Outside foil questions

[greiswig]

Hi, all,

Couple of basic questions about using the "outside foil to low impedance side" issue:

1. How do you determine which is the low Z side of the cap if you're not a whiz with circuits?

2. Suppose you hook two foil caps in series...what is the "right" way to take advantage of the outside foil as shield? Would you hook them up so that the inside foils are connected together, and the outside foils to the circuit?

Thanks!

[roberto]

Please read this thread. For a series connection, I think the better way is to connect together the inner foils of both. (but I think doesn't mean I'm sure).

[mlp-mx6]

For a series connection, I think the better way is to connect together the inner foils of both. (but I think doesn't mean I'm sure).

Just a dissenting opinion here - but it seems to me that this approach violates the principle. You would have one cap oriented toward the high-impedance end, right? 2 caps in series is the same as 1 smaller cap, so I would think you just put both of them oriented in the same direction, toward the low-impedance side. JMHO, YMMV, etc.

[greiswig]

For a series connection, I think the better way is to connect together the inner foils of both. (but I think doesn't mean I'm sure).

Just a dissenting opinion here - but it seems to me that this approach violates the principle. You would have one cap oriented toward the high-impedance end, right? 2 caps in series is the same as 1 smaller cap, so I would think you just put both of them oriented in the same direction, toward the low-impedance side. JMHO, YMMV, etc.

This is the gist of it for me: I can see both sides of this argument, and both seem plausible. So I'm still puzzled...

[roberto]

My thought is: outer foil has to shield the signal from external noises, so you have to "ground" it in the shortest way you can.
But a cap is not conductive, so if you want to "ground" the outer foil, you have to solder the two in opposite direction. Isn't it?

[klingo]

If you have nothing connected at the common point of your caps in serie, you may connect it to a DC voltage (9v batterie ?). i read it dampen the foils from moving with the AC swing(if no DC on this cap)....this is not answering your foil orientation question...sorry.

[David Root]

For a coupling cap, outer foil goes back towards the plate, inner to the load resistor. Similarly for cathode bypass cap, outer to the cathode, inner to ground.

[roberto]

Well, no. In a cathode bypass cap, the outer foil is to ground, the inner to cathode. Otherwise no shield effect will appen.
But in this thread we are not talking about general cap orientation, but the orientation of a series of two cap.

[Ears]

My thought is: outer foil has to shield the signal from external noises, so you have to "ground" it in the shortest way you can.
But a cap is not conductive, so if you want to "ground" the outer foil, you have to solder the two in opposite direction. Isn't it?

Caps are AC conductive. The outside foil will only serve as an effective shield for extraneous junk if it is connected to AC ground.
No matter how you connect a series pair I cannot see how that beneficial effect will work for both of them. If the inners are connected then one outer goes to AC ground the other to point of higher impedance . When connected inner to outer then one cap is nestled and so only one sheild effect is available. If two outers are connected then no effective shield possible.
Of course I could be quite wrong.

[roberto]

Ears, we are both quite wrong. The shortest way to ground depends on tube, plate resistance, capacitance and load.
I said the shortest way to ground for the second cap is the load because reactance is 1/(2 pi f C). Usually you have 1n to 22n caps. A 22n at 5kHz has a reactance of 1450ohm at 1kHz 7250ohm. 1n has 32kohm and 160kohm.
I have no amps to check this out until next week, otherwise the simplest way is.. try!

[David Root]

I was bassackwards about the cathode bypass cap.

Interestingly, in parallel, if you connect two nonpolar caps with the inside foil to outside on both sides, you get an essentially non-inductive compound cap, also with reduced ESR and impedance. Almost the perfect capacitor. Used in audio circuits but not to my knowledge in MI amps.