DC Elevated Heaters Vref?

[UltraHookedOnPhonix]

In Kevin O'Connor's TUT book (chapter 7 on effects loops) there's a fixed-bias cathode follower driving the send signal of the FX loop.

My immediate though was to use the 43.5V reference point as a "freebie" to apply DC stand-off heaters in order to avoid exceeding the tube's Vhk(max) rating and reduce hum. Is this a good idea or is there no such thing as a free lunch?

[martin manning]

I think yes, you could use it. However I wouldn't be too worried about a 43V Vh-k. Most 12AX7's are spec'ed at least to 100V Vh-k max.

[John_P_WI]

Usually the elevated heaters are taken off the Vscreen cap node using a couple of resistors and a cap. The greatest benefit is hum reduction. In the drawing posted the tube grid is isolated by the ratio of the voltage divider created by the 1M and paralled 33k2 rground and 226k r to the B+. This ratio will dictate the amount of heater hum imposed to the tube grid or the amount of signal injected onto the heaters if you were to tap off at that point.

Honestly, pull it off the low impedance point of Vs and avoid the tube grid.

[martin manning]

The 33k2 has a 22u cap across it so the AC impedance to ground at the top of the 33k2 is pretty low. That cap could be increased if necessary, but again I don't think there is any need for an elevated heater circuit here.

[John_P_WI]

Point taken about the cap, reviewing and posting on a phone has its limitations...

If I could guess, UHOP is interested in using the above to generate a vref not necessarily to inject a heater vref at that point. Maybe chasing hum on a Dlator?

[UltraHookedOnPhonix]

Thanks for the input guys!

Martin, I'm thinking about the 12AT7 which has a Vkh(max) of 90V.

Honestly, pull it off the low impedance point of Vs and avoid the tube grid.

I was hoping to consolidate the voltage divider for the fixed-bias 12AT7 cathode follower stage and the heater elevation point to save some parts, avoiding having to set up two different nodes with roughly the same reference voltage.

If I could guess, UHOP is interested in using the above to generate a vref not necessarily to inject a heater vref at that point. Maybe chasing hum on a Dlator?

I'm interested in generating a Vref for the voltage-referenced 1M grid resistor and to DC elevate the heaters using that same voltage.

Actually, I'm designing a built-in loop for something Dumble-ish. Seeing as the Dlator has around 340V on the recovery stage, I could feed the CF driver and recov amp from the same PS node. Again, to save parts and streamline the design.

Here's a drawing of the concept. Loop string is in parallel with the screen node. PS is of the "precision" style.