Yes its system problem, if system is small there is huge excursion happening on bass which then affects midrange played with the same driver. In this case there is benefit with current drive to reduce some of the distortion that comes with the excursion. On the other hand a system that does not have huge excursion and perhaps bass and mids are handled with separate ways for example, would have less benefit. Current drive gets system sound a little bit better at the edge performance, in other words system gets little bit louder before bad sound.
Yeah one can use current drive amplifier with very good drivers as well and get lowest distortion. Question is of course whether its audible improvement or not, marketing aside. In general the lower tech driver it is, and too small of a system for SPL and bandwidth requirements the more there is audible benefit from CC but it won't still fix the system, just make it wee bit better
We DIYers have possibility to design the distortion out by scaling up the system enough for distortion be not a problem, 15" three way speaker goes circles around any 5" driver two way system with current drive amp but not everyone can fit one in their homes or ideologies. For anyone looking for best sound then why not do current drive as well what ever the speaker is, or a hybrid like active system with traditional voltage control amps with some passive crossover parts tossed in to increase impedance to reduce distortion. What ever rocks the boat. Understanding on the stuff helps to figure out how to improve a system at hand.
Last edited:
Right.
Mixed feedback is used in most SS guitar amps today.
It HAS been done; and in fact transformer driven amps work in that mode, even if visibly speaker is "connected to emitters" in fact it´s in series with collector, since drive is floating, not ground referenced.
Power transistors work as current sources.
I remember it and even built a couple, problem is that power transistors add voltage gain to the mix (they current drive loads) and that often introduce instability.
This is one version:
Best numbers.
If you're looking for the best sound, you have to always ask the question what that even means and what is important and what's not.
For best numbers a passive filter most certainly is not gonna help btw.
I'm optimizing for best sound so talk for your self of course, when system is ideal, what ever that means in each situation, all audible issues are minimized like driver distortion if its audible. Its audible sooner or later with rising SPL.
My ideal system is active loudspeaker and instead of max driver distortion I have now reduced it with cost of few currency units. Perceived sound did change at least there is difference in noise so its better than before, there is some drom my DSP. Numbers are also better. There is no inductor on my big woofers as they didn't seem to help much but need to test again when more time.
Think about it, some passive components to manipulate impedance, low pass distortion, and all the rest is DSP like before with electrical damping resonance and all. I don't see benefits having all passive filters for my system, DSP is much more versatile. Noise would be better but its fine already. I can see why other people want passive filters and why they might sound better to some active systems, they reduce driver distortion!
of course, when system is ideal, what ever that means in each situation, all audible issues are minimized like driver distortion if its audible. Its audible sooner or later with rising SPL. My ideal system is active loudspeaker and instead of max driver distortion I have now reduced it with cost of few currency units. Perceived sound did change at least there is difference in noise so its better than before, there is some drom my DSP. Numbers are also better. There is no inductor on my big woofers as they didn't seem to help much but need to test again when more time.
Think about it, some passive components to manipulate impedance, low pass distortion, and all the rest is DSP like before with electrical damping resonance and all. I don't see benefits having all passive filters for my system, DSP is much more versatile. Noise would be better but its fine already. I can see why other people want passive filters and why they might sound better to some active systems, they reduce driver distortion!
They reduce driver distortion?
That statement is just simply not true at all.
Weirdest thing I have read in a while.
The only thing they can do, is reduce noise coming out of the system (amplifiers + other things)
But it will also cost you an awful lot more power.
For a good VC amplifier, impedance doesn't matter. (which is true for 99.8% of amplifiers today)
For a CC amplifier it does, but as long as you're using a good performing speaker, with a relative straight impedance (meaning plenty of demodulation) it doesn't matter either.
If you want to go for maximum performance, I can absolutely NOT see why someone wants to bother to go for passive filters.
They make the whole speaker-filter interaction non-lineair, which is even worse in series-crossover-filters.
Meaning for example if the Le(X) shifts with amplitude, it literally shifts the whole crossover around.
No, thanks. Frequency manipulation can also be done in a DSP.
In fact, this can even be done with a feedback loop to get even better results.
Btw, you still haven't described "best sound" you are still describing best numbers.
With that approach just hoping that it will be good enough.
Fact is, that a system can never be ideal, there are compromises. Which already starts by the recording and mixing.
Second it's a rather graceless (And I guess unscientific) approach.
Where you even run the risk of overlooking things or putting a lot of effort (time, money whatever) in things that aren't even important.
Improving something that isn't noticeable anyway, is not an improvement, but just a waste of resources.
Which is 100% fine from a subjective point of view, but it doesn't have anything to do with a "better sound".
That statement is just simply not true at all.
Weirdest thing I have read in a while.
The only thing they can do, is reduce noise coming out of the system (amplifiers + other things)
But it will also cost you an awful lot more power.
For a good VC amplifier, impedance doesn't matter. (which is true for 99.8% of amplifiers today)
For a CC amplifier it does, but as long as you're using a good performing speaker, with a relative straight impedance (meaning plenty of demodulation) it doesn't matter either.
If you want to go for maximum performance, I can absolutely NOT see why someone wants to bother to go for passive filters.
They make the whole speaker-filter interaction non-lineair, which is even worse in series-crossover-filters.
Meaning for example if the Le(X) shifts with amplitude, it literally shifts the whole crossover around.
No, thanks. Frequency manipulation can also be done in a DSP.
In fact, this can even be done with a feedback loop to get even better results.
Btw, you still haven't described "best sound" you are still describing best numbers.
With that approach just hoping that it will be good enough.
Fact is, that a system can never be ideal, there are compromises. Which already starts by the recording and mixing.
Second it's a rather graceless (And I guess unscientific) approach.
Where you even run the risk of overlooking things or putting a lot of effort (time, money whatever) in things that aren't even important.
Improving something that isn't noticeable anyway, is not an improvement, but just a waste of resources.
Which is 100% fine from a subjective point of view, but it doesn't have anything to do with a "better sound".
Yes the first part of my question was redundant, the driver impedance always changes with load 😀* load impedance does vary with frequency, and that´s the point of taking the extra job of specially creating an amplifier with >0 generator impedance.
End result is that you introduce some extra equalization in the mix, which is usually deemed desirable.
* you mean adding some extra circuitry so generator impedance tracks load impedance?
Not me, know no one else who does, it would fill more of an intellectual desire than a practical one, but hey, if somebody wants to walk the extra mile, please be my guest
FWIW last Saturday there was an "Entrepreneur´s Fair/Show", including those related to Musical Instruments, and I was asked to send a sample, so I gladly obliged.
It was held at the City Stadium at La Plata , the province´s Capital, some 45000/55000 young visitors assisted.
View attachment 1094510
why do I mention it? .... because it was the only SS amplifier amongst all kinds of Tube ones also shown there ... and musicians kept coming back, go figure.
part of the "special sauce" behind that? ....
well, high impedance low damping mixed feedback, of course.
Might have used pure current drive, but then sound becomes too extreme, razor sharp chest thumping, "too much of a good thing is not good any more"
View attachment 1094513
You got the real question I was after, whether you adjusted the feedback network to maintain constant damping across frequency. I haven't seen it before but was curious if you had given your experience. Interesting concept at least but perhaps counterproductive to good sound (too much of a good thing as you say about current drive). Mixed impedance provides more damping at resonance, where you generally want it most.
Tom, I'm very interested in your Transconductance Amp. Could you post the schematic please. I'm hoping that, like your other designs, it would be fairly simple ... and yet still maintain Zo > 1k from 20 - 20k Hz.
I wrote this in another forum circa 2006.
Richard Lee :
It is entirely possible for an "amplifier" to know what the speaker is doing.
It needs to sense the speaker current. An amp which twiddled its Output Z using both current & voltage feedback does this. Speakers act as accurate microphones (sense the actual sound) if operated into Low Z.
See "Loudspeakers as Microphones" - Peter Baxandall special lecture London AES (early 80s, late 70s?)
If operated into High Z, then the voltage at the terminals is a measure of cone velocity.
Both these mechanisms obey superposition & Thevenin so if you're clever, you can look at this while the amp is giving zillion volts and amps to the speaker. But non est tantum facile.
There are several tried & tested methods of using this "controlled output Z" or "current + voltage feedback" or "actual sound & motion feedback" (different descriptions of the same thing) if you incorporate the amplifier design in the speaker. Some of these are in the zanier incarnations of my Powered Integrated Super Sub technology.
The simplest is the negative output R that Fons mentions.
More sophisticated but similar (??!) is ACE technology by Erik Stahl which was used by Audio Pro, Sweden for subs. Unfortunately, since he left, there isn't anyone there who understands it. Anyone have a contact for Erik? Or a clean copy of his original AES preprint?
These methods have the distortion reduction and dynamic overload protection features discussed in Mills & Hawksford. However, they are badly affected by heating of the voice coil.
David Birt (?) did an excellent IoA paper at Windermere where he arranged speaker and amp in a bridge so he could measure and compensate for heating on the fly. Anyone have an email for him?
These are the most elegant methods and they can be analysed from many viewpoints. Some of these viewpoints don't show up the distortion reduction advantages clearly.
I'm contemptuous of methods which rely on extra transducers or extra windings (like Mills) or zillion point DSP EQ especially if they don't give ALL the advantages of the elegant methods.
A brute force zillion point approach possible today is measure accurately speaker Z (not that easy) and tailor the output of a High Z amp to suit. This would give some but not all the advantages of the above systems cos it wouldn't "know" what the speaker is up to.
Richard Lee :
It is entirely possible for an "amplifier" to know what the speaker is doing.
It needs to sense the speaker current. An amp which twiddled its Output Z using both current & voltage feedback does this. Speakers act as accurate microphones (sense the actual sound) if operated into Low Z.
See "Loudspeakers as Microphones" - Peter Baxandall special lecture London AES (early 80s, late 70s?)
If operated into High Z, then the voltage at the terminals is a measure of cone velocity.
Both these mechanisms obey superposition & Thevenin so if you're clever, you can look at this while the amp is giving zillion volts and amps to the speaker. But non est tantum facile.
There are several tried & tested methods of using this "controlled output Z" or "current + voltage feedback" or "actual sound & motion feedback" (different descriptions of the same thing) if you incorporate the amplifier design in the speaker. Some of these are in the zanier incarnations of my Powered Integrated Super Sub technology.
The simplest is the negative output R that Fons mentions.
More sophisticated but similar (??!) is ACE technology by Erik Stahl which was used by Audio Pro, Sweden for subs. Unfortunately, since he left, there isn't anyone there who understands it. Anyone have a contact for Erik? Or a clean copy of his original AES preprint?
These methods have the distortion reduction and dynamic overload protection features discussed in Mills & Hawksford. However, they are badly affected by heating of the voice coil.
David Birt (?) did an excellent IoA paper at Windermere where he arranged speaker and amp in a bridge so he could measure and compensate for heating on the fly. Anyone have an email for him?
These are the most elegant methods and they can be analysed from many viewpoints. Some of these viewpoints don't show up the distortion reduction advantages clearly.
I'm contemptuous of methods which rely on extra transducers or extra windings (like Mills) or zillion point DSP EQ especially if they don't give ALL the advantages of the elegant methods.
A brute force zillion point approach possible today is measure accurately speaker Z (not that easy) and tailor the output of a High Z amp to suit. This would give some but not all the advantages of the above systems cos it wouldn't "know" what the speaker is up to.
Betcha could if you wanted to. Design your own - with vanishing low distortion too
^ iron core is subject to hysteresis and it would be better to use air core inductor in this application. I have no experience how much of a difference it would make but better be sure as we are trying to reduce system distortion. Here is really good illustration on hysteresis and seems to result worst kind of distortion as it is not masked https://purifi-audio.com/2020/04/28/dist/
Yes it seems so, system distortion is reduced which is mostly due to non-linearities in driver. Its application dependent if power loss is something to worry about isn't it, small/expensive inductors don't have much resistance.
Frequency manipulation in DSP and increase series impedance with passive parts yeah.
I've been testing mid driver with 1mH coil in simulator in series and if looking Audioxpress Test bench driver measurements Le(x) voice coil inductance varies say 50uH at worst with very high excursion. The crossover is pretty much unaffected by such small difference but impedance in the circuit is now doubled or quadrupled for high frequencies. I mean the coil is just to increase impedance above pass band and actual filtering/shaping is done in DSP. Driver motor non-linear effects like hysteresis and Le(x) turn into acoustic sound through current and reducing current by increasing impedance in the circuit reduces distortion emitted acoustically. Same happens with current drive, but without losses as you state, current amplifier has very high output impedance which makes series impedance in the circuit very high and non-linear driver motor does not produce about any (distortion) current into the circuit to turn into acoustic sound.
I'm not too familiar with all of this but this is the general idea, to increase impedance to reduce current also error current due to fluctuating impedance and what not. Voltage control amplifier in active system hooked directly to a driver is practically a short between driver terminals and driver motor non-linearities are able to have maximum effect on current in the circuit. Increase this impedance and less current, less distortion.
Compromises are obvious and I'm very much into them thinking about stuff and trying to taking the right ones, what matter and what not. As I'm DIY hobbyist there are different objectives and freedoms businesses have. When compromise is between reduced distortion to small amount of money its an obvious one for me, a company could take the less money approach. Think this, active system voltage amp, woofer low passed below cobe breakup with DSP. Now the distortion generated by the driver motor non-linearities are not low passed and get even amplified with the cone breakup and this happens above the woofer passband, less masking. In a system this is basically woofer emitting treble (distortion) and my hypothesis is that its this bare naked distortion is audible if any. Hense low passing it makes sense and perhaps audible difference but need to make tests to see if it really is audible. Waiting for free time kids are not home to be able to measure properly. As I have very big system, excursion stay low, its possible there is no uadible diffrence other than reduced noise, which I gladly take.
I always play around with ideal concepts, have you tried it? Try and imagine / simulate / pencil system for your application with ideal drivers and see what you come up with, lots of fun. Then just replace the ideal parts with best, or good enough, real world parts and there you have it.
Last edited:
This is what we need to know.
In the other thread where we were talking about this, an iron cored inductor is used in series with a pair of woofers and is said to be an improvement.
Yeah, there's that. But isnt it just a single knob to tweak, a.k.a. the old "Damping Factor" control? Perhaps "Output Z;" 2 - 20 - 200 Ohms in one turn of a knob. I thought I read dave saying he had such an amp (or was demo'd) that gave some interesting effect on some of his speakers.
Anyway, such a thing would be a specialty piece, like Pass's F1J. "As you may already know, the F1 is an active current source amplifier which offers unique characteristics, particularly to sensitive full/wide range loudspeaker drivers".
Iron core inductor is certainly smaller and cheaper for same function with downside on possible issue with hysteresis affecting perceived sound. Perhaps there is still more benefits than downsides even with the iron core but it could also be expectation bias. What measurements show? Perhaps speculating the hypothesis that distortion products emitted past crossover, out of band of the driver, really are quite audible and reducing them is audible improvement? Also hysteresis distortion is low passed with the inductance in series with voice coil, from both the iron core inductor and the driver if it has ferromagnetic material in the motor.
Truth is I'm not too professional on this or have about any experience on passive crossover and hysteresis for example so my writings are based on reasoning and what feels logical to me with the knowledge I have. Here is the philosophy: I think its the end use, acoustic sound, that matters and everything from sources to amplifiers to speaker structure more or less affects it, what the sound is and how its radiated. Its philosophical thing as applications and subjective preference vary its hard to give more exact advice / thoughts on specifics. As many/most/all issues are minimized in the whole system then perceived sound ought to be sublime. Distortion is just one among many "issues" and not very significant one in comparison to other aspects like frequency response, or bass extension, or room acoustics. If going for best sound be sure to understand the application well enough to be able to strive for best possible sound, what ever that is, minimize issues with given design freedom from starting from the worst. I'm sure everyone have their applications and philosophies and not trying to push mine, just some background info for my output.
Last edited:
The hysteresis curve, a closed locus of points, becomes very narrow if the flux density is kept sufficiently low. The disadvantage is that would require a larger core and thicker copper windings. This is what simple intuition tells, but one has to do some calculations to determine whether using a larger core actually results in less energy being taken by the core, which is what contributes to distortion. Since air can support very large magnetic fields without becoming saturated in the same way as iron, air cored inductors seem to be a better choice.