[not again ?]bypass cap in singal circuit

[sagitis]

I searched the forum in regards of quality and brand of bypass capacitor. Eveyone discuss the affect it did in PSU and speaker only but none of them are talking about how the brand of a bypass capacitor did affect when installing in the signal circuit.

Anyone mind share their opinion, I don't want to waste too much money on this issue.

The pre amp I am going to build is this
http://www.diyaudio.com/forums/blogs/rjm/1324-crystalfet-phono-stage.html
and I mainly concern in C9 - C20
http://www.diyaudio.com/forums/anal...lifier-development-thread-15.html#post4689009

 

[DF96]

My opinion is that the best way to avoid wasting money on bypass caps is to avoid wasting money on bypass caps! That is, don't use bypass caps. It is rare that they do anything useful in audio circuitry, but they can sometimes create problems of unwanted parallel resonances. If you must use them (e.g. in order to follow fashion) then make sure that either they or the electrolytic they are bypassing is sufficiently normal and cheap that there is enough resistive damping to kill off any unwanted resonances.

 

[Enzo]

I may be wrong, but I think he is referring to the whole row of filter caps in the power supply rails as "bypass caps." The term means something else to me, but that is how I read it. One doesn't want to waste money, but one probably wants to retain supply filter caps.

 

[DF96]

When he said

and I mainly concern in C9 - C20

I assumed he meant the two 0.1uF bypass caps C19 and C20.

 

[adason]

He should spend his money on c2, c3 and c4...
bypass caps are just those two as DF96 said, rest are normal filter caps, not much to do with signal, except keeping the hum as low as possible

 

[Enzo]

Ah, when I read it I saw it as 9 through 20, which would be the entire row of eight caps on the print. Your interpretation is probably correct.

 

[sagitis]

Ah, when I read it I saw it as 9 through 20, which would be the entire row of eight caps on the print. Your interpretation is probably correct.

yeah it is those 8 caps with the electrolytics and ceramics ones

 

[DF96]

Most of them are not bypass caps - only C19 and C20 are bypasses. As I said, you can omit them.

The rest are local decoupling capacitors. Any ordinary electrolytics from a reputable dealer will be fine.

 

[Monte McGuire]

Here's another data point to support the "use good electrolytics" suggestion. I've cobbled together a test jig to measure the impedance of a power supply capacitor, and found that most high quality electrolytic caps have a very low impedance, even at somewhat high frequencies. I measured a 470µF 35V Panasonic FR cap using a swept oscillator and a series resistor, and found that, while the cap's self-resonance is somewhere around 4-500kHz, the low impedance region around resonance is broad and wide, unlike a ceramic or film cap's narrow and high-Q resonant region. This cap has a far lower impedance magnitude at 1MHz than any ceramic bypass will have, even though this is well above its self-resonance. Using this jig, it's also instructive to see just how little a typical small bypass cap does when added in parallel to a quality large-ish electrolytic cap.

So, my suggestion is to spend your money on quality electrolytics, and if you're concerned about low impedance at extremely high frequencies, put some small SMD bypasses right near the finicky component, not across the large electrolytic.