Cathode Follower Voltages

[hylaphone]

Hi gang,
I am messing around with a standalone 6SL7 preamp, with a 12au7 cathode follower similar to V3 in the Ampeg VT40.

It sounds awesome as-is, but is clearly exceeding the Heater to Cathode rating of the tube (heaters at ground, cathode at 180v)
This is a ship-in-a-bottle type of project, and I don't really have space to make a voltage divider for elevated heaters, are there any changes I can make to the CF circuit to lower that cathode voltage and still keep the output impedance low enough to drive a 10k SS power amp?
Would adding a small anode load be a bad idea?

[pdf64]

Yes, drop the plate voltage with a resistor.
It should reduce plate dissipation too.
The main constraint on it being that the cathode voltage must be sufficient to support the max signal voltage, so you should have plenty of scope.
Why the 10k resistor in series with the signal output?
I'm sure that there's still plenty of headroom but it seems wasteful to dump half the signal there.
Pete

[hylaphone]

Cool, I'll tinker around with that. Thanks for the tips.
First time messing with a CF, I just grafted that circuit from the VT40.
I'll yank the 10k too.

Would it be bananas to set this up like a concertina with equal Rk and Ra, but with output only taken from the cathode?

[GainFreak]

To avoid the high cathode to heater voltage you can also make the CF look like a cathodyne PI or take a look at ENGL Fireball schematic to see how it's done there. Check out also the Framus Cobra schematic.
Currently your CF has and output impedance of ~9k, so I would also suggest to dump the 10k resistor.

[g1]

To take a different tack, wouldn't elevating the heaters be a viable option?

[g1]

On the other hand, it has never been a concern in the Ampeg's (225V on cathode) so do you really need to worry about it?
My only experience of preamp tube failures in cathode follower applications is with specific types of 12AX7 in Marshall CF circuits.
If others here have seen CF failures elsewhere, please mention them.

[ReadyTeddy]

You can make the cathode voltage any value you want by properly selecting the cathode resistor. I would start by picking the tube current you want, and the grid voltage you want. Then calculate the lower cathode resistor (V/I). The resistor will be smaller than 47k, but the only thing that will mean is that the gain of the follower will be less than one by a bit more. Now calculate the resistor value that goes from the cathode to the spot at which the grid connects (Vgk/I), where Vgk is the voltage between the grid and cathode--tweaking that value will allow you to move the cathode voltage up or down. You could also set the grid voltage by a resistor divider from the plate voltage--there's always more than one way to get the job done.

[Rhodesplyr]

If others here have seen CF failures elsewhere, please mention them.

I've seen 7199s die of triode section heater-cathode shorts in Dynaco ST-70s where the circuit exceeds the tube's datasheet specs (if you read the fine print). It's a common mod to elevate the heaters to ~+50V or more to extend 7199 life.

There are a few tube amps that use positively biased heaters on one or more tubes to avoid h-k voltage problems. The Marantz 9 comes to mind. It uses a 6CG7 as a cathode-follower driving the output tubes, and there's a dedicated heater winding on the PT just for this one tube.

[Tom Phillips]

...This is a ship-in-a-bottle type of project, and I don't really have space to make a voltage divider for elevated heaters, are there any changes I can make to the CF circuit to lower that cathode voltage and still keep the output impedance low enough to drive a 10k SS power amp?...

I know you didn't ask for this but I'll offer the idea of using a MOSFET to build your follower. Especially since you are space limited. If not used for your current project maybe the idea will be useful for another project. You can read about it in the MOSFET Follies article at http://www.geofex.com/Article_Folder...osfetfolly.htm
Cheers,
Tom

[loudthud]

If there is a negative rail like in an SVT (-150V?), you could use that for the bottom side of the cathode resistor so as to lower the grid and cathode voltages as low as zero volts.

[hylaphone]

Dudes, thanks for the insight. This has been illuminating.
Earlier on in this build I toyed around with a Mosfet, but decided to tackle the classic CF first.
Much more fun, for me at least.

Santa's elves managed to shoehorn a few things, now the filaments are referenced to +50v and the 12AU7 is running within spec.
Thanks again!