What's the goal of small coupling caps

[jhaas]

I have a couple questions referencing the attached partial schematic, which happens to be V1 of the Trainwreck Rocket, but this is more of a general question, so I'm posting here...

two questions:
1. In this stage we have a 500pF coupling cap in series with a 100pF bright cap. In this case, I'm assuming the overall capacitance is measured as with any capacitors in series (1/C1 + 1/C2 = 1/Ctotal), giving us 83pF. Is that correct, or does that fact that C2 is bypassing a resistance have an effect on the overall capacitance?

2. What is the design goal of very small coupling caps like this. I know reducing a coupling cap value will reduce lows, but 83pF, or 500pF is significantly less than the common .022uF coupling cap, which makes me wonder what the goal is.

A better understanding of this will help me fine tune my latest build.

Regards,

-John

[Andy Le Blanc]

You know how most fenders start to fart out when you crank them.
The smaller the coupling cap the less it farts.

So you restrict the frequency response with a small cap and it also focuses
the amps response to a range and wont waste current in the PS on the lows
and will present more strongly in frequency's above the bass roll off that is
determined by the cap size. It cleans up the high gain distortion, tube bean-o

The brite cap simply compensates for treble attenuation at low volume.
turn it all the way up and the cap isn't really there.

[jhaas]

Thanks Andy,

That is as I suspected, and I've gone from .022uF to .01uF to tighten things up, but 500pF, well, for like-units, 22nF to 0.5nF is such a drastic difference, I wondered if there was something else going on.

You mention the high-gain tube bean-o and that makes sense. I've associated this kind of arrangement with high gain designs, so it caught my eye when I saw it in the Rocket/DC-30 schematic, which is not a high-gain amp.

-John

[Andy Le Blanc]

It is also associated with a vox "brilliant" or bright channel.

[jhaas]

OK, I'm still not clear about the math side of it - calculating the total capacitance of series capacitors, when one is bypassing a resistor (or if that matters).

Here is a partial schematic of the amp I'm currently fine tuning. This time the bright cap is on a voltage divider, not a pot, so it's always in the circuit.

In this case, is the net coupling cap value .005uF?

[iknowjohnny]

I went to a .0047 from a .022 and it helped remove a lot of the fartiness in the lows. I tried many sizes, but what REALLY helped and in fact fixed it 100% was a 500PF ! Unfortunately i could not get the low end back that i lost. What i came to realize trying smaller couplers is that there never seems to be a perfect size, but going small enough to be problematic as far as bass loss works great otherwise. If i could somehow add the bass back in after that i would be using a 500pf right now. But apparently you can add something back thats no longer there. So i stuck with a .0047 because it seemed to be the best compromise.

[tubeswell]

The coupling cap combines with the output impedance of the stage in front and the input impedance of the stage behind to shape the frequency. Lots of things can affect output impedance and input impedance including the type of tube(s) you are using - so you see a 500pF going into a paralleled 12AT7 reverb driver, but a 0.022uF going into a single 12AX7 stage.

[Gee]

1. In this stage we have a 500pF coupling cap in series with a 100pF bright cap. In this case, I'm assuming the overall capacitance is measured as with any capacitors in series (1/C1 + 1/C2 = 1/Ctotal), giving us 83pF. Is that correct, or does that fact that C2 is bypassing a resistance have an effect on the overall capacitance?

C2 and C3 are not in series because of the volume pot between C2 and C3. So what you have is a 500pF coupling cap followed by a volume pot with has a bright cap (C3) across the top of the pot. The purpose of the bright cap is to allow high frequencies to pass without being attenuated by the volume pot. Also when the volume pot is up full, the bright cap is effectively not in the circuit but the coupling cap is still 500pF.