Differing Capacitor Characteristics.

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

jimmy123

New member
Joined
Feb 23, 2009
Messages
2
Hi guys this is my 1st post so I hope i'm not asking an obvious question...!

I'm looking for advice from clone builders out there. I was reading that different types of capacitors work better in certain applications (ceramic for filtering and bypass, polystyrene for timing, tuning circuits etc). I am looking into building the GSSL and doing some research and before I jump into buying parts it would be good to know whether I should pay close attention to this or just choose one type of capacitor (obviously with differing values) and run with it.

Any help is appreciated. 

Thanks,

JN
 
While the complete answer is a chapter in my book, simply consider how the capacitor is being used.

If the capacitor is in an audio filter that has poles at audible frequencies, there will be changing terminal voltage across that capacitor so voltage coefficient and other terminal voltage related nonlinearities will matter.  If it is a DC blocking capacitor set for a low enough pole frequency that terminal voltage doesn't change with audio, the primary concern is non-ideal series impedance. So beside the pole frequency the termination impedance or loading will influence the significance of ESL and ESR. To wit a blocking capacitor in a loudspeaker path is a different animal that blocking DC out of a 100k pot.

etc.....

JR
 
It's unfortunately one of those questions that has many answers depending who you ask and what time of day.  Most of those answers are already on the forum and I'd go search if you want to know all of the details.

But the short answer is to use quality electrolytics for both DC blocking(AC coupling) as well as supply rail decoupling.  Audio frequency devices don't need overly crazy schemes of capacitor usage(such as bypassing with film caps, etc) like a lot of the audiophile crowd will lead you to believe.

If you really want to bake your noodle, go look up Dr. Howard Johnson's research on decoupling and signal integrity.  Most of his work centers around high speed data but physics is still physics and a lot of it applies to low speed audio too.
 
Probably never... I'm not smart enough to write the whole thing...

What I wanted to write (or edit) was the book I always wanted to buy.

Start with the chapter on resistors,,  Begins with very rudimentary what a resistor is , maybe some basic ohms law.. then progresses into more detail and esoterica.. types of resistors and why, tolerances, power dissipation, thermal noise, excess noise, atomic effects (?)... whatever.

A beginner or novice, reads only as deep into the resistor chapter as comfortable then jumps ahead to the next chapter on capacitors or whatever. Later, or when the need arises to understand more about a topic you return to the chapter to dig deeper. If you don't understand, you back up a few pages and there should be enough foundation to lead you in.

Perhaps a web version with hyperlinks to jump around, but my concept is a reference for self learning that can work from multiple levels of background or understanding. The discussion starts simple and gets progressively more difficult. As much as you can bear, or need to know.

Most books stop too soon, or start in the deep end on topics. Most of the really esoteric stuff is not even assembled in this kind of organization, since the esoteric capacitor section needs to written by a different expert than the esoteric resistor section. Only the style and presentation needs to be managed so the reader can jump in and follow.

Might be a good idea to get say application engineers from a resistor company, and cap company, and whatever to participate.

but don't hold your breath from me... Between wiki and the web this may almost morph into reality. All that is missing is an organizing force or intelligence (aka technical editor). With a web or hyperlink based media (Kindel?) instead of footnotes referencing more detail, you just hyperlink them to the references.

JR

 
JohnRoberts said:
A beginner or novice, reads only as deep into the resistor chapter as comfortable then jumps ahead to the next chapter on capacitors or whatever. Later, or when the need arises to understand more about a topic you return to the chapter to dig deeper. If you don't understand, you back up a few pages and there should be enough foundation to lead you in.

That's a brilliant publishing model - for tech related subjects especially. I've never seen it done but I'm sold on the idea. Hopefully you can find an enthusiastic co-author or editor with the technology to make it happen.
 

Latest posts

Back
Top