AC coupling between driver tube and output tube?

[theGopher]

This may be a silly question as I'm a relative newbie to OTL tube amps, but why not AC couple the plate of the input tube to the grid of the output tube? That would allow each tube's operating point to be optimised separately wouldn't it?

Was this considered but rejected for Crack?

[Paul Birkeland]

This is certainly something that's doable, but it adds six parts to the amplifier and introduces a coupling cap that doesn't need to be there.  With the Speedball installed, the DC current through each tube is locked in, so then there is even less purpose to adding these parts.

[theGopher]

Thanks for the quick reply Paul.

Would AC coupling make rolling input tubes easier if the output tube grid voltage could be set independently of the input tube plate voltage?

Why 6 components? Isn't it just a capacitor and a resistor to ground for each channel?

[Paul Birkeland]

Would AC coupling make rolling input tubes easier if the output tube grid voltage could be set independently of the input tube plate voltage?

It's also important that the operating point for a given driver tube makes sense in other ways.  For example, if you wanted to use a 12AX7 in the Crack (a bad idea overall), you would have to change the LED out for a cathode bias resistor and you would basically have to use the Speedball.  These constraints have nothing to do with the 6080.  For some tubes that are a little off, it might give you a tiny bit more flexibility, but not worth it IMO.

Why 6 components? Isn't it just a capacitor and a resistor to ground for each channel?

2 x coupling capacitors
2 x grid leak resistors
2 x extra cathode bias resistors

I attached an image from one of the old RCA manuals that shows what this looks like.  C2, R6, and R4 are not present in the Crack design.

[Tom-s]

For example, if you wanted to use a 12AX7 in the Crack (a bad idea overall), you would have to change the LED out for a cathode bias resistor and you would basically have to use the Speedball.

Isn't the speedball set for a way higher current 8ma or so? vs the 1-2ma needed for 12ax7? Or would this be just fine?

[Paul Birkeland]

It's not what current the C4S board is set for in the Speedball, it's that there's not enough voltage available to use an adequately large plate loading resistor.