Here's some speculation as to why/how different caps sound different. This isn't my own idea, it was inspired by the late Cyril Bateman who conducted some seminal research into capacitor distortion.
First, consider what a capacitor is. It's two plates of conductive material separated by an insulator. The simplest capacitor would be two flat plates of copper (or some other metal) with an air gap between them. The capacitance is a function of the area of the conducting plates and the distance between them. If the distance between them changes, the capacitance changes too. This effect can be used to make a capacitor (aka condenser) microphone. In a microphone one of the plates is a rigid metal conductor and the other is an extremely thin metal conductor (or metalized plastic conductor) which is free to move as sound waves impinge on it. The movement of the thin metal plate causes the distance between it and the rigid metal plate to be modulated, thus the capacitance is modulated, and that in turn can be used to generate and electrical signal. However, the distance between the two plates, and therefore its capacitance, is different when the capacitor has a polarizing voltage on it compared to when it doesn't have a polarizing voltage. The effect can be seen by looking at a reflection in the the shiny surface of the flexible plate (usually referred to as the diaphram) and seeing how it is deflected when a polarizing voltage is applied (usually from "phantom power").
The converse of a capacitor microphone is an electrostatic loudspeaker. In the case of an electrostatic loudspeaker there is a fixed/rigid plate and a flexible conductive plate which is forced into motion by modulating a polarizing voltage between the two. It can be easily observed that the "flexible plate", ie the diaphram, of an electrostatic speaker is limp when the speaker is switched off but is pulled tight when the speaker is powered-up.
So what does all that have to do with capacitors in (non-electrostatic) loudspeaker crossovers? Cyril Bateman's theory was that the physical construction of capacitors was significant if it allowed the two conducting plates to move. In a practical capacitor, as opposed to our super-simple "two conductors with an air gap between them" example, there's usually two thin conductive metal foils separated by a film of insulating plastic material, and the whole thing is rolled-up into a cylinder-shaped object. Alternatively, instead of using two metal foils and an insulating film, a metalized film can be used, ie a plastic film which has had a thin conductive layer applied to its surface by chemical deposition.
Either way, as a voltage is applied across the two terminals of the capacitor the plates will be attracted to each other and if they're free to move they will move closer to each other and the capacitance will change. If a DC voltage is applied the plates will move and stay in their new position with a new capacitance value. However, if an AC voltage is applied the plates will move in sympathy with the voltage and the capacitance will change over the course of the AC cycle. That modulation of capacitance over an AC cycle causes distortion.
Now it isn't difficult to image that a capacitor that is made of very thin material and is wound very loosely into a cylinder is able to expand and contract as AC is applied to it, whereas a capacitor which is made of thicker material and tightly wound and encased in epoxy resin has its tendency to expand and contract restricted. Which capacitor would you expect to suffer less capacitance modulation over an AC cycle? Therefore which capacitor would have less distortion?
Anyway, that's speculation to some extent, but I hope you can see the scientific logic in the argument. Maybe it explains why some capacitors sound different from others in loudspeaker crossovers.
Or maybe it doesn't.
First, consider what a capacitor is. It's two plates of conductive material separated by an insulator. The simplest capacitor would be two flat plates of copper (or some other metal) with an air gap between them. The capacitance is a function of the area of the conducting plates and the distance between them. If the distance between them changes, the capacitance changes too. This effect can be used to make a capacitor (aka condenser) microphone. In a microphone one of the plates is a rigid metal conductor and the other is an extremely thin metal conductor (or metalized plastic conductor) which is free to move as sound waves impinge on it. The movement of the thin metal plate causes the distance between it and the rigid metal plate to be modulated, thus the capacitance is modulated, and that in turn can be used to generate and electrical signal. However, the distance between the two plates, and therefore its capacitance, is different when the capacitor has a polarizing voltage on it compared to when it doesn't have a polarizing voltage. The effect can be seen by looking at a reflection in the the shiny surface of the flexible plate (usually referred to as the diaphram) and seeing how it is deflected when a polarizing voltage is applied (usually from "phantom power").
The converse of a capacitor microphone is an electrostatic loudspeaker. In the case of an electrostatic loudspeaker there is a fixed/rigid plate and a flexible conductive plate which is forced into motion by modulating a polarizing voltage between the two. It can be easily observed that the "flexible plate", ie the diaphram, of an electrostatic speaker is limp when the speaker is switched off but is pulled tight when the speaker is powered-up.
So what does all that have to do with capacitors in (non-electrostatic) loudspeaker crossovers? Cyril Bateman's theory was that the physical construction of capacitors was significant if it allowed the two conducting plates to move. In a practical capacitor, as opposed to our super-simple "two conductors with an air gap between them" example, there's usually two thin conductive metal foils separated by a film of insulating plastic material, and the whole thing is rolled-up into a cylinder-shaped object. Alternatively, instead of using two metal foils and an insulating film, a metalized film can be used, ie a plastic film which has had a thin conductive layer applied to its surface by chemical deposition.
Either way, as a voltage is applied across the two terminals of the capacitor the plates will be attracted to each other and if they're free to move they will move closer to each other and the capacitance will change. If a DC voltage is applied the plates will move and stay in their new position with a new capacitance value. However, if an AC voltage is applied the plates will move in sympathy with the voltage and the capacitance will change over the course of the AC cycle. That modulation of capacitance over an AC cycle causes distortion.
Now it isn't difficult to image that a capacitor that is made of very thin material and is wound very loosely into a cylinder is able to expand and contract as AC is applied to it, whereas a capacitor which is made of thicker material and tightly wound and encased in epoxy resin has its tendency to expand and contract restricted. Which capacitor would you expect to suffer less capacitance modulation over an AC cycle? Therefore which capacitor would have less distortion?
Anyway, that's speculation to some extent, but I hope you can see the scientific logic in the argument. Maybe it explains why some capacitors sound different from others in loudspeaker crossovers.
Or maybe it doesn't.
Last edited:
