"My friend/colleague/professor agrees with me" does not carry any weight here. You need to accept this, and come up with sensible arguments instead of a mixture of imprecision and confusion.Joe Rasmussen said:
It is not entirely clear to me that you understand the difference between magnitude of impedance and impedance. They are not the same thing.
Joe, look up Norton. Apply it to the terminal node of the speaker.
Model the currents of the amp as well as the three snub branches and the driver. They must sum to zero.
JC cannot state publicly that I understand anything audio. He consistently lies about me in that regard.
The conversion from electrical power to mechanical ouput via magnetics is a subject so many get wrong. Magnetics is far less understood than electrostatic.
Ps..as to your compatriot who conveniently works in the same field as me, I recommend you inform him of this discussion, what I've said, and ask him if it is ok to give out his name. If he is not explaining the exact same thing we have told you, it might be better his name not get out. But it should be his decision.
Jn
Model the currents of the amp as well as the three snub branches and the driver. They must sum to zero.
JC cannot state publicly that I understand anything audio. He consistently lies about me in that regard.
The conversion from electrical power to mechanical ouput via magnetics is a subject so many get wrong. Magnetics is far less understood than electrostatic.
Ps..as to your compatriot who conveniently works in the same field as me, I recommend you inform him of this discussion, what I've said, and ask him if it is ok to give out his name. If he is not explaining the exact same thing we have told you, it might be better his name not get out. But it should be his decision.
Jn
Last edited:
Alas, it depends on the feeding we provide to it.
Go back to beginnings of May 2018 in this same thread.
Pages upon pages, the same saga and we were all there.
As I’ve said it is a very predictable pattern.
post # 3764
“Everybody in this room is wearing a uniform and don’t kid yourselves”
“you’ll hurt your throat, stop it!”
(“Burnt Weeny Sandwich” album, Side 2 right before “Valarie”)
George
Attachments
Perhaps you should understand the posts. He has just claimed that his three little elements now make the driver current flat and constant phase.JN,
Just to tweak you, yes adding the resistor does drop the loudspeaker driver current. Of course a bit due to the effective internal amplifier resistance (correctly impedance since it does vary with frequency.)
As this can in a well designed voltage amplifier with global feedback be as little as a few milliohms at low frequencies. Probably lost by the connecting wires' resistance.
So if you get the propeller cap you can spin something more material than your hypothetical wheels.
!!!
ES
Confusing the input terminal node current with the driver current is not a good thing.
All he needs to do is clamp some high bandwidth current probes across the actual wires and sweep frequency. At speaker current levels and frequencies, the clamp ones are not that expensive. It's not like he needs to use five Rogowski coils.
Jn
Ah no. Life is finite. And I admit I just wanted to poke you in your ear.
It is possible that the issue of current drive and transducers designed for it has been examined by the folks who make large quantities of self powered loudspeakers that go on sticks. But I won't claim anonymous friends beamed data to me.
At one time a single manufacturer was making so many of those toys, that if they had spun off the division it would have been third largest loudspeaker manufacturer by sales dollar consideration.
Last edited:
What I find much more interesting is what Joe's technique does to the wire and it's response.
Given a zero ohm amp with a 20 or 30 foot length of zip, a constant z load will force the settling time of the wire to become independent of frequency. IOW, ITD variation will drop close to zero, will be frequency and amplitude independent.
This removes the wire from the soundstage equation.
In addition, the zobel will prevent a high capacitance cable from making an amplifier unstable as the load goes high z below the unity gain bw point of the amp.
I've waited many months to see if Joe understood this as I didn't want to spoil his possible "novel" approach. Now it's not novel.
Jn
Given a zero ohm amp with a 20 or 30 foot length of zip, a constant z load will force the settling time of the wire to become independent of frequency. IOW, ITD variation will drop close to zero, will be frequency and amplitude independent.
This removes the wire from the soundstage equation.
In addition, the zobel will prevent a high capacitance cable from making an amplifier unstable as the load goes high z below the unity gain bw point of the amp.
I've waited many months to see if Joe understood this as I didn't want to spoil his possible "novel" approach. Now it's not novel.
Jn
I would like to try to clear this controversy with Joe.
It seems to be, as usual, a question of words and the way we use them. (Not easy for me, with my poor English ;-)
Looking at the *impedance response curve* measurement of a speaker, in free air, shows a bump at the resonance frequency. I think everybody will agree.
This is, obviously, due to the EMF voltage produced at the mechanical resonance of the speaker. I think everybody will agree, including Joe.
Now, some people, me included, think that compensating the variation of the impedance of the speakers with frequency (whatever its cause), and trying to get a flat impedance curve on all the audio range brings some benefits. (i'm too lazy to explain which and why.)
People, that never tried by themselves, think not. (they are free to "think" what they want).
To distingue this "resonance" bump from the effects from the one due to the inductance of the coil+magnetic assembly, we call this bump "motional " and, using an appropriate resonant circuit (RCL) as a parallel charge, we compensate this bump of *measured* impedance. We call this "motional impedance compensation". Not a good definition ? What the hell ?
We compensate the increase of the impedance with frequency due to the inductance with an other appropriate circuit (RC). "Inductance compensation".
which is more commonly accepted, but bring the same kind of change.
La correction d'impédance RC et RLC série
La correction d impédance RLC série
All this controversy is, on my opinion, due to a confusion between the causes and the effects.
The both changes (with and without compensation) can bee seen, with all and every speaker I have measured, on the acoustic response curve and water falls, whatever the way they are charged (Bass reflex, closed box, free air, 1/4 or 1/2 wave). Even with amplifiers with VERY low output impedance and big short cable.).
Let-us the people that are modest enough to believe they cannot think to ALL the effects at work in a speaker assembly that can have an influence on the results free to make their mind by a listening comparison.
And the others stuck in their certainties.
A speaker can work with or without this compensation networks, try and chose the best for you. But there is No snake oil in all this.
PS: To understand all the effects at work, some need is to use their knowledge of both Ohm law in alternative currents and understand Thiel and Small parameters .)
It seems to be, as usual, a question of words and the way we use them. (Not easy for me, with my poor English ;-)
Looking at the *impedance response curve* measurement of a speaker, in free air, shows a bump at the resonance frequency. I think everybody will agree.
This is, obviously, due to the EMF voltage produced at the mechanical resonance of the speaker. I think everybody will agree, including Joe.
Now, some people, me included, think that compensating the variation of the impedance of the speakers with frequency (whatever its cause), and trying to get a flat impedance curve on all the audio range brings some benefits. (i'm too lazy to explain which and why.)
People, that never tried by themselves, think not. (they are free to "think" what they want).
To distingue this "resonance" bump from the effects from the one due to the inductance of the coil+magnetic assembly, we call this bump "motional " and, using an appropriate resonant circuit (RCL) as a parallel charge, we compensate this bump of *measured* impedance. We call this "motional impedance compensation". Not a good definition ? What the hell ?
We compensate the increase of the impedance with frequency due to the inductance with an other appropriate circuit (RC). "Inductance compensation".
which is more commonly accepted, but bring the same kind of change.
La correction d'impédance RC et RLC série
La correction d impédance RLC série
All this controversy is, on my opinion, due to a confusion between the causes and the effects.
The both changes (with and without compensation) can bee seen, with all and every speaker I have measured, on the acoustic response curve and water falls, whatever the way they are charged (Bass reflex, closed box, free air, 1/4 or 1/2 wave). Even with amplifiers with VERY low output impedance and big short cable.).
Let-us the people that are modest enough to believe they cannot think to ALL the effects at work in a speaker assembly that can have an influence on the results free to make their mind by a listening comparison.
And the others stuck in their certainties.
A speaker can work with or without this compensation networks, try and chose the best for you. But there is No snake oil in all this.
PS: To understand all the effects at work, some need is to use their knowledge of both Ohm law in alternative currents and understand Thiel and Small parameters .)
Last edited:
What I studied in college was the technique of taking large and extremely complicated systems, and breaking them down into easily understood and manageable subsystems. What I have done for the last 25 years is, granted, an extreme example of that technique.
But, it was not "debating"
It was called...wait for it....
Engineering.
Jn
A decade or two ago products in the high end consumer loudspeaker market started to use networks to provide a constant impedance load. Nothing really new. The first high end models did get good reviews.
I am always amazed at all the "new" stuff that comes out regularly. Shame so many folks did the same thing 50 years earlier. Not limited to individuals. Bose has some great patent attorneys.
I am always amazed at all the "new" stuff that comes out regularly. Shame so many folks did the same thing 50 years earlier. Not limited to individuals. Bose has some great patent attorneys.
Well as it has been posted on here by the designer of the amps for at least one popular active speaker that he did that hardly 'news'
Anyway, for those that use active speakers, it is a fact that compensating speakers offers less improvements. (But still sensible)
But with passive filters, consider the serial impedance that they bring between the near 0 Ohm amplifier, the cable resistance in between it and and the speaker, and the help a constant impedance load offers, while we are at calculating the filters and their slopes.
But with passive filters, consider the serial impedance that they bring between the near 0 Ohm amplifier, the cable resistance in between it and and the speaker, and the help a constant impedance load offers, while we are at calculating the filters and their slopes.
Last edited:
A useful technique of course, that of serial elements to modify driver current.
Joe does not put a serial element in the driver loop.
And when asked to show the driver current, incorrectly posts the sum of the parallel structures.
An error in simulation. That is why I mentioned actual measurements of the branch currents with either clamp ons, or if one has the money, 4 DCCT's.
Jn
Last edited:
Yes, somewhat perceptive, but it isn't about impedance correction, I thought long and hard about that, discussed it with a number of persons that I respect. I now view it, not so much about impedance correction (like a Zobel across the VC is), but about current EQ.
True, been known a long time, but the voltage model is very persuasive as it seems to explain a lot of things. That is why it is hard to get past it in many people's minds. The dynamic driver is a current device and there is significant evidence to support that, including sound mathematics.
I think differently, at least I have an open mind that 'knowledge' might lead to better things. That is not unreasonable, it has in the past.
But electrostatic speakers are voltage devices. They 'obey' only the voltage in the same way that dynamic drivers, coils in magnetic gaps 'obey' only the current. Some here might not like that because it sounds like non-EE language. I get that, but I am a visual guy and and think conceptually.
This may surprise you, and again I have had this discussion with a number of persons, we tend to over complicate things. Often the best way is to distil things down to the lowest number of variables. This stops us from over thinking things and just get lost in minutia. In this case I decided to ask the question, what does the amplifier see? Since everything is in series and the only stable impedance is the Re, the DC resistance of the VC, then what happens at the Re impedance and what it does to the amplifier on the current side, this has been fruitful.
But in the end, some kind of measurement will be required. And it will be about distortion, harmonic distortion I believe. Is there a distortion mechanism, one that can settle it for good? I think there is. Time will tell and it might be quite soon.
WHERE?
Sigh...
Yes, I pretty much understand that, I have claimed nothing 'novel' so I can't be guilty, right?
Sigh...
I am going to bed.
There's your problem in a nutshell...
"Everything is in series"
It isn't. Not from your schematic.
The only thing in series with your driver is the driver inductance, both motional and Le.
Jn
I've said the same thing but it's not just words. There are possible merits to a speaker that presents a constant resistive load, but as Joe calls it "current friendly" it does not mean the distortion benefits of true current drive apply.
Probably unlike most here I visited Joe's physicist friend's web site. He has long article on how class A/B amplifiers are inherently incapable of driving a load where the voltage and current are out of phase. I don't follow it unless you are talking about 1970's era badly designed amplifiers. That's OK a physicist can have a marginal knowledge of EE but no one with a PhD in physics could possibly not have a deep knowledge of complex algebra.
This is where the "controlling" the current in the amplifier comes from. I find that Joe mixes all these things together in a way no one can follow and saying the you have to go beyond Thevenin and Norton shows at least some ignorance of the basics.
Last edited:
Forcing current and measuring voltage is one statement of Ohm's law, Dr. Thiele did not invent it
- Status
- Not open for further replies.