Paralleling class 1 & class 2 MLCCs for audio?

[Zoios]

I routinely use J-Fets in audio switching in DC coupled circuits. Why do you need coupling capacitors?

I would need to see what circuit you are trying to realise, to be specific.

Attached is the relevant part of the circuit. A trim adjust in the form of a baxandall volume control and the output of my EQ block below it, connected to a JFET soft changeover switch in most part taken from Douglas Self's design. The switch essentially bypasses the EQ block at the push of a button. The reason for DC blocking caps on both outputs is that they are also routed to other parts of the circuit, most notably the CUE bus (also after some switching, CMOS in that case).

To answer the question above, he mentions the necessity to use DC blocking caps as even small amounts of DC can cause switching transients. My EQ design does introduce some DC at its output, which (in my belief) needs to be removed before routing it to the switch.

I always try to reduce the number of coupling caps in the signal chain.

I tried that approach leaving me with capacitors only directly before switching circuits and in the summing stage of the mixer. When building my first prototype of this circuit I used the combination of X7R and C0G caps (the doubts of their usefulness causing me to make this thread) for DC blocking, which sounded better when compared against some DJ mixers (XONE:92, XONE:96, Pioneer DJM-V10) both when analysed in a DAW and in plain hearing AB tests.

My original question was answered (thank you guys!), yet now I'm very curious to try out Thor's approach of using smaller C0G capacitors with higher resistance FB paths to further improve noise performance

 

[thor.zmt]

I'm very curious to try out Thor's approach of using smaller C0G capacitors with higher resistance FB paths to further improve noise performance

You misunderstood what I wrote.

What I am suggesting is to separate DC feedback loops and AC feedback loops, so it becomes possible to use low value, high quality capacitors to block DC.

I demonstrated in my quick approximations that doing so does not materially degrade the noise performance of modern FET/JFET Op-Amp's.

So noise performance remains determined by the AC feedback components and the Op-Amp's.

Your circuit as shown looks like it inverts polarity with EQ and has correct polarity without. That is a bad idea. Of course the actual EQ part is not shown.

Most of the DC blocking capacitors are pointless and should not be needed, nor are the circuit values chosen well. The circuits themselves also contain many possible improvements.

Try not to copy D. Self.

He is a source for great references on on how stuff works and measures. Great learning material.

And a great source of specific circuits and implementations that are best avoided.

Old British consoles from the more well reputed makers of the 80's and 90's are a good source of inspiration.

Thor