AC30 cathode follower Q

[Chuck H]

Thanks Ian. See post #10. I changed the design and am planning to dedicate a 12ax7 to the circuit posted above. If there is a better way to run the tube correctly wrong I don't know about it. And I've never heard that the cathode follwer in AC30's is prone to fail. I'll probably just duplicate the circuit..

[tubeswell]

Thanks. I actually burned some oil last night and I'm pretty happy with the new schem.

Lets be seein' it then.

[Chuck H]

You betcha. But it's only in rough pencil right now. Since I don't need to present this design professionally I probably won't CAD it. But when I have it drawn up nice I'll scan it and post.

[Wombaticus]

Chuck,
I would not expect the same sonic results from using a 12AU7 for the cathode follower.
In the part schematic you posted you can see that the 12AX7 cathode follower is running with zero bias (180V on grid and cathode). 180V across 56K means that the tube is running 3.2mA, that is a hell of a lot for a 12AX7 and to get that current the bias is down to zero. I expect it is drawing a fair amount of grid current from the previous stage and I also expect it to simply not respond at all to positive signal swings on the grid, that is it will pretty much only reproduce the negative going half cycle as it is already saturated. I can see that I'm going to have to dig out the Wiz'es book and have a look a what he says. When ever I see a 12Ax7 running at 3mA I immediately think that its not good design and there should be a better way to achieve the same thing - sort of like a friend of mine with his classic car restorations, when he reaches for the hammer I always say "hang on , there has to be a better way" - 90% of the time I'm right.
Cheers,
Ian

Ian,

I'm virtually certain you are mistaken about the 56K CF biasing out at 0V (and also, therefore, in your conclusion about the current through the stage) Cathode followers are surprisingly forgiving in terms of their ability to self-bias to workable points across a large range of cathode resistances. And sure enough, looking more closely at that schematic, I believe that it says something like 183V, not 180V (although it's admittedly quite indistinct).

The other thing to bear in mind here is that it's not straightforward to accurately measure the grid-ground voltage of a DC coupled CF, as the probe of a standard voltmeter will load down the node sufficiently to meaningfully alter the reading, and I think this has been occasionally the source of significant confusion regarding the bias point of DC coupled CFs. I always take the readings at these nodes on schematics with a grain of salt for this reason. Big differences can be explained away just by whether one is using a relatively low-impedance probe vs. a high impedance VTVM.

Cheers,

Paul