wouldn't consider a dome tweeter nowadays for anything other than a casual listening bookshelf design, certainly not for main speakers.
I don't know about Magnetostats and Electrostats, but is it really fair to say that an (aluminium foil) Ribbon tweeter is operating permanently in breakup mode ?
Yes, the foil is so soft that if you were to drive it in only one location as is the case with a dome tweeter, the diaphragm would be bending and thus in breakup mode across it's entire frequency range.
But it's not being driven in one place, the entire radiating surface of the foil diaphragm is the "voice coil" and is within a uniform* magnetic field. The force acting on all visible parts of the diaphragm is equal, so all parts of the foil move back and forth in perfect unison.
Even though the foil has little bending strength it is not being subject to any bending forces across it's surface due to the uniform drive, so in practical terms it is not operating in "cone breakup" at any frequency.
* Yes in a practical ribbon the magnetic field is not perfectly uniform through the volume occupied by the ribbon, but in a good design the variation is so small that it might as well be.
I would disagree that "without substantial smoothing the frequency response looks horribly non-flat" in the case of a foil Ribbon - it may be true for the other two though.
I've measured Ribbon's such as the AC G2 narrow band swept sine with no smoothing and they're exceptionally flat, with the only deviation of the G2 being a very mild and broad emphasis around 10-20Khz on axis which is due to the design of the wave guide.
There is no way the CSD of a Ribbon would be as good as it is if it was either operating in cone breakup through it's entire range, or had a horribly non-flat un-smoothed frequency response.
Neither do I. I didn't buy/listen to my first pair of Ribbons until 2003 but there really is no going back once you've experienced them in a properly set up design.
I haven't heard any dome tweeter at any price that can compete with even a mid range Ribbon like the AC G2 when it comes to crisp, airy, spookily realistic and natural, listening fatigue-free sound.
I wouldn't consider a dome tweeter nowadays for anything other than a casual listening bookshelf design, certainly not for main speakers.
Last edited:
I'm afraid it works according to Murphy's law - if it can happen it will happen - more or less...
Slightly different case but nevertheless:
YouTube - Resonantie
Michael
I've watched the video, but I don't see the relevance to Ribbon tweeters.
From what I can see from the poor video quality it's a square steel plate which is mounted to a vibrational device near the centre - with the edges of the plate left unsupported.
That breakup modes occur and standing wave patterns form on the plate at certain frequencies is therefore no surprise. Vibrational force is only being applied to the middle of the plate, so the inertia of the edge reacts with the stiffness of the plate to apply bending stress.
This is much like case of a standard cone speaker - driven only near the centre and "largely" unsupported at the edge. At high enough frequencies it breaks up in a similar fashion due to inertia and less than infinite stiffness.
The ribbon tweeter foil however is directly driven across it's entire surface area by a uniform force, so no bending mode stresses are applied despite the relative softness of the material.
In a real tweeter the field is probably not perfectly uniform, which would cause a very small amount of bending, but because the entire diaphragm is still magnetically controlled, I doubt whether it could develop into full blown cone breakup resonance - certainly CSD's of decent Ribbon tweeters don't show any evidence of typical dome type breakup.
I am basically guessing as you do, because I've not seen relevant data so far - but -
- it does not matter if the diaphragm structure is supported at the frame of not (this only changes the standing wave patterns with respect to frequency (and therefor I said its not exactly the same case)
- it does not matter (that much) that the force is applied more uniformly in case of planars - its IMO a very basic "stability problem" that is involved and the standing waves that are possibly at certain frequencies will build up - no matter how.
You've seen the patterns that are embossed to the Raal?
Its a pretty save bet that they are there for a reason and that reason actually can only be "non uniform" movement over the area - otherwise it would not make any sense.
Another "anecdotal" evidence may be found in the many ESL threads. There people are pin point aware of that such break up occurs and they invest quite some effort to "tune" the placement of spacers accordingly.
Michael
- it does not matter if the diaphragm structure is supported at the frame of not (this only changes the standing wave patterns with respect to frequency (and therefor I said its not exactly the same case)
- it does not matter (that much) that the force is applied more uniformly in case of planars - its IMO a very basic "stability problem" that is involved and the standing waves that are possibly at certain frequencies will build up - no matter how.
You've seen the patterns that are embossed to the Raal?
Its a pretty save bet that they are there for a reason and that reason actually can only be "non uniform" movement over the area - otherwise it would not make any sense.
Another "anecdotal" evidence may be found in the many ESL threads. There people are pin point aware of that such break up occurs and they invest quite some effort to "tune" the placement of spacers accordingly.
Michael
There is also my own experience with the NEO3 versus the AMT.
The NEO3 does show a distinct sonic pattern I was not quite happy with. My guestimation is that this is caused by
- the membrane material used
- the specific breakup patterns that occur
compared to such a planar an AMT has a fundamental advantage almost no one is aware of:
very very small diaphragm areas that shift such breakup towards extreme high frequencies.
Michael
The NEO3 does show a distinct sonic pattern I was not quite happy with. My guestimation is that this is caused by
- the membrane material used
- the specific breakup patterns that occur
compared to such a planar an AMT has a fundamental advantage almost no one is aware of:
very very small diaphragm areas that shift such breakup towards extreme high frequencies.
Michael
A planar driver with a diaphragm that is not all metal is not the same as an all aluminium foil ribbon though - in the ribbon case the entire moving mass is directly controlled by the magnetic field, whilst in a planar with a hybrid diaphragm the possibility of material resonances and cone breakup is still present, as only the conductive parts of the diaphragm are directly controlled.
Last edited:
I certainly agree that there is a difference in where the force is applied for all metal versus hybrid diaphragms - but - as said, the standing waves (= brake up) can not be "controlled" by the uniform force across the entire area.
Ask the ESL guys if you do not trust my technical explanation with respect to "basic stability issues of *any* diaphragm".
🙂
Michael
Ask the ESL guys if you do not trust my technical explanation with respect to "basic stability issues of *any* diaphragm".
🙂
Michael
- Status
- Not open for further replies.