That has made it's rounds about the purported superiority. The fact, as per Ohm's Law, is that current flow is dependant on resistance, or an impedance and the voltage signal present. A current cannot be pushed. Don't believe a word of what is in that link about current drive being audio perfection and voltage drive being inferrior. Rod Elliot shared his feelings on this paper and I agree with his assessment- "current drive" is another "me too" topology wherein empty promises become just that. There is a reason that you can't buy a so-called "current drive amplifier".
Last edited:
search will find some more detailed discussions - current drive does avoid a few distortion mechanisms
but loudspeaker drivers, multiway loudspeaker crossovers are currently designed, optimized for Voltage drive amplifiers
if you want to repeat the experiments you can find going back 3-4 decades it could be entertaining - but it won't affect the dominant paradigm in audio today - given the long time it has been known and hasn't made even 1% inroads yet
but it appears that the distortions that Current drive reduce aren't in the top 2-3 issues of loudspeaker audible performance when you use moderately good drivers
but loudspeaker drivers, multiway loudspeaker crossovers are currently designed, optimized for Voltage drive amplifiers
if you want to repeat the experiments you can find going back 3-4 decades it could be entertaining - but it won't affect the dominant paradigm in audio today - given the long time it has been known and hasn't made even 1% inroads yet
but it appears that the distortions that Current drive reduce aren't in the top 2-3 issues of loudspeaker audible performance when you use moderately good drivers
Last edited:
I see.
If it's in the 4th or 5th tier in speaker performance when the drivers are efficient, it still seems like an interesting DIY project to consider nonetheless.
Yes you are right, I haven't seen any commercial Current drive amplifiers.
If there are any available to look at, let us know.
If it's in the 4th or 5th tier in speaker performance when the drivers are efficient, it still seems like an interesting DIY project to consider nonetheless.
Yes you are right, I haven't seen any commercial Current drive amplifiers.
If there are any available to look at, let us know.
Am I allowed to mention valves (tubes) in this thread??? Anyway, here goes:
An output tube operating in 'pentode mode' behaves approximately as a current source. (Because we are usually working on the 'flat' part of its characteristic).
This corresponds to the amplifier having a high output impedance.
If we change to 'triode mode' or 'ultra-linear mode', and/or introduce global negative feedback, the output impedance is reduced and we move more towards a voltage source.
An output tube operating in 'pentode mode' behaves approximately as a current source. (Because we are usually working on the 'flat' part of its characteristic).
This corresponds to the amplifier having a high output impedance.
If we change to 'triode mode' or 'ultra-linear mode', and/or introduce global negative feedback, the output impedance is reduced and we move more towards a voltage source.
All amplifiers are transconductance amplifiers. Wehave acouple amps that are variable tranconductance amps with output impedance variable from close to zero to VERY high.
Both the mentioned amplifiers have damping factor <1 so are in the current amp end of the spectrum.
dave
Not at all, that is more marketing talk. Essentially all amplifiers are in fact transimpedance amplifiers. They use resistive or active feedback wherein the difference causes a current to flow, that then causes the device, or multiples thereof to produce a voltage signal on the output.
Transconductance is largely the opposite and works by conveying a current signal that is equal to the difference in input voltage. This is sometimes a lower noise means by which to transfer and amplify waveforms between stages. However, to drive a loudspeaker, the signal must also comprise a voltage magnitude, as per Ohm's Law.
You are talking in circles. A current amp has fixed current output and variable voltage, a voltage amp has fixed voltage and variable current -- things get a bit more "wavy" near the zero crossing point.
Transconductance=impedance, conductance=transimpedance are all the same but for a change of sign.
dave
dave
Transconductance=impedance, conductance=transimpedance are all the same but for a change of sign.
dave
dave
I'm talking in circles? No offence intended, but you don't appear very educated in the engineering arena. It would do you some good to read some articles on the subjects of transconductance and transimpedance by people other than site sponsors, preferably written by licensed embedded systems engineers and pioneers in the fields who don't name companies after themselves, as the terms are not to be casually interchanged where they do not apply, or fail to otherwise act as descriptors thereof.
What about the F4?
It has no Voltage Amplification, there is actually a bit of loss.....
How does it fit into the "Current Drive" situation?
It has no Voltage Amplification, there is actually a bit of loss.....
How does it fit into the "Current Drive" situation?
Well, guys, there is some misunderstanding here.
The Ohm law works in any case. Either it's a voltage drive or a current drive, there is some voltage at the load and some current through the load. The point is - which one you control. You can have a low output impedance and control the voltage - current will follow the voltage change, or, you can have a high output impedance and control the current - voltage will follow according to the Ohm's law.
Why current drive is not really popular - is a completely different question. Definitely not because it's not possible to build one. But more because there's probably no need to build one. It requires at least different speaker setup and sonic advantages are not obvious... even though - the current is what actually makes the coil moving, so controlling the current seems to be a good idea... benefits don't look significant comparing to draw-backs... at least for me.
Cheers,
Valery
The Ohm law works in any case. Either it's a voltage drive or a current drive, there is some voltage at the load and some current through the load. The point is - which one you control. You can have a low output impedance and control the voltage - current will follow the voltage change, or, you can have a high output impedance and control the current - voltage will follow according to the Ohm's law.
Why current drive is not really popular - is a completely different question. Definitely not because it's not possible to build one. But more because there's probably no need to build one. It requires at least different speaker setup and sonic advantages are not obvious... even though - the current is what actually makes the coil moving, so controlling the current seems to be a good idea... benefits don't look significant comparing to draw-backs... at least for me.
Cheers,
Valery
I did not see that thread, sorry!
Post #8 there was entertaining, he seems to think this new book by Esa M is a revolution in amplifier design.
A novel idea that 99.9% of all amplifiers today are sonically flawed.
Post #8 there was entertaining, he seems to think this new book by Esa M is a revolution in amplifier design.
A novel idea that 99.9% of all amplifiers today are sonically flawed.
There are a few simple but important differences to be disambiguated. In the image above, there are two transconductance amplifiers and one transimpedance amplifier. Both of the shown transconductance amplifiers, the first one being a Yamaha discrete design (we know it today as the monolithic Burr Brown OPA627) and the other being an early Mark Levinson (this design now known among Parasound products), are being used in transimpedance mode. The Yamaha design can operate in both transconductance and transimpedance modes, the Mark Levinson pictured center can only operate in what is transimepdance mode, but functions under the rules of transconductance, and the transimpedance amplifier on the right can only operate in transimpedance mode. At an early date, Yamaha was awarded a variety of patents for their newer form of transconductance active Vas and bias mechanism (not shown, but consists of a small handful of additional devices) to operate in the current domain between gain, drive and output stages. Outside of the scope of interstage conveyance, the most practical means of operating the final load, ie loudspeakers was as vzaichenko mentioned, to remain in voltage operation. In recent years Krell has delved into interstage current domain operation in the same manner as Yamaha, only extending it to signal transfer over the interconnect wires. The First Watt designs appear identical to the Mark Levinson, but with the outputs replaced by Mosfets and an interstage autoformer for voltage operation. As noted in another thread, I've seen another company/individual build with the exact same idea around 1998.
Last edited:
'Transconductance' describes an amplifying stage (or stages). So, for example, if we regard the input to a stage to be a variation in voltage and we regard the output of the stage to be a variation in current, then we can regard that as a transconductance amplifying stage.
This thread is more about how the output of the 'output stage' in the amplifier interfaces with (or 'controls') the speaker. To characterise this interface we can examine how the amplifier output would react to a small change in speaker load (call it a thought experiment if you like). For example, if the amplifier is instantaneously delivering 1 amp (and 8 volts) to an 8 ohm resistive load, if the load suddenly increases to 9 ohms (it's just a thought experiment - I know it won't happen in practice) the voltage might stay at 8 volts and the current drops to 0.89 amps. That would be 'voltage drive' or we can say that the amplifier is behaving as a voltage source. Or, the current might stay at 1 amp and the voltage goes up to 9 volts, which would correspond to 'current drive' and the output stage behaving as a current source. Somewhere in between these two extremes would be found in practice, of course.
This thread is more about how the output of the 'output stage' in the amplifier interfaces with (or 'controls') the speaker. To characterise this interface we can examine how the amplifier output would react to a small change in speaker load (call it a thought experiment if you like). For example, if the amplifier is instantaneously delivering 1 amp (and 8 volts) to an 8 ohm resistive load, if the load suddenly increases to 9 ohms (it's just a thought experiment - I know it won't happen in practice) the voltage might stay at 8 volts and the current drops to 0.89 amps. That would be 'voltage drive' or we can say that the amplifier is behaving as a voltage source. Or, the current might stay at 1 amp and the voltage goes up to 9 volts, which would correspond to 'current drive' and the output stage behaving as a current source. Somewhere in between these two extremes would be found in practice, of course.
Going back a few years we used to have (more than) two design strategies of speakers.
There were the oriental origin lightweight speakers with cones with lightweight spiders. These speakers were comfortably driven by the common-all-garden amplifiers of the day.
Then there were the (predominantly) British speakers like the B&W's that utilised much stiffer suspensions. These speakers sounded dull when paired with the amplifiers above but came into their own with amplifiers that were designed to provide higher output currents - ie British amps like the Arcams.
Just my two penniworth.
There were the oriental origin lightweight speakers with cones with lightweight spiders. These speakers were comfortably driven by the common-all-garden amplifiers of the day.
Then there were the (predominantly) British speakers like the B&W's that utilised much stiffer suspensions. These speakers sounded dull when paired with the amplifiers above but came into their own with amplifiers that were designed to provide higher output currents - ie British amps like the Arcams.
Just my two penniworth.
- Status
- Not open for further replies.