In this thread we will discuss current vs voltage drive of loudspeakers in an open and friendly way.
Here is the EDN article to kick the discussion off.
http://www.edn.com/design/consumer/...periority-of-current-drive-over-voltage-drive
I've put it here under solid state since we want to discuss not only the transducer and acoustical aspects, but also the amplifier and sensing requirements.
Here is the EDN article to kick the discussion off.
http://www.edn.com/design/consumer/...periority-of-current-drive-over-voltage-drive
I've put it here under solid state since we want to discuss not only the transducer and acoustical aspects, but also the amplifier and sensing requirements.
Last edited:
In this thread we will discuss current vs voltage drive of loudspeakers in an open and friendly way.
Here is the EDN article to kick the discussion off.
Loudspeaker operation: The superiority of current drive over voltage drive | EDN
I've put it here under solid state since we want to discuss not only the transducer and acoustical aspects, but also the amplifier and sensing requirements.
What about filters? It is all good-looking stuff when you use one speaker. How about calculate a filter for multi-way speaker? I bet all programs that are available for us assume Rout of amplifier to be zero or almost zero.
Last edited:
You are probably right.
If this current drive is to ever take off, it will mean a that speakers have to be characterized differently (or at least additional parameters will have to be provided).
I think this technique will probably also require some feedback mechanism - perhaps an extra winding on the voice coil for example. Ideally you want to control accurately the instantaneous position of the diaphragm. Difficult in practice, but not insurmountable. Don't know how this would work at HF though.
A technique probably suited to a DSP approach.
If this current drive is to ever take off, it will mean a that speakers have to be characterized differently (or at least additional parameters will have to be provided).
I think this technique will probably also require some feedback mechanism - perhaps an extra winding on the voice coil for example. Ideally you want to control accurately the instantaneous position of the diaphragm. Difficult in practice, but not insurmountable. Don't know how this would work at HF though.
A technique probably suited to a DSP approach.
some thoughts on Merilainen and current drive
I stand back in amazement at how many people, including knowledgeable engineers, are so dismissive of the work of Esa Merilainen. It is really obvious that they 'know about' this, but have never experienced it or worked with it. From my perspective, this argument reminds me of the story of the parade of circus elephants where everyone in the parade is holding the tail of the elephant in front of him: bitching about the smell and the view, but afraid to let go. So everyone in the parade criticizes the elephant goes of in a different direction, and achieves significantly different results.
Esa's work deserves some serious consideration and experimentation before all these judgements are given forth. The list of experts picking at his article in EDN is just amazing. I think he has delivered some of the most constructive, new thinking in a very long time. His comprehensive analysis of the problems that occur with voltage drive is very complete. Many of these problems are either greatly reduced with current drive, or disappear altogether. Have you actually read his book? Or do you just know that current drive can't work? Have you actually experimented with it and done some listening tests? Based on the most of the published comments the answer is obviously NO.
PMA: At the risk of offending, I think that problems of inductance, and especially resonances CANNOT be 'fixed' with a DSP. Chopping holes in the audio program just adds insult to injury. It does simply not fix the resonance problem(s), any more than the elaborate 1/3 octave filters of yesteryear did. Those who believe you can deal with resonances with a DSP have a very poor understanding of what resonance is and how it works, but then the audio business has always been very long of selling 'band-aids' to fix its perceived problems.
As far as it being a fashion topic: F=Bli ... I don't see a term for voltage anywhere in that formula! Maybe in fact, it is a direction worthy of serious exploration. I find it to be some of the most promising and exciting work I've done in a long time.
Loudspeakers are one of the nastiest, reactive loads you can place on an amplifier. And nearly every amplifier made today has global feedback right on its output ... so all the load caused foibles are introduced back into the amplifier ... which is my opinion, functions as an active distortion generator as a result. I'm convinced that this is the major cause of active distortion in music playback today. Lowering source impedance, and fancy speaker wires do not correct phase shifts and other problems in the loudspeaker.
I stand back in amazement at how many people, including knowledgeable engineers, are so dismissive of the work of Esa Merilainen. It is really obvious that they 'know about' this, but have never experienced it or worked with it. From my perspective, this argument reminds me of the story of the parade of circus elephants where everyone in the parade is holding the tail of the elephant in front of him: bitching about the smell and the view, but afraid to let go. So everyone in the parade criticizes the elephant goes of in a different direction, and achieves significantly different results.
Esa's work deserves some serious consideration and experimentation before all these judgements are given forth. The list of experts picking at his article in EDN is just amazing. I think he has delivered some of the most constructive, new thinking in a very long time. His comprehensive analysis of the problems that occur with voltage drive is very complete. Many of these problems are either greatly reduced with current drive, or disappear altogether. Have you actually read his book? Or do you just know that current drive can't work? Have you actually experimented with it and done some listening tests? Based on the most of the published comments the answer is obviously NO.
PMA: At the risk of offending, I think that problems of inductance, and especially resonances CANNOT be 'fixed' with a DSP. Chopping holes in the audio program just adds insult to injury. It does simply not fix the resonance problem(s), any more than the elaborate 1/3 octave filters of yesteryear did. Those who believe you can deal with resonances with a DSP have a very poor understanding of what resonance is and how it works, but then the audio business has always been very long of selling 'band-aids' to fix its perceived problems.
As far as it being a fashion topic: F=Bli ... I don't see a term for voltage anywhere in that formula! Maybe in fact, it is a direction worthy of serious exploration. I find it to be some of the most promising and exciting work I've done in a long time.
Loudspeakers are one of the nastiest, reactive loads you can place on an amplifier. And nearly every amplifier made today has global feedback right on its output ... so all the load caused foibles are introduced back into the amplifier ... which is my opinion, functions as an active distortion generator as a result. I'm convinced that this is the major cause of active distortion in music playback today. Lowering source impedance, and fancy speaker wires do not correct phase shifts and other problems in the loudspeaker.
There was very little new in there that hadn't been covered thoroughly by Mills and Hawksford.
My first exposure to the concept was from Nelson Pass. It looked interesting and I spent some time playing with it. I came away convinced that you could make it work well if you could design a driver from the ground up. Better than voltage drive? Maybe not.
it really isn't 1980 anymore
and even then this was quickly shot down
wrong - no one with actual engineering understanding of audio power amps, feedback theory, global feedback is going to talk about the issues this way
this is straight out of the 1970's-80's - repeated "theories" put forth to build egotistical Audio Guru's names, then analysis, peer reviewed replies, with hardware tests, real measurements showing the "theory" is vapor - effects in the noise with "conventional engineering" practice
you can search the AES eLibray for “interface distortion” or interface IMD as a directly relevant example of buzzword invention, self promotion, over issues covered by engineering textbooks for decades preceding
I suggest as long as you're into books try Cordell, Self's latest editions, anything on feedback by Putzeys
and even then this was quickly shot down
wrong - no one with actual engineering understanding of audio power amps, feedback theory, global feedback is going to talk about the issues this way
this is straight out of the 1970's-80's - repeated "theories" put forth to build egotistical Audio Guru's names, then analysis, peer reviewed replies, with hardware tests, real measurements showing the "theory" is vapor - effects in the noise with "conventional engineering" practice
you can search the AES eLibray for “interface distortion” or interface IMD as a directly relevant example of buzzword invention, self promotion, over issues covered by engineering textbooks for decades preceding
I suggest as long as you're into books try Cordell, Self's latest editions, anything on feedback by Putzeys
Last edited:
Once you dive into position feedback, isn't it instantaneous drive power that is controlled, and whether the power amplifier tends toward infinitely high or low impedance sort of inconsequential?
A must read for anyone interested in taking advantage of current drive.
http://www.diyaudio.com/forums/vendors-bazaar/157787-secret-tube-amplifiers-revealed-much-more.html
We have a couple of variable transimpedance amplifiers here, and there is an optimum output impedance for every loudspeaker.
dave
Hi,
The style is extremely grating. We already know current drive
can have some advantages (and difficulties) over voltage
drive, but it is a ludicrous position to take that voltage
drive is wrong, when your discussing using a 40mH
inductor in a basic 1.5 way to allow current drive.
rgds, sreten.
The style is extremely grating. We already know current drive
can have some advantages (and difficulties) over voltage
drive, but it is a ludicrous position to take that voltage
drive is wrong, when your discussing using a 40mH
inductor in a basic 1.5 way to allow current drive.
rgds, sreten.
I'm not convinced.
FYI, I've read Self and Cordell. Cordell's latest book is sitting right next to Merilainen's.
I arrived at this belief from years of practice of creating passive conjugate networks which when combined with the loudspeaker drivers, reflect a uniform load to the amplifier. The results far exceed what you might expect: not the least of them being a dramatic reduction in distortion in complex program material. The clarity improvement is not subtle.
Everytime I mention this belief, I can count on a number of people coming to shoot it down ... but the my audible results tend to suggest otherwise. The resultant sound output is much lower in distortion. Passive loudspeakers do not produce distortion all by themselves.
FYI, I've read Self and Cordell. Cordell's latest book is sitting right next to Merilainen's.
I arrived at this belief from years of practice of creating passive conjugate networks which when combined with the loudspeaker drivers, reflect a uniform load to the amplifier. The results far exceed what you might expect: not the least of them being a dramatic reduction in distortion in complex program material. The clarity improvement is not subtle.
Everytime I mention this belief, I can count on a number of people coming to shoot it down ... but the my audible results tend to suggest otherwise. The resultant sound output is much lower in distortion. Passive loudspeakers do not produce distortion all by themselves.
Member
Joined 2009
Paid Member
Good topic - I think you have started another epic thread
My own observations are a bit limited - but I have found that low output impedance (so-called voltage drive) is more hi-fi and dynamic to my ears than a higher output impedance. But the higher output impedance makes for some pleasant easy-listening. I don't consider one better than the other as it depends on the music and my mood.
LdSpkrPro - I'm very interested in your observations regarding dramatic improvements - do you have additional reading material ?
There is some truth in that statement. It depends on the power amplifier used, sometimes additional passive components added to "equalize" impedance vs. frequency do decrease distortion - or, better say, distortion is less dependent on frequency then, rather than decreased. But, the components to equalize bass resonance peak are really big and the method is thus not very practical. But the method leads to more flat amplitude response of the speaker. If you ask me for measurements of improved amplitude response flatness - I do not have them, but my colleagues who design speakers have such measurements.
Last edited:
