Hi Allen
Actually, it only responds to the current. There are many misunderstandings here. Voltage can only create a potential for current. The driver responds to that current when it presents a load. If a specific current exists at a certain frequency through the coil, if that current is the same under both current drive and voltage drive, the output will be the same dB-SPL with either drive at that frequency.
Also, and I know this may cause some to see red, with dynamic drivers there is no such thing as voltage drive. Yes, there is such a thing as a voltage source and amplifiers can be called thus. But way back in a 1947 Wireless World, Langford-Smith realised that the DC resistance would need to be zero and later Neville Thiele came up with an actual equation that proves it. I know this is perhaps difficult to wrap one's head around, but there you are. Yes, it took me time to come to this.
Think about for a moment, the huge majority of the voice coil is outside the magnetic gap (overhung). If we were to measure the inductance, it would most come from the gap. On average, the 70% outside the gap would contribute little inductance. That means that 70% of the DC resistance is outside the gap and is acting little more than as a low inductance wire-wound resistor. But when you think about it, the DC resistance inside the gap is also to be added anyway, so it doesn't even matter whether the driver is underhung or overhung. It is the very fact that the coil has any DC resistance that prevents any voltage drive, because it appears in series with the output impedance of the amplifier.
But adding anything to that DCR (Re) will alter that alignment and as Small taught that the damping of the system (LF) is entirely a function of the alignment you end up with. So yes, damping is affected by any change in alignment.
Current drive is a different matter, but here the idea that the coil is driven is open to questions, IMO.
Hope nobody gets upset by the above. It has been well thought out in my mind and maths is there to back it up.
Last edited:
Scottjoplin
The book of engineer Esa Meriläinen named "Current-Driving of Loudspeakers Eliminating Major Distortion and Interference Effects by the Physically Correct Operation Method" explain your question in detail. You can have more details in Current-Drive - The Natural Way of Loudspeaker Operation
Last edited:
For voltage drive and current drive Linear part will be the same if equalized for the same impulse response. (both phase and frequency). If someone use linear phase EQ here the responses will be different.
I am sure it is well thought out in your mind, but it is incomprehensible to those with training in the subject under scrutiny.
Measurements?
I will state the facts and people can make up their own minds. Here are some hard facts:
I have a driver before me here, this is for real and not vapour. It is a high quality Danish driver which has some state-of-the-art specs, it is not exactly a cheapie. It also happens to be an excellent sounding driver.
Some facts about this driver are:
Voice coil = 16mm
Voice Coil Gap = 5mm
Peak Linear Excursion: 11mm (do the maths)
Inductance @ 1KHz = 0.2mH (this is measured, not rated)
Wire type = Aluminium or CCAW (copper-clad aluminium wire)
DC Resistance of Coil = 5.7 Ohm
Wire Diameter = 0.2mm
Turns = 120 approx
Amount of Turns inside Gap = 38 approx (do the maths)
Amount of Turns outside the Gap = 82
The key to the above numbers is looking at the inductance of 0.2mH is almost entirely formed by the 38 turns that are inside the gap. The external 82 turns create almost no inductance at all. That 0.2mH inductance is associated with the BL force factor created inside the gap, no such force is exerted outside the gap.
DC Resistance per turn = 0.048 Ohm approx.
That means the external resistance of 82 turns = 3.9 Ohm DCR (do the maths)
That means 41 turns are in front of the gap = 1.95 Ohm DCR. That is 2 Ohm rounded off.
That means that 41 turns are at the rear of the gap = 1.95 Ohm DCR.
So we basically have 2 Ohm DCR in front of the gap and another 2 Ohm behind the gap.
4 Ohm DC resistance in total.
This is nothing more than a low inductance aluminum wire-wound resistor that is added in series with the output impedance of the amplifier. The output impedance of the amplifier might well be 0.0001 Ohm and have extreme 'damping factor' and yet the truth is that in this case and in all other cases, that vanishingly small output impedance is swamped by the DC resistance of the voice coil, and not just by the resistance outside the gap, but also the total resistance because they are in series it is the total resistance that matters. The amplifier can only see a single value because they are all in series.
The fact that the external resistance of the voice coil is 70% outside the gap is just a way of proving that sometimes the proof is hiding in plain sight:
Just look at the voice coil of an overhung driver and you can see it with your own eyes!
So Scott, the above contains measurements (can be done with most multi-meters and one that can measure inductance too) and totally solid maths. If that doesn't satisfy you, then you win and I lose... not? Let others make up their own mind, I have done the inquiry as best as I can.
So to me there are amplifiers that are voltage sources, but that does not guarantee that there is such a thing as voltage drive. Current drive is a different beast however, but even here that are facts that do not add up with the popular view of what current drives is supposed to be? Some have suggested that current steering is a better term.
Strangely, electrostatic speakers are voltage/force-driven, so that is indeed voltage drive. So it does exist.
People can make up their own minds. I just present the facts.
I will state the facts and people can make up their own minds. Here are some hard facts:
I have a driver before me here, this is for real and not vapour. It is a high quality Danish driver which has some state-of-the-art specs, it is not exactly a cheapie. It also happens to be an excellent sounding driver.
Some facts about this driver are:
Voice coil = 16mm
Voice Coil Gap = 5mm
Peak Linear Excursion: 11mm (do the maths)
Inductance @ 1KHz = 0.2mH (this is measured, not rated)
Wire type = Aluminium or CCAW (copper-clad aluminium wire)
DC Resistance of Coil = 5.7 Ohm
Wire Diameter = 0.2mm
Turns = 120 approx
Amount of Turns inside Gap = 38 approx (do the maths)
Amount of Turns outside the Gap = 82
The key to the above numbers is looking at the inductance of 0.2mH is almost entirely formed by the 38 turns that are inside the gap. The external 82 turns create almost no inductance at all. That 0.2mH inductance is associated with the BL force factor created inside the gap, no such force is exerted outside the gap.
DC Resistance per turn = 0.048 Ohm approx.
That means the external resistance of 82 turns = 3.9 Ohm DCR (do the maths)
That means 41 turns are in front of the gap = 1.95 Ohm DCR. That is 2 Ohm rounded off.
That means that 41 turns are at the rear of the gap = 1.95 Ohm DCR.
So we basically have 2 Ohm DCR in front of the gap and another 2 Ohm behind the gap.
4 Ohm DC resistance in total.
This is nothing more than a low inductance aluminum wire-wound resistor that is added in series with the output impedance of the amplifier. The output impedance of the amplifier might well be 0.0001 Ohm and have extreme 'damping factor' and yet the truth is that in this case and in all other cases, that vanishingly small output impedance is swamped by the DC resistance of the voice coil, and not just by the resistance outside the gap, but also the total resistance because they are in series it is the total resistance that matters. The amplifier can only see a single value because they are all in series.
The fact that the external resistance of the voice coil is 70% outside the gap is just a way of proving that sometimes the proof is hiding in plain sight:
Just look at the voice coil of an overhung driver and you can see it with your own eyes!
So Scott, the above contains measurements (can be done with most multi-meters and one that can measure inductance too) and totally solid maths. If that doesn't satisfy you, then you win and I lose... not? Let others make up their own mind, I have done the inquiry as best as I can.
So to me there are amplifiers that are voltage sources, but that does not guarantee that there is such a thing as voltage drive. Current drive is a different beast however, but even here that are facts that do not add up with the popular view of what current drives is supposed to be? Some have suggested that current steering is a better term.
Strangely, electrostatic speakers are voltage/force-driven, so that is indeed voltage drive. So it does exist.
People can make up their own minds. I just present the facts.
Hi Joe. My local bottlo availed itself of some Carlsberg and I don't mind a pilsner so I brought some home 
The basic premise is simple, would you agree? Whether you drive from a current source, or a Voltage source or anywhere in between.. the response changes but after this is equalised conventionally, the response becomes the same, and the impulse becomes the same.
Except the main difference is the non-linear behaviour.
What may also be interesting is what this means and what we could do...
The basic premise is simple, would you agree? Whether you drive from a current source, or a Voltage source or anywhere in between.. the response changes but after this is equalised conventionally, the response becomes the same, and the impulse becomes the same.
Except the main difference is the non-linear behaviour.
What may also be interesting is what this means and what we could do...
Hi Joe. My local bottlo availed itself of some Carlsberg and I don't mind a pilsner so I brought some home
Brewed locally in Victoria I suppose, under license. Dan Murphy usually sells Carlberg Elephant Beer at 7.2% alcohol and that is brewed at the original brewery in Frederiksberg (in the south-western suburb of Copenhagen) and not far, less the 2 K's I believe, Frederiksberg Hospital where I was was born.
But locally, I think that Tuborg is more popular, but that is owned by Carlsberg too. If you ever get to the brewery, ask for Carlsberg Master Brew which is a full 10%, just to try it, but then maybe not drink it again, it is just too much.
Precisely. But will the non-linear behaviour be different? If you pick 50 Hertz and figure out the current to produce a target dB-SPL, let us assume 90dB is achievable at 50 Hertz. Now connect to a current source and adjust the level of current to match that of the voltage source. The dB-SPL will be the same. Will the current source be more linear than the voltage source, or vice versa? I think I have an inkling what the answer is - but I want to actually confirm it. And I won't be rushed. And I won't rush to report it here. "Don't throw pearls before swine" was a quote by you-know-whom two thousand years ago.
But ScottJ is right, acoustic distortion, with nearfield LF measurements to get the ambient noise floor down, this should be done. But he seems to think this is an easy thing to do. What stimuli is but one question?
The idea that a current source cannot be used to create a proper electrical LF Butterworth, Bessel or Chebychev (sealed box) when driving from a current source, is not correct, This is not theory, it has been done and I have one example right next to my left foot right now.
EDIT: PS, the name Carlsberg means "Mountain of Carl" who was the Founder's son. But it was more of a hill than a mountain. I know, I am a local native.
Last edited:
Yeah, I know, this can be a bit daunting when it is first explained. I was first exposed to this in 1975 and it took me a while to understand it too. It's actually imbedded in this equations:
In these equations, the Re is the total resistance seen by the amplifier, including its own output impedance and the DC resistance of the driver.
The example, I gave just makes it visually obvious. Most of the voice is nothing more than acting as a wire-wound resistor and it would be no different that putting that right on the output of the amplifier. Even the amplifier had a zero Ohm output impedance, it now becomes a 4 Ohm output impedance.
But it isn't just the "wire-wound" resistor part that is the problem, but in fact, the whole DC resistance, this includes the other 30% that is inside the gap, is also part of that resistance. That total is the Re.
The idea of voltage drive is simply not what is happening because Re prevents it. What happens is that the voltage of the amplifier causes a current to flow when a load is connected. That is not voltage drive since the driver is responding to the current via Re load and never directly to the voltage.
Current drive may be a different matter. There 100% of the current flows through the voice coil. But the word 'drive' implies some kind of control. Yet some will say we loose control with current drive, so even here it is not so simple.
So I hope that helps make it a bit clearer. But when Richard Small sets you straight, you just can ignore those equations above.
Last edited:
Joe I find all this very confusing and unnecessary obtuse.
If you consider a driver as a black box, and you connect it to a voltage source , then voltage drive that driver. By definition.
Connect it to a current source and you have current drive, by definition.
I believe most of the confusion comes from the fact that you seem to try to separate the voice coil resistance from the inductance, and that you consider the resistance separately. Then you conclude that, because there is a resistance in the circuit, you cannot have voltage drive.
But that is a fallacy; the driver is not something without the resistance, it wouldn't work without it. The driver has two terminals, where you connect the driving source, which can either be a voltage source for voltage drive, or a current source for current driving (or something in between).
In the case of voltage drive, the load current is determined by the complex impedance of the driver; there's the inductance, the DCR of the voice coil, the reactance of the magnetic circuit and probably some more, and that complex determines the load current depending on frequency and level. It appears illogical to just separate the DCR out of that complex and then conclude there's no voltage drive. It's just nonsensical.
I feel you make it needlessly complex, and wrongly so.
Jan
Last edited:
My daughter hopes it's a cat named Mr. Butterworth.I have one example right next to my left foot right now.
You could say the same thing about a Voltage source, but would it still be one then? A current source?
SA actually, at the Coopers factory.
Our cat was named Coco and we took him to the vet last Friday with chronic kidney problems. Sigh. We are currently without a house cat.
Who does that?
Last edited:
Hi Jan
I am sorry that you find it confusing. Obtuse? Did you ever watch The Shawshank Redemption? Not obtuse at all. But it's a great movie.
Shawshank Redemption (1994) - how can you be so obtuse? - YouTube
I don't view the driver as a black box. To me it is anything but that. And I have many reasons for saying that.
If you mean that you have 'conventions' and if I was to adopt a conservative mindset, does not that also stopping looking if these are based more on convenience and a STOP sign? I went and looked behind the sign and said oops!
Definition of a convention: A way in which something is usually done.
Yes, it is a convention, but based on what I know with certainty, I am bothered by it. There is stuff that doesn't add up.
The same goes for a definition, what is that exactly? And I find the idea of a black box even less than satisfying. Just saying.
Wrongly so implies a moral judgment, I can't see that applying here.
I agree that thinking in terms of conventions and black boxes is simplifying things. But the devil is in the detail.
When I pull a driver apart, and I have done a few
, try to figure out what the designer had in mind, so many choices and variables. Then should I gaslight my own eyes when I see on either side of the magnetic gap what amounts to nothing more than two passive wire-wound resistors that you might as well put directly on the positive (+) output of the amplifier? That I cannot do.The point I was trying is in fact to keep it simple and even so it could be visualised. Then I am told that I am complicating things when in fact I am trying to make it as simple as possible. Clearly I failed.
But imagine you adding a 3-4 Ohm resistor on the output of your amplifier? Can that amplifier still be considered to performing 'voltage drive'?
I think not.
The Thiele-Small Parameters, please take a look at them. The truth is in the numbers.
Lowthers have extremely low Qes and Qts. The reason? Strong magnets? That helps. Short voice coils? That has a lot to do with it. Four layers rather than the usual two, I am very familiar with the maths regarding short and long voice coils, the BL force-factor, and the patterns that changing one thing has on the other (doubling Re lowers electrical damping by two (Qes>)) and what it does to the real damping which is always the alignment. Lowthers have an Xmax of 1mm and max linear excursion of 2mm. What would happen if we lengthen the voice coils, the diameter and type of wire, the DCR and so on. Please Jan, I have done my diligence and not just talking of the top of my head. I understand these patterns well and they occupy driver designers too. I know what happens when I change where the resistances are and their effect on Re and how that affects... etc, etc... I have done the hard yards and I trust the maths and reality more than conventions and undefined definitions.
I can only hope that you see that. And thanks for joining in. BTW, the DIY current source amplifier that this thread is about, I built this to be a tool and have been using in surprising ways and the results would surprise many. For example, a current source can lead to higher damped alignments than voltage drive (no, I am not kidding and the way to do it is rather simpler than you'd expect). The above is based on hands-on experience and not just theory. I will document these and make it available. Until then, maybe I should say less and just get on with it - even if these things take time and patience.
Cheers, Joe
Last edited: