It is often tempting to allow shunt impedance to stray an order of magnitude below the circuit impedance. It is often done unintentionally. Damping can become dominated by component parasitics. Capacitors can sound different for the wrong reasons.
Well, yes I suppose. It could be argued that falling back on capacitor ESR to set the lower impedance (resistance/damping) limit on a shunt band-stop filter for example amounts to inelegant circuit design. Some inductor DCR or a small value of resistor may make more sense as the limiting element.
In the topic context this would expose subjective evaluations to criticism. I don't buy that shunt components are less 'in' the circuit than series. It depends.
No, they are not less "in" the circuit, you are correct. All these differences are simply the result of builders who do such a poor job of designing a circuit that the only reason we hear any difference is underlying crappyness of the design. Let the beating of the dead horse begin.
So are you saying that having a cap in a shunt circuit without an also-in-shunt series coil and its DCR or that of just a resistor is not proper electrical circuit design? I fail to see the issue here.
The only way that can be a problem is if a tank circuit is used and the ultrasonics see a short-circuit path that causes amp instability in the ultrasonic range.
Amp.If you mean a straight second order electrical filter with only series coil and shunt cap as being 'inelegant', you have to consider that the reactance of the coil increases impedance with increasing frequency to where nil ESR on said shunt cap is not going to be a problem in the range where it starts to short out higher frequencies. This is the same reason a bottomless trap or LC shunt is an okay thing to do- the series lowpass coil's reactance eliminates the short-circuiting impedance issues.
This is of course all relying on the fact that the shunting components are far enough into the rolloff for the coil to counteract the impedance issue. ie- optimal design considerations.
Later,
Wolf
No, it can be done. Consider though, that a series/shunt combination each part having an impedance an order of magnitude from the load would offer 40dB of influence which is quite useful with speakers. In some circuits requiring further attenuation a second filter is indicated.
Do you have any concerns as to the linearity of capacitor ESR? Inductor resistance should be as linear as that of a resistor, but as with ESR is likely to be shrouded with distributed reactance. As you say, this may not be an issue if it is properly considered.
The context here is that certain things are being taken for granted. In some circuit positions, ESR is the factor that might cause one capacitor to appear to sound different to another. In some circuits it might be the presence of DC. Sometimes it may come down to microphonics, or the dielectric.
Of course ESR can change amplitude levels of the signal, and reducing the resistance in the circuit elsewhere can make up for the loss. However, I don't believe that is the only cause for the difference in sound, nor does it actually bother me. I purchase parts off the shelf as they are for the reasons I've come to expect, and making the ESR the same in caps will likely not make them all sound the same. You have to adjust levels for components and drivers in a system as a whole.
I don't really believe the presence of DC on a speaker would be a cap's fault, and is likely due to poorly designed or operating amplifiers with high DC offset. I don't really build electronics, so I have minimal experience with part substitutions in those applications.
Obviously some caps sing on their own. Usually this is when there is no speaker termination on the xover, IME. IE- only xover hooked up to source.
I don't really sweat ESR differences, nor do I sweat DCR linearity variations. I just balance the system tonally, and shoot for flat FR. I've only had to change attenuation differences a couple times in a speaker to return the sound they had previously before the change of parts. I always factor in DCR in a sim, but ESR- not so much. There isn't always a place to add that integer.
I'm usually VERY close from sim alone, and have not had to adjust several designs away from sim after the fact. This is even when using the uber-parts.
Suffice to say- parts do sound different, and there's not much I can do about it. I just use what I want to use, and change by voicing when it is necessary.
Later,
Wolf
This is some interesting stuff. I wonder if anyone has done any research into the possibility of a capacitor ringing, or storing energy electrically. Not acoustically, or from vibrations. But in the same way that we judge a driver by the waterfall plot, I wonder if caps could be measured this way? I know I have read claims that capacitor X or Y has been optimized to prevent similar effects.
Like, let's say a cap passes a perfect cycle of a 1kHz sine wave. Is the second and 1,000,000 also going to be perfect? Is there going to be any additive effects so that say the 1,000th cycle now looks differently? If I do a single 1kHz cycle followed by a 5kHz, is the 5kHz ever affected?
Sadly I do not have the experience, or tooling to run these tests, but that's one area where I'd like to see real testing, as opposed to theory. Maybe there is never a difference. Maybe it only happens with blue and black capacitors... 🙂 but it's fun to think about.
Best,
Erik
Like, let's say a cap passes a perfect cycle of a 1kHz sine wave. Is the second and 1,000,000 also going to be perfect? Is there going to be any additive effects so that say the 1,000th cycle now looks differently? If I do a single 1kHz cycle followed by a 5kHz, is the 5kHz ever affected?
Sadly I do not have the experience, or tooling to run these tests, but that's one area where I'd like to see real testing, as opposed to theory. Maybe there is never a difference. Maybe it only happens with blue and black capacitors... 🙂 but it's fun to think about.
Best,
Erik
Yes, I know what it should do. I'm wondering if anyone had tested temporal effects of signals on actual capacitors.
This is why I wanted to know if anyone had done this type of testing. To say "I tried it and it's bs" or not. I mean, for all I know, capacitors heat up and then act differently, or they exhibit this but only under conditions where (some condition like excessive internal inductance, oxidation of the outer layers converting the signal to an acoustic wave...). Nature is cagey. You think you know everything about it until you don't.
"Science" without testing is no longer science, it's opinion.
This is why I wanted to know if anyone had done this type of testing. To say "I tried it and it's bs" or not. I mean, for all I know, capacitors heat up and then act differently, or they exhibit this but only under conditions where (some condition like excessive internal inductance, oxidation of the outer layers converting the signal to an acoustic wave...). Nature is cagey. You think you know everything about it until you don't.
"Science" without testing is no longer science, it's opinion.
Last edited:
For instance, furthering the idea of science needing test and research, wouldn't it be really great if we did find out if and why capacitors sound differently? We could measure that quality, and then model it when we build speakers. There could be a table, or a formula we can convert to a spreadsheet or code into our software. Then dial in "liquid treble" or "darker midrange" as desired.
This is what I want, instead of having to listen to a lot of marketting bs and the opinion of drunken audiophiles (like me). This is why people turning to existing theory as if it's all that could be known without bothering to consider future or maybe existing research as exciting.
Those who lack any sense of excitement at the idea of discovering something new or considering new ideas (even if skeptically) are dead inside, say I. 🙂
This is what I want, instead of having to listen to a lot of marketting bs and the opinion of drunken audiophiles (like me). This is why people turning to existing theory as if it's all that could be known without bothering to consider future or maybe existing research as exciting.
Those who lack any sense of excitement at the idea of discovering something new or considering new ideas (even if skeptically) are dead inside, say I. 🙂
- Status
- Not open for further replies.