Split load cathode?

[lowell]

I'm sure this isn't the correct name for this circuit. Can anyone explain the cathode setup on the preamp stages in the SVT Classic? Specifically stages 2, 3 and 4?

Here's what I think is happening. Please confirm or correct me. The Vgk is nothing new as there's still a 1.5k from cathode to grid leak resistor . I'd guess this is still Vgk of ~1.5vdc like your typical 12ax7 bias setup.

The tube now has added resistance with the other cathode resistor to ground when drawing current so maybe this adds some compression?

Maybe this extra resistor is used to place Va more accurately to B+/2 for cleanest output? I think this is probably the correct answer.

Finally, the grid is now seeing positive feedback. Granted ~50% of Vg on stage 2. Is this enough to be noticed? And how does this not cause oscillations?

Here's a link to the schematic. Preamp is in file 0736...
Amp Archives/Ampeg/Schematics/Ampeg SVT-CL Schematics

[pdf64]

For V1 2nd stage of the 001 scheme, I think that the idea is to increase the value of resistance in the cathode ground return, so that when the partial bypass 'ultra bright' cap is switched in, there will be a bigger boost (ie more difference between bypassed and unbypassed stage gain).
So the stage is biased with a 1k5, whereas for gain purposes, the cathode resistor is 3k.
Re positive feedback, the grid leak is bootstrapped to some degree; the bootstrapping may be frequency dependant when the ultra bright is switched on. As it's a 1M grid leak, I can't see it having much affect.

[Chuck H]

I think the intention of these circuits is less lofty than you propose. I think instead that the cathodes are elevated to provide enough poop for the tone shaping circuits on the cathodes and the coupling to the grids is simply a bias resistor providing somewhat more center biasing with this elevated cathode. Yes there will be more local NFB and this will linearize the amplifier. But I think it's a byproduct and not the primary goal. The reason it doesn't oscillate is that there is no PFB happening. The cathode is out of phase with the plate output.

Stage two and three ARE set up this way. Stage four is just a cathode follower.

EDIT: Pete simulposted and beat me by three minutes. Now I know who I would hire for a morning shift here

[Dave H]

Stage 4 is a cathode follower and as it's AC coupled the 4k7 is there to bias the tube.

Stage 2 is a gain stage with a 15k cathode resistor to reduce its gain and the 1k5 is to set the bias point. The cathode resistors are bypassed by a capacitor to increase gain at high frequencies on the ultra-high setting.

Stage 3 is set up the same as stage 2 but the selected mid frequencies can be either boosted or cut by either increasing the gain of the stage (mid pot CW) or by attenuating its output (mid pot CCW). I think that's how stage 3 works.

[Dave H]

EDIT: Pete simulposted and beat me by three minutes. Now I know who I would hire for a morning shift here

And you both beat me by by more than five minutes. It's usually Enzo who does that.

[lowell]

Oh it's a 15k resistor under the 1.5k biasing resistor? Hard to see with the resolution.

[Chuck H]

I thought it was hard to tell also. I looked at the digit spacing and figured it was PROBABLY 15k also.

[Enzo]

Oh, I'm getting older and slower.

[Dave H]

I'm getting older and slower.

No way!

[Chuck H]

Oh, I'm getting older and slower.

Nah! You're just so experienced now that you don't need to be fast

[Enzo]

I think my wife agrees with you, why just earlier today she said I was half-fast.

[lowell]

So I'd think the tube would be in cut-off with a 16.5k Rk. Or not? Seems like alot. And there'd be little to no gain if plotted on a load line.

[Chuck H]

The WHOLE tube is elevated by the cathode and the grid bias is derived by the split load. There IS, however, additional local NFB which is selectively bypassed by the tone shaping circuit. So there is gain reduction via elevating the tube (because there's less "working voltage" across it) and the local feedback loop, but not necessarily a cold bias condition. Sort of the same way a LTP phase splitter cathode is elevated and with similar gain losses.

I think

[lowell]

Ah right...brain fart. Scuse me. 😉