Just to be clear:
The nonlinear part of material deformation is not the cause of CMP behavior either - in fact - CMP behaviour with respect to break up modes is fully independent of that.
The cause of break up basically is a relativistic one: no material whatsoever can be stiff - and I mean *absolutely stiff* in a relativistic sense - in the time domain movement (change thereof) and thus deformation can only happen within limited velocity.
So - it simply is not possibly to have "piston movement" at whatever low movement (frequency) we look at - *if* we take it "hairsplitting".
Its just not bothering us that much as long as the joint to the VC *roughly* is in phase with the rest of the membrane towards the rim / surround.
CMP is caused by the *delay time* it takes to built up standing waves which is related to the speed of sound of the diaphragm material and the distances to travel for different break up modes.
Michael
Last edited:
You're lack of materials science knowledge is starting to show. There are all kinds of deformation - elastic is the obvious type in this case. There is no requirement that deformation be permanent when diaphragms operate in breakup - that is your own delusional thinking that I'd prefer you not project onto me.
And yes, the response of the driver is "highly and obviously dependent upon the level of input signal". Distortion levels - particularly non linear distortion is almost always heavily influenced by drive level. I don't know what drivers or data you've been looking at. But none of the driver's I've been working with over the past 25 years support your contention that non linear behavior is independent of drive level. The Thiel-Small small signal model itself was developed because of the difficulties that arose in comparing the performance of different drivers at drive levels significantly above 1V. I'd like to continue having a legitimate, productive conversation but you're going to have to get real with some facts first. And as mentioned earlier, distortion plots are an absolute necessity to prove the usefulness of UE. Without them, this is all just a gum flapping exercise. So let's dispense with the handwaving and bring out the measurements.🙂
fntn, I believe you are the individual who said "just like yield behavior of a piece of steel", not I.
And was it not also you who said, "The nature of material deformation (what diaphragm breakup actually requires and represents) is highly non linear." I reitterate, nonlineary is nor a prerequsit for breakup.
The intent of the UE is not to reduce nonlinear distortion any more so than any other passive or active crossover is, with the exception of allowing the driver to operate over an optimum band width, thus reducing distortion from out of band radiation.
Breakup, in and of itself is not necessarily a mechansim for generation nonlinear distortion.
Perhaps you would like to explain to those interested just how breakup requires nonlinear deformation?
And was it not also you who said, "The nature of material deformation (what diaphragm breakup actually requires and represents) is highly non linear." I reitterate, nonlineary is nor a prerequsit for breakup.
The intent of the UE is not to reduce nonlinear distortion any more so than any other passive or active crossover is, with the exception of allowing the driver to operate over an optimum band width, thus reducing distortion from out of band radiation.
Breakup, in and of itself is not necessarily a mechansim for generation nonlinear distortion.
Perhaps you would like to explain to those interested just how breakup requires nonlinear deformation?
Thats only half ways true and certainly not at the critical time slots I pointed at.
We are used to think of non-linear distortion in terms of a periodic event related to the signal periodicy - but thats simply no longer the case with CMP systems.
You might remember that I pointed it out under the term "CMP framing" - but it did not create too much attention either - LOL
Anyway I don't want to interfere too much with the discussion you have with fntn - which is only punctual related to CMP as I hopefully made clear for the record.
😉
Michael
Last edited:
The reference to yield was the change in the modulus of elasticity that occurs at high rates of deflection just prior to plastic deformation or yield for materials having a non linear modulus of elasticity. The stress strain curve of steel is linear until yield. Obviously, transducer diaphragms aren't made of steel.
You're clearly in combat mode here. I'm not interested in talking more non sense with you. The initial observation still stands - in my experience, distortion rises dramatically when applying "EQ" in the form of even a modest 4-5db high Q boost or cut to an actively driven loudspeaker. If you can show distortion plots in addition to the multitude of CSD's and FR plots - your theory about UE's usefulness might garner some merit. Until that occurs, if it ever occurs, I will remain skeptical of any and all claims - especially those relating to a reduction in distortion as a result of amplitude and phase "adjustments" to the applied signal.
Last edited:
I believe that the only person who discussed reduction of distortion due to the UE was ... me. It was you who mis-characterized what was said.
1) Flattening the FR will reduce distortion in that FR irregularities are distortion - linear distortion. You made the erroneous assumption that a reference was made to non-linear distortion. That was incorrect. I also did not say that any linear distortion was totally eliminate, only the flattening the FR will reduce the linear distortion.
2) No one, not me, not others, made any claims that when the UE EQ's phase to flat that it was any sort of change in distortion, though in a strict sense it is and easily demonstrated in the impulse response. Whether or not it is audible is another topic. We did discuss correction of tweeter "ringing" or breakup, but only from the aspect of it being correctable or not.
3) I did point out that there was one limited possibility of reduction of non-linear distortion and that was through the use of high order slopes in order to prevent exciting the driver where it produces some of its distortion, essentially what any crossover does, to whatever degree it may.
4) Statements as to linear response of breakup being level dependent was just that, whether or not the breakup is itself linear, that is, if one doubles the signal does the breakup magnitude double. In my experience it does, in a linear fashion. Any increase in distortion will likely be primarily motor and suspension issues unrelated to any breakup within a diaphragm. The latter was the initial focus of discussion.
You're certainly free to remain skeptical of anything you wish. If you care to discuss it in a civil manner, I'm sure everyone here would welcome that. But as far as "combat mode", it's a case of the kettle calling the pot black.
Last edited:
No, it's simply one aspect, the linear response aspect, and it's correct. If you want to expand the discussion, no problem. And there is no "towards" anything, something is right or it is wrong. In this case, it is right.
Dave
Quite interestingly, I had the an aluminum dome tweeter with a ring in front. It did not help too much. With a another metal tweeter with a ring in front of it, some listening testes were conducted, and the ring seemed to mask some detail, once cut off, the detail was there, but the breakup was much audible. However the breakup mode can be dispersed. Beryllium and Diamond somehow produce a different sonic signature, I am not quite sure where it is comming from, but my guess is it's related with the manufacturing and forming process.
Inverted domes probably seem to give a wider off axis response, but at the expense of obtaining a uniform wave front because the edge becomes radiators and act like diffraction with a different time relation. Not particularly good if one is tring to reduce diffraction
Last edited:
Think its wrong then "in this case" 😀
I'm not particularly aware of that I brought "non-linear distortion" into that discussion but anyway - as always - I have my difficulties with the separation between linear and non-linear distortion (linear-distortion not being recognized as distortion at all where I come from).
*If* we take a closer look at so-called linear distortion (which is kind a oximoron in my book) the "real" distortion happens at the point where the change in amplitude happens - simply as any change in amplitude of a sinusoid needs some additional "frequencies" to happen.
So if we look at 2nd 3rd or any other non-linear distortion these changes in amplitude happen on a basis periodically to the signal examined - but with CMP systems this change in amplitude happens only twice - at the time slot of delay and at signal shut down
This is a fundamental difference one has to grasp first !
Michael
Last edited:
No disagreement about the latency associated with the energy stored and released in a diaphragm undergoing deformation and breakup. The relationship between stiffness, mass, and distance of energy traveled is pretty easy to understand. However, for the purposes of designing and operating a successful loudspeaker - I think we can dispense with the relativistic concerns - bandwidth is limited by human hearing and power that exceeds 120db in a typical listening environment is neither necessary nor desirable. 😀
Loud sound does not sound as loud if the duration is short enough. However, the perceived dynamics makes a difference.
Soon, when you scroll down the page you can see an energy storage investigation that Linkwitz did on a metal dome tweeter with difusor. I think it was a Scan speak model.
Issues in speaker design - 2
Issues in speaker design - 2
Yes, I have read that before. Unfortunately, more data of mine is left on a different system. Basically, I always try to avoid having anything in front of a driver for the past 20 or 30 years as an audiophile.
Yep. I came across that a little while back. Interesting investigation as always with Linkwitz. Clearly not related to any diaphragm deformation or breakup but this part of his comment agrees with my experience and what I've tried to convey earlier in this thread:
"Equalization of the modified driver might be a possibility. Whether it is worthwhile depends on the driver's non-linear performance, which requires different tests to determine."
That's a pretty blatant acknowledgement of the limitations of EQ - something most of us with a little experience will readily acknowledge.🙄
No SL's "stored energy" term isn't interesting at all - its mere B*S*T - and anybody is jumping at it because it applies to each and everything that makes up for the slightest wiggle in FR and because it was coined by an audio icon - sorry Siegfried ...
Would he have had investigated any deeper instead of being superficial (in this respect) - he immediately would have come across the fact that there are fundamental differences in so called "stored energy" - and - to lump it together in a phony term will not make understanding of whats actually going on any easier.
Michael
No SL's "stored energy" term isn't interesting at all - its mere B*S*T - and anybody is jumping at it because it applies to each and everything that makes up for the slightest wiggle in FR and because it was coined by an audio icon - sorry Siegfried ...
Would he have had investigated any deeper instead of being superficial (in this respect) - he immediately would have come across the fact that there are fundamental differences in so called "stored energy" - and - to lump it together in a phony term will not make understanding of whats actually going on any easier.
Michael
You have a point. In this case, the "stored energy" is systemic and not related or confined to the moving mass itself. In that sense, Linkwitz was a little loose with the term "stored energy". It would have been more precise if he referred to the shield as an external resonator or "energy storage mechanism".
- Status
- Not open for further replies.