DC coupled cathode follower bias tomfoolery

[mxrshiver]

a humble lil 12AX7 cathode follower stage tweaked a la the school of Merlin Blencowe that i've built into a modded Showman, but the voltage readings seem very strange. capacitances in purple, resistances in pink, voltages to ground in blue. based on the loadlines this sucker should be too hot if anything, biased around 0V - i figured i could warm up the driver stage's bias w parallel cathode resistors till the cf got to a more reasonable bias.

but... the cathode ended up 11.1V higher than the grid somehow... was not expecting that. here's the funny thing though... it occurred to me to measure voltage directly between grid and cathode, and whaddya know, -0.36V, right where i thought it'd be...

i'd like to make sense of what i'm lookin at, and why the voltage measurements from the cf grid and cathode to ground, respectively, do not match up with the voltage measurement directly between grid and cathode. Merlin's articles & book make no mention of a grid stopper or arc protection changing the bias (that i could find), but it does make sense to me that a 330K grid stopper drops that 16V from the plate of the driver to the cf grid, esp if the cf is pulling grid current. but i don't see why that would make the cf's grid-to-cathode voltage appear so much larger than it is...

[sluckey]

Your meter is the culprit. Even a 10M input on your DMM will load down the voltage on that high impedance grid. Cathode followers, cathodynes, and LTP PIs all have very high impedances so the grid voltage cannot be measured accurately.

[mxrshiver]

ahhhh that does track, thanks sluckey! are there common workarounds to that? is measuring grid voltage in reference to other electrodes typically more accurate? or should i also take that -0.36V with a grain of salt?

[sluckey]

I don't have a workaround for that fancy DCCF circuit. The only DCCF circuits I've worked with simply have the CF grid connected directly to the previous plate, so that makes voltage measurement easy. The method I use for AC coupled CF, cathodynes, and LTP PI is to simply measure the voltage at the bottom end of the grid resistor, as in these examples...

     http://sluckeyamps.com/lil_maggie/Magnatone_M2.pdf
     http://sluckeyamps.com/phoenix/phoenix.pdf

Measuring voltage directly between grid pin and cathode pin should give a more realistic voltage reading in most cases but may not work in your circuit. Meter loading could also be a factor.

[pdf64]

With a 10M meter load. I think the measured grid to cathode voltage should be the effective bias voltage, near as dammit.

Grid current can be assessed by the V DC across the 330k ‘grid stopper’.

[cfortner]

Or an oszi with 100:1 divider, that means 100MOhm