At some point, Tony, you're going to have to do either a quickie Excel spreadsheet, or, a Bode plot demonstrating what you're convinced "is". The capacitor drops Z (the A/C signal impedance) to a remarkably low value - dependent as you say on frequency - which over the band-of-frequencies that we're interested in (20Hz to 20kHz, with skirts on both extremes) has an effective resistance so low, as to emulate a constant VOLTAGE source. That's its purpose. There are many ways to achieve this too ... but having the front-end line stages' cathodes given a positive voltage bias that self-adjusts to the age of the tube, the manufacturer, the era, the variances in manufacturing ... is a damned good thing. Especially for "consumer electronics" - where the very most one can expect the consumer to do is occasionally take all the tubes to a friend for testing and replacement for the "weak" ones.
The more fixed, more absolute constant(-ish) voltage sources are nowhere near as accomodating to tube aging, lot variance, manufacturer variance and "era" (1950s, 1960s, 1970s, Chinese modern) variance. This is why I continue to advocate unbypassed resistors (I simply prefer the cleaner, though admittedly much lower gain signal-tracking action of an unbiased resistor), along with conventional plate-load resistors to develop the gain multiplier. It works, its simple, and as I said ... I like it.
GoatGuy
The more fixed, more absolute constant(-ish) voltage sources are nowhere near as accomodating to tube aging, lot variance, manufacturer variance and "era" (1950s, 1960s, 1970s, Chinese modern) variance. This is why I continue to advocate unbypassed resistors (I simply prefer the cleaner, though admittedly much lower gain signal-tracking action of an unbiased resistor), along with conventional plate-load resistors to develop the gain multiplier. It works, its simple, and as I said ... I like it.
GoatGuy
That is simply not true.
Different capacitors have different sound signatures. Some are natural, some are boring, or harsh, or dry, with more or less microdetail, etc.
There are hundreds of people around who can hear this, and in general their opinions about a specific capacitor converge.
Of course, you need a good set of ears, a good music source, an excellent amp and transducer to hear this kind of differences.
why listen to the sound of caps? why not just listen to the music instead?
Because some capacitors sound better than others, in my system, with my ears and my speakers/headphones.
Sorry, but I don't build amps that "just work". If that's what you do, ok, it's your business.
I, for one, want the best musical experience out of my amps, and I do everything I can to get it.
You miss the point. In a properly designed circuit with the value of coupling capacitor properly selected there is no audible difference. I am not saying there is no audible difference. I am saying that if you hear a difference it is due to poor design. I say again, publish the circuit of this wonderful system of yours that allows you to detect the sound of different coupling capacitors.
Cheers
Ian
Quite to the contrary: when your whole audio system (from data bits or vinyl grooves to ear drums) is good enough, the individual sound of the different components becomes sensible.
I'm not here to convince you I'm right. I said what I had to say. Now, gentlemen, I bid you adieu and go play elsewhere.
This formula is often quoted, but quite wrong! It may explain why some people think they can hear capacitors; they have used the wrong value.GoatGuy said:
The R in the formula is not the cathode resistor, but the parallel combination of the cathode resistor with 1/gm - using gm at your chosen operating point not merely the one quoted in the valve datasheet.
In some cases what people may be hearing is a combination of wrong (usually too low) value combined with wide tolerance of electrolytics. The result is inter-channel phase shift at low frequencies; the capacitors take the blame when all they are doing is what it says on the can: 100uF +-20% when maybe you actually need 220uF +-5% or 1000uF +-20%.
On the one hand you ask me to accept that some people can hear differences yet you refuse to provide any evidence other than your opinion. Not surprisingly, I remain unconvinced.
Cheers
Ian
Or that a circuit is competently designed such that the AC voltage across the capacitor is minimized. Granted, there's a lot of incompetent tube design out there, especially among hobbyists.
I designed out the cathode bypass cap and the coupling caps because I could hear them. DC coupled with self adjusting fixed bias.
When I did use cathode bypass caps I always went significantly over spec in uf to avoid phase issues.
My experience is that when people are talking about differences in the sound of caps, it is almost exclusively a hi-frequency issue rather than an inadequate sizing issue.
We can all talk about poor design, but ultimately I have been through the alternatives and chosen a different path.
Shoog
Last edited:
If I want an engineers amp, I will buy a chip amp where I know the specs will beat anything that a valve design can offer. Did that and moved onto better things.
I don't buy the argument that if it measures the same it sounds the same - or we would all be using the same op-amps and none of us would be building valve designs.
Shoog
Speaking as a musician as well as an engineer, there is alot about sound that is beyond theory and measurement. There are low level effect to the sound other than frequency response, phase, and distorsion. We use simple models of capacitors, resistors, inductors, transformers, etc. to engineer an amplifier, however if we build identical circuits with different components having the same top level (i.e., first order) performance characteristics, we get a different sound quality. Realizing that some may deny there would be a difference, many of us readily perceive a different sound as a result. Components are not ideal, a capacitor has resistive and inductive characteristics. They are also microphonic and non-linear in the micro sense (as distinct from the "macro" non-linearity given by its application in a circuit). Different dielectrics, film thickness, electrode material (foil, metallized), geometry, winding tightness, lead material and attachment, and other such manufacturing and construction methods means that two different capacitors having the same value are not physically identical. Are we to believe that these have no influence on the sound? Now speaking as a musician, you can have two musical instruments that by all appearences, are physically identical, and yet they have a different response and sound quality when played. Is that just the imagination of the player? There are very low level characteristics in musical instruments that are difficult to quantify, and take artistry and not neccessarily engineering to duplicate. Likewise, when you play back a recording made of instruments, can you identify the difference in the sound characteristic between the two instruments? I maintain that these subtle low level effects are real (because I notice them when playing), however are likely immeasurable by conventional methods. A listener should be able to hear these, given an ear that has the sensitivity for such differences. Now back to capacitors or any other circuit component. I maintain that if you have a physical difference between two real-world components that are otherwise identical in their top-level characteristics (i.e value), that the performance of these components at the micro level will be different. They will exhibit subtle differences that may or may not be audible, but they will be there because the are an artifact of how the component is designed and constructed. I cannot just catagorically deny that they are not audible, even if I cannot hear them, who am I to say that someone else who is sensitive to such low level differences is just deluding themselves. I would turn it around and say it is the deniers who are deluding themselves into thinking that just because they cannot hear something, it does not exist.
You get a little bit of local NFB which may reduce distortion but this is more than offset in most cases by the increase in ra which leads to an increase in distortion. Perhaps its this extra distortion that some people think sounds better.
Cheers
Ian
At the risk of repeating myself, any amplifier that can be made to sound different by changing the make of capacitor is just designed badly. Any designer worth his salt will be able to make a circuit where these subtle effects are inaudible. The only reason there is so much said about the sound of components is because there are so many incredibly bad designs out there. It is no wonder people can hear the differences.
Cheers
Ian
Not buying that argument at all sorry.
What we are talking about here is people (large numbers of them) reporting differences in sounds with changes of capacitors and a set of consistent results across listeners about what they say sound best. The differences are often very small and marginally important overall - and frequently they just fade into the background, but they exist.
I have heard it myself. I am not obsessive however and have never bought anything more fancy than Philips mustard blocks which have served me well in most application where I have used them. But to deny that people hear differences is arrogance as far as I am concerned.
Shoog
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.