A question about the 'gimmick' capacitor on my SF champ

[eschertron]

(NOTE: since the 'gimmick' cap under discussion is not indicated on my schematic, I did not link a schematic.)

I've been looking over the CBS-era champ that I recently acquired, and one thing that I saw - that others have already talked about here on the boards - is the 330pF cap across the grid to cathode on the output 6V6. I understand that it is a roll-off filter cap to help avoid oscillations. While poking around the 6V6, I also noticed (The part that has me scratching my head):

There is a 'gimmick' of a few twists of the lead wire to the 6V6's grid with the HT wire for the preamp tubes. OK, so the small couple of pF capacitance (multiplied by mu, giving a slightly less small capacitance) will also act as to roll off high frequencies. Why did Fender include this when the cap across grid to cathode exists?

Does the reactance between the HT and the power tube grid add some kind of noise-cancelling effect? I'm thinking that if the B+ has a spike on it, then the capacitance will cause the grid to be tugged slightly in the same direction, and the effect on the plate tends to cancel out the excursion. Is there any merit to this? How far off base am I, and if there are discussions or resources that deal with this I'd love to see them (haven't dug anything up on it yet!).

[Enzo]

A photo would be great.

A gimmick would be a couple short pieces of wire twisted, but they don;t go anywhere. If you are talking about two wires that go from place to place, and are twisted together, I wouldn;t call that a gimmick.

Rolling off high end with a cap wired like the guitar tone control is one way to squash oscillations, but there are others. Two wires mighy be aligned to couple, so that an out of phase signal is fed back into a stage. Just like the big NFB loop in the power amp.

An amp can have more than one point of instability, and addressing one doesn't necessarily help the other. Like more bandages on your arm won;t make your leg feel better.


And sometimes wires are twisted together simply because it make the amp neater, and is cheaper than wire ties.

[eschertron]

Enzo, thanks for the reply! here's the picture, and a relevant schem:



The red wire goes from the last PS node to the 100K plate resistors for V1. The white wire goes from the coupling cap / grid resistor to the grid of the 6V6. I'm inclined to agree about your point on 'neatness'. I'm also becoming aware of how the layout (or layout in general) might be arranged to save an eyelet here or there. I'm way too literal-minded to foresee some of the solutions I've come across in these builds.

Question: the grid resistor for the 6V6 is grounded with the cathode circuitry at the bottom of stage two (where the NFB returns). I would have thought to join it with the cathode grounds at the 6V6 itself, at the expense of an extra eyelet and some wire. Am I over-thinking this?

[Gingertube]

OK - the red wire is B+ and so is at AC ground - you get a little shield effect and a little capacitance by twisting the grid feed with it like this. Neither effect is significant (thet is, crap shielding and a few pF)
The added grid to cathode cap is I guess for stability, you need a dominant pole (high frequency roll off) somewhere in the forward path, inside that feedback loop.
It bothers me that there is no grid stop on the 6V6.
Cheers,
Ian