Hear, hear.
Apart from the EQ and stability issues at low frequencies, pumping current into a resistive load seems equivalent to allowing a voltage to set the current. At high frequencies it's a different matter because of inductance that changes depending on cone position*, and that's where you want the amplifier's output impedance to be as high as possible to keep the current steady.
*The iron core surrounding the voice coil has an air gap, preventing it from being strongly saturated, so the coil can store energy in the iron. Therefore, as low frequencies modulate the geometry, this changes the impedance seen at high frequencies.
Say the impedance at 5kHz for a 12" woofer varies from 20 to 25 ohm. That's 1/5th variation in current, probably making it distort quite badly in that frequency range. Add 10 ohms and then it's only 1/7th. At 100 ohm output impedance, the variation is only 1/25th.
I think you will find, for ANY given unit, you can go closer to Xmax with Current Drive compared to Voltage Drive without THD going through the roof.. I refer you to the 2 Mills & Hawksford AES papers.
There's a HUGE Current Drive thread where I pontificate on other methods (eg ACE bass) which give you all the advantages of Current Drive without some of its Cons like susceptibility to drift.
I wait with bated breathe for someone to simulate this with 'realistic' stuff .. or better still, to build such a beast and measure it There's a slight difference (like > 10x) in bandwiidth between audio and motors/SMPSs
Last edited:
Thanks for this.
I have to confess that though I was confident to pontificate on the Impedance curve of the Heil AMT. I'm much less sure about the Magnepan LRS.
About the only thing I can vaguely conclude is that it is very inefficient cos hardly any wriggles can be seen including the 'bass' resonance. ie Electrodynamic Damping is very small.
So small that hardly any of the mechanical/acoustic load is reflected in the Impedance.
Well, I had a quick look at their paper here:
https://www.pearl-hifi.com/06_Lit_A..._Malcolm/Distortion_Reduction_with_iDrive.pdf
and their equation (1) says that for small Re and Le, cone velocity is inversely proportional to Bl. This doesn't make sense.
My point is that back-emf varies with Bl, such that, in the voltage-driven case, current varies to counteract variations in Bl. So, as Bl diminishes at the excursion extremes, more current is drawn, and since force (and therefore velocity) is proportional to current AND Bl, the effect of the diminished Bl is reduced.
If that's true then that's definitely good news for the OP and others who are currently interested in current drive.
Honestly, I don't think anyone here would do it for you because they wouldn't get paid for their time or efforts. You asked for details of existing feasible alternative techniques and I shared the same with you. Nevertheless, at this juncture, it is worth remembering that the requirement for a current drive amp is yours, and therefore it's going to be your time and your effort.
I was really directing you to the measured THD measurements at high level compared with voltage drive.
Behaviour of MC drive units near Xmax is complex and often non-intuitive as you've noted above. But in general, Current Drive gives less THD near Xmax ie the yucky effects of the coil running out of the gap, suspension non-linearities etc are alleviated.
In the specific case you mention, as the coil starts coming out of the gap (your 'Bl diminishes'), with Voltage Drive, current & force increases and the coil tries to jump out .. giving more THD. This is a dynamic and unstable situation and the 'cures' are often non-intuitive
BTW, in a MC drive unit, while Force is usually proportional to Current ... for a large part of its working range, Cone Velocity is proportional to Voltage This was realised by Rice & Kellog in their invention of the Moving Coil Loudspeaker and is fundamental to ALL MC speakers. Which is why Voltage Drive is universal today.
newvirus, I apologise for appearing to diss your suggestions for Current Drive amps. Mea culpa, mea culpa, mea maxima culpa
In my previous life, I've seriously investigated Current Drive amps several times over 2 decades in da previous Millenium. Ideas, which appeared good in theory, on closer examination or build, have exhibited practical cons and/or not given, what I consider true Current Drive. That's why I'm always asking for evidence of practical performance .. or even an accurate LTspice sim.
This Millenium, I'm a beach bum so my interest is purely academic ... as I believe is yours
It appears 'Current Drive', to most on this thread, is a tool to bodge the sound. If so, there are better and more consistent ways to bodge things eg ACE bass. For those interested in a beach bum (and some gurus) pontificating on the subject ...
Current Drive for Loudspeakers
But I'm interested in the original post. Someone suggested a guru in Jurassic times tossed a coin and went for Voltage Drive. But Rice & Kellog chose it cos they saw how practical speakers would work well (sound good) with that.
If they had chosen Current Drive, they would have needed MUCH bigger speakers. Apologies to da true speaker gurus for simplifying stuff.
I was hoping this thread would re-visit this fundamental issue and da true speaker gurus would join in. Alas, most people seem more interested in bodging.
Perhaps if a simple but 'true Current Drive' amp was available, there would be more people interested in designing Loudspeakers for Current Drive.
I have a serious question. And for real I am not sarcastic or cynical.
Do people ever use the search function on this forum?
Very recently there have been two topics going on about this subject, basically in parallel, where you can already find so many answers.
It's like repeating the same thing all over again......
I might as well just copy-paste the entire discussion here.
Do people ever use the search function on this forum?
Very recently there have been two topics going on about this subject, basically in parallel, where you can already find so many answers.
It's like repeating the same thing all over again......
I might as well just copy-paste the entire discussion here.
Although open-mindedness would be great, being skeptical is still normal and certainly not the problem.
What is disrespecful is that you're trying to get others to do all the dirty simulation work for you (using software of your choice) and simply "gift" you with a ready-made schematic of a fully functional amplifier, and that too, for free !!
My interest has surely been deeper that that. I suggested the non-linear school of thought because I've realised their potential in the real world. For example, I already had Mr. Putzeys' method successfully performing current control a few years ago. The thing is that I'm just not looking for a current-drive amplifier right now.I'm a beach bum so my interest is purely academic ... as I believe is yours
But don't you think it's somewhat disrespectful for a reference to be made, but without an appropriate link, so the intended recipient is left to their own devices to conduct their own search?
I just did so in #87 ... to find my posts in a parallel thread Current Drive for Loudspeakers. I do find it a chore to search the AES library for my own and other stuff which I have lost recently in several HD crashes
Please excuse senile oldies like me who are forgetful. If you go through that parallel thread, you'll see I forgot what I'd posted several years earlier.
Don't forget people post for different reasons
- Some people post to show off
- Some post to help or educate others
- and there are those, like me, who post in a naive attempt to persuade others to help them with difficult stuff they are too lazy or incompetent to do
Definitely disrespectful if 3. is intended. Want people to help you? Make it easy for them to do so.
Last edited:
Someone mentioned output-transformerless valve amplifiers for current drive. Well, I'm not so sure that the presence or not of the transformer is particularly important, but what about the following:
The SET (single-ended triode Class-A) amplifier is highly regarded in some circles - and this is because it offers lowish output impedance without the need for feedback (and resulting in an ear-friendly harmonic distortion profile). The low output impedance is desirable for the usual voltage drive, of course.
However, if you want current drive, the use of a pentode (or beam-tetrode) in a very similar configuration would be ideal. Same basic circuit as for the amplifier from a 1050s radio - but with a much higher quality output transformer. Also, the cathode bypass capacitor should be deleted, thus linearising the output stage and raising output impedance via local current feedback. (And of course the global feedback connection should be deleted.)
The SET (single-ended triode Class-A) amplifier is highly regarded in some circles - and this is because it offers lowish output impedance without the need for feedback (and resulting in an ear-friendly harmonic distortion profile). The low output impedance is desirable for the usual voltage drive, of course.
However, if you want current drive, the use of a pentode (or beam-tetrode) in a very similar configuration would be ideal. Same basic circuit as for the amplifier from a 1050s radio - but with a much higher quality output transformer. Also, the cathode bypass capacitor should be deleted, thus linearising the output stage and raising output impedance via local current feedback. (And of course the global feedback connection should be deleted.)
It's a slightly different question, though. My instinct is that it's a bit back-to-front to seek a new type of speaker to suit a particular type of amplifier. I prefer to think in terms of "active speakers" a la instrument amplifiers, where the components are permanently mated together, and in the optimum case you're either picking from a smorgasbord of options or designing everything from the ground up. But there's nothing wrong as such to try and push the envelope.
The 'promise' of current drive is just extrapolating out from existing results. If we can make the top half of an ordinary speaker sound way better, than maybe the bass could also be improved?
Interesting. Even if there would be a bump in bass response due to current drive this does not seem to bother at all in a listening test because everything sounds less distorted and very convincing better than before.
Now in the bass you have often room resonances playing mountain and valley with you.
Then many people manipulate highs and lows to their taste if they have this possibility on their preamp.
So it seems to me that the "bass resonance argument" is the last remaining defense against current drive by "voltage amp fans"?
Now in the bass you have often room resonances playing mountain and valley with you.
Then many people manipulate highs and lows to their taste if they have this possibility on their preamp.
So it seems to me that the "bass resonance argument" is the last remaining defense against current drive by "voltage amp fans"?