Joachim,
Nice one. Of course, all ...(x) related distortion sources need to be solved by using the actual driver design.
For example problems with the main surround stiffness [Kms(x)] can be compensated for by a non-linear spider that becomes softer as excursion increases.
Issues with magnet field linearity [BL(x)] can be addresses by improved magnet geometry, the same is true for the the modulation of Le by voice coil position [Le(x)].
For those who qualify for a Voice Coil Subscription, Vance Dickason who measures several drivers each month does have a Klippel analyser and the results are published...
On the other Hand, continuing from Mr. Klippel's quite useful poster, the change of voice coil inductance with current [Le(i)] cannot be influenced by driver design, but requires the Amplifier design to deal with it.
If the driver is driven by a current source Le(i) distortion is negated, as are additional distortions created indirectly due BL(x) modulation and of course thermal compression is nullified as well.
Ciao T
Nice one. Of course, all ...(x) related distortion sources need to be solved by using the actual driver design.
For example problems with the main surround stiffness [Kms(x)] can be compensated for by a non-linear spider that becomes softer as excursion increases.
Issues with magnet field linearity [BL(x)] can be addresses by improved magnet geometry, the same is true for the the modulation of Le by voice coil position [Le(x)].
For those who qualify for a Voice Coil Subscription, Vance Dickason who measures several drivers each month does have a Klippel analyser and the results are published...
On the other Hand, continuing from Mr. Klippel's quite useful poster, the change of voice coil inductance with current [Le(i)] cannot be influenced by driver design, but requires the Amplifier design to deal with it.
If the driver is driven by a current source Le(i) distortion is negated, as are additional distortions created indirectly due BL(x) modulation and of course thermal compression is nullified as well.
Ciao T
John,
If we define "success" as commercial it is fairly easy to make a successful speaker, it needs great looks (90%) and acceptable acoustic performance (10%). Then it will sell well.
If we define successful in any acoustic sense we need to first need to define requirements that extend by far beyond what is commonly considered.
In fact many of them are already well understood by makers of Pro-Audio Studio Monitors and yet are completely ignored by many High End Speaker designers (and by many DIY Speaker designers), such as directivity (rated DI) and power compression and indeed even general distortion.
Of course, a speaker with a high and even DI, low compression and low distortion is not necessarily one that sounds good or is successful either... However I found that many times subjective problems I looked for in the FR (where they where absent) could be found by looking at the DI or Distortion levels...
Ciao T
If we define "success" as commercial it is fairly easy to make a successful speaker, it needs great looks (90%) and acceptable acoustic performance (10%). Then it will sell well.
If we define successful in any acoustic sense we need to first need to define requirements that extend by far beyond what is commonly considered.
In fact many of them are already well understood by makers of Pro-Audio Studio Monitors and yet are completely ignored by many High End Speaker designers (and by many DIY Speaker designers), such as directivity (rated DI) and power compression and indeed even general distortion.
Of course, a speaker with a high and even DI, low compression and low distortion is not necessarily one that sounds good or is successful either... However I found that many times subjective problems I looked for in the FR (where they where absent) could be found by looking at the DI or Distortion levels...
Ciao T
The idea that back EMF is not a real and measurable quantity is absurd.
Just take a speaker, block it in some way from moving its cone, and measure its impedance over frequency. You will get the combination of the inductance of the speaker and its DC resistance.
Now, unblock the speaker and measure its impedance over frequency. It will be different. Now, what makes this difference in measurement, besides back EMF that is generated by the cone motion in the magnetic field?
Just take a speaker, block it in some way from moving its cone, and measure its impedance over frequency. You will get the combination of the inductance of the speaker and its DC resistance.
Now, unblock the speaker and measure its impedance over frequency. It will be different. Now, what makes this difference in measurement, besides back EMF that is generated by the cone motion in the magnetic field?
I don't want to argue, but even JBL has sometimes failed to keep up with new developments, and at some time, forgot the fundamental trade-offs in speaker design. I know, because I measured the problems, myself. When I confronted JBL with it, (we were a JBL wholesale dealer at the time) they were NOT happy. Today, they might do better, but the point is that even PROS make mistakes. It is true that there are too many loudspeaker box developers in this world. Everybody and his brother seems to want to try his hand at it, it would seem. Most are disappointing, but the experienced ear can still differentiate and chose.
Directivity is not my greatest concern in a speaker design. I have addressed it, seriously, in pro designs, however. Low distortion, cabinet damping, and reasonable frequency response is big on my list.
Directivity is not my greatest concern in a speaker design. I have addressed it, seriously, in pro designs, however. Low distortion, cabinet damping, and reasonable frequency response is big on my list.
John, how could I understand reasonable? Does that mean that you allow for say ±3dB dips and peaks, or do you prefer equalizing it with many crossover components influencing the specific frequency range?
John,
In my old trade we generally used to quip that JBL was short for "Just Brutally Loud" (we also quipped that B*SE was short for Bring Other Sound Equipment and many other of the majors also got their deserts).
I find the kind of culture engendered in large scale corporations produces mostly mediocrity and some times startling accidents of innovation and performance. When JBL was small they turned out really innovative stuff, nowadays it's so hit & miss it's unfunny.
If I look at a 1930's Eckmiller driver, with it's time aligned coaxial arrangement and almost pulse coherent series crossover, it's mechanical damping of the drivers resonance (instead of relying on electrical damping) and combined with the open loop pentode (current source) driving Telefunken amplifier, I cannot but own that this system over seven decades is still so far ahead that current HiFi / High End / Studio Speakers have not caught up... Not sure who is to blame...
Ciao T
In my old trade we generally used to quip that JBL was short for "Just Brutally Loud" (we also quipped that B*SE was short for Bring Other Sound Equipment and many other of the majors also got their deserts).
I find the kind of culture engendered in large scale corporations produces mostly mediocrity and some times startling accidents of innovation and performance. When JBL was small they turned out really innovative stuff, nowadays it's so hit & miss it's unfunny.
If I look at a 1930's Eckmiller driver, with it's time aligned coaxial arrangement and almost pulse coherent series crossover, it's mechanical damping of the drivers resonance (instead of relying on electrical damping) and combined with the open loop pentode (current source) driving Telefunken amplifier, I cannot but own that this system over seven decades is still so far ahead that current HiFi / High End / Studio Speakers have not caught up... Not sure who is to blame...
Ciao T
Anatoliy,
I don't think we are actually in any disagreement.
As you know there are actually standards on how the loudspeaker impedance is to be rated. The rating should not be higher than twice the minimum impedance. So if someone produces a loudspeaker that is 20 ohms over the entire rated frequency range but drops to 2 ohms for even a few hertz, it must be listed as a 4 ohm loudspeaker.
Now I don't consider that an artificial standard as it tells me that I will need an amplifier rated to drive 4 ohm loudspeaker loads. It also tells me if I build an amplifier rated to drive 4 ohm loudspeakers it must be capable of handling 2 ohms worst case.
ES
The "back EMF" is just a result of compliance, mass and mechanical resistance recalculated to electrical side of the speaker schematic circuit. Usually we get something like 500uF//20mH//20 ohm parallel resonance circuit. But one has to realize that these "electrical components" originate in mechanical and acoustical impedances.
The question is who is an "amateur" here, as I graduated in electro acoustics. Speaker analysis and measurements were of the main subject. There are very few here who are educated in electro acoustics, and we can see it in thoughts posted.
The question is who is an "amateur" here, as I graduated in electro acoustics. Speaker analysis and measurements were of the main subject. There are very few here who are educated in electro acoustics, and we can see it in thoughts posted.
There was a Charles Rodrigues cartoon about that. Two homeless guys sitting in a doorway, one asks the other to tell his fortune. The other then, errr, lets out a puddle which runs down the sidewalk. He stares at it and says, "You'll be taking a trip to Florida..."
Pavel,
Reductio ad absurdum.
Actually, Back EMF MAY BE APPROXIMATED with an electrical model, however the actual process producing is literally that of the diaphragm movement inducing a voltage in the voice coil, due to the fact that the coil moves in a magnetic field, precisely the same process that yields a microphone.
In fact you can a use a Speaker in double fasion, in the 80's I worked on an Intercom design that used the speaker as microphone and by using a bridge circuit where able to go directly bi-directional, no stupid Talk/Listen Switch. They key was they could be fitted to the old systems with a small PCB.
So clearly, in this case speaking INTO THE SPEAKER produces a signal that is usable as (not so high quality) microphone.
This effect cannot just be brushed off as a lumped LCR Model.
Really. And they taught you that Back EMF from the generator of a Speaker moving in a magnetic field is actually just a mechanical equivalent of a Choke, a Resistor and a Capacitor?
And the Back EMF in a rotary motor (which I can use to actually sense it's speed and use to keep it constant is also just a choke, resistor and capacitor?
The whole system has non-linearities in so many areas that this model is not just limping, it is ruddy legless. So yes, you are right, we can see it from the posts.
Ciao T
Reductio ad absurdum.
Actually, Back EMF MAY BE APPROXIMATED with an electrical model, however the actual process producing is literally that of the diaphragm movement inducing a voltage in the voice coil, due to the fact that the coil moves in a magnetic field, precisely the same process that yields a microphone.
In fact you can a use a Speaker in double fasion, in the 80's I worked on an Intercom design that used the speaker as microphone and by using a bridge circuit where able to go directly bi-directional, no stupid Talk/Listen Switch. They key was they could be fitted to the old systems with a small PCB.
So clearly, in this case speaking INTO THE SPEAKER produces a signal that is usable as (not so high quality) microphone.
This effect cannot just be brushed off as a lumped LCR Model.
Really. And they taught you that Back EMF from the generator of a Speaker moving in a magnetic field is actually just a mechanical equivalent of a Choke, a Resistor and a Capacitor?
And the Back EMF in a rotary motor (which I can use to actually sense it's speed and use to keep it constant is also just a choke, resistor and capacitor?
The whole system has non-linearities in so many areas that this model is not just limping, it is ruddy legless. So yes, you are right, we can see it from the posts.
Ciao T
- Status
- Not open for further replies.