Cream "Gibson" Color Survey?

[Mike Sulzer]

Ahh. It does not work that way. A rat's-nest of capacitors (no matter how many or how they are connected together) acts as if it were a single capacitor of merged or ~averaged properties.

The above principle has a name, Thevenin's Theorem.

1. What you mean is that impedances can be combined algebraicly so that a network with two external terminals composed of an arbitrary number of impedances can be replaced by a single one. Thevinen's theoreom is this (Wikipedia): "In circuit theory, Thévenin's theorem for linear electrical networks states that any combination of voltage sources, current sources and resistors with two terminals is electrically equivalent to a single voltage source V and a single series resistor R. For single frequency AC systems the theorem can also be applied to general impedances, not just resistors."

In the original definition, Thevinen's theorem applies to voltage sources, not both voltage and current sources and Norton's theorem is the more general one. But neither theorem is required to merely combine impedances. (Each theorem does imply that you can combine impedances, but more is involved when there are sources. The combination of impedances follows from nothing more than the fact that such networks are described by simple linear differential equations.)

2. Especially in tube amplifiers where the ac voltage across a capacitor can be large, one of the key concerns about the sound of capacitors is their lack of perfect linearity. In a situation involving non-linearities, you cannot simply add impedances in general. So even the weaker condition which you mean, combination of impedances, does not apply.

" It does not work that way." It certainly does; how you combine non-linear circuit elements is important.

[Joe Gwinn]

Mike, you have misunderstood the reference of "it does not work that way".

[Mike Sulzer]

Mike, you have misunderstood the reference of "it does not work that way".

Belwar: ".....taint the outcome ..."

My interpretation of Belwar's statement: he was reluctant to combine capacitors because it might confuse the results.

My interpretation your statement ("It does not work that way."): "not to worry" because many capacitors combine to be just like one capacitor.

I said that many capacitors do not combine to be just like one when one or more of them is non-linear.

Example: If you put a non-linear capacitor in parallel with a linear one you tend to reduce the distortion, but it is not a simple averaging of properties. You tend to "short out" the harmonics, but also the level of generated harmonics has changed because you have changed the current through the non-linear capacitor.

What did you actually mean?

[David Schwab]

Belwar: ".....taint the outcome ..."

My interpretation of Belwar's statement: he was reluctant to combine capacitors because it might confuse the results.

I believe Belwar was referring to putting caps in series. You left out part of his quote... the important part.

The original post was:

I have to make a decision though as to whether or allow caps in parallel or caps in series.. Doing both expands my choices for caps.. I figure im going to allow caps in parallel but not in series.

Joe asked why series was a problem.

but in series whatever is in front could taint the outcome

This was about what ever sonic signature a cap might have.

[Mike Sulzer]

I believe Belwar was referring to putting caps in series. You left out part of his quote... the important part.

When Belwar wrote this:

Well i would be worried that there is some effect that would happen in the first capacitor that would happen in the subsequent ones - In parallel, whatever happens would be split between the caps equally but in series whatever is in front could taint the outcome - I don't have an indication of what effect (if any) this would have. Any thoughts? This was/is my only hesitation.

I thought he was asking for any thoughts about his thinking on combining capacitors, something like "am I right in thinking series is a problem but parallel is OK?"

Apparently Joe did too, because he wrote about the general case, and then about both series and parallel:

If you have a series string of capacitors, it makes no difference whatsoever what order the capacitors are in. There is no concept of "in front", as the same current must flow through all.

If you have capacitors in parallel, the overall current is simply the algebraic sum of the currents through the individual capacitors.

[RedHouse]

That's the way I read it too, and the reason I chimed-in with that Neve thing.

[David Schwab]

Apparently Joe did too, because he wrote about the general case, and then about both series and parallel:

Right, but if you string the question and answer together it looks like:

Belwar: "but in series whatever is in front could taint the outcome"

Joe: "it does not work that way"

To me that was the question that was being addressed.

Now we will have to see if the parties in question agree.

[RedHouse]

Ahh, I see.

[Dave Kerr]

To me that was the question that was being addressed.

Now we will have to see if the parties in question agree.

What fun is there in doing that? It's much more interesting to parse out every word of every post and debate like Talmudic scholars.

[David Schwab]

What fun is there in doing that? It's much more interesting to parse out every word of every post and debate like Talmudic scholars.

Well for one I can follow along.. otherwise it becomes confusing.

[Joe Gwinn]

Belwar: ".....taint the outcome ..."

My interpretation of Belwar's statement: he was reluctant to combine capacitors because it might confuse the results

....

What did you actually mean?

Dave Schwab has it right.

Look in message 157, where Belwar worries that the first capacitor in a series string will somehow affect the response of the others, the implication of worrying about the first capacitor being that it matters which capacitor is first. Well, it doesn't work that way - the order does not matter. Nor is linearity or its lack an issue.

The rest is mere pedagogy.

[Mike Sulzer]

Dave Schwab has it right.

The rest is mere pedagogy.

Then I suppose I was commenting on the pedagogy, and why not? It is the majority of your comment. But in any case, I would be happy to hear any comment you might have on that. We seem to be far from the original topic, and I do not think anyone will mind.

[Joe Gwinn]

But then we would become mere pedagogs. Better to quit while we are ahead.

[Dave Kerr]

But then we would become mere pedagogs.

Well, you each seem to not only be able to teach, but to do, so I wouldn't worry about that.

[belwar]

But then we would become mere pedagogs.

and wouldnt be allowed within 300ft of elementary schools...