MOSFET PI, how to Bias?

[Doug H]

I've added a irf820 as a cathodyne PI. After much fussing around I am finally getting sound out of it, great. Volume seems about half and even though the amp generally sounds good, it doesn't sound quite right. I feel I don't have the mosfet biased properly.

I've got a dedicated power node at 235v running the PI and voltage divider for biasing, but my readings make no sense at all.

Just trying to setup the voltage divider doesn't work as expected. 1M/1M gives me ~60v at the junction, that's without the mosfet in the circuit. I add the mosfet, with a 12v zener source to gate and 47k top and bottom, the source and drain both read ~110v to ground. I think the gate still reads around 60v. I measure the voltage drop from gate to source and it's 3.5v.

I don't understand all these wonky readings. Is my PS fried? I get expected resistance along the b+ line. I think I even got the correct 4 or 5k from the PT side of one of the rectifier diodes to the PI power node.

The advice I've read is to try and set it up with 1/4 b+ on the source. I'd estimate I'd have to use something like 15k source resister to ground, maybe because my b+ is so low? Isn't that going to pull more current through the mosfet and drain a lot of the signal strength? Is my b+ too low for a mosfet here?

At this point I'm swinging in the dark with different bias resisters hoping some combination magically works out. I have some lnd150s coming which I understand can be biased off the source like the tube counterpart, but I don't get what's going on here, and I have no scope or signal generator.

Thanks

[loudthud]

You want to set the Voltage divider to about 1/4th the B+ Voltage. When you try to measure the Voltage with your DVM, you probably won't get an accurate reading because the meter loads the Voltage down too much. Just measure the Source Voltage of the MOSFET and if it's close to 1/4th, and the Drain Voltage is 3/4ths, everything is good.

[Doug H]

Thanks a lot, that worked.

Doesn't like my cross line MV though, it gets all gated before it effects volume much. Maybe I screwed up something else, anyhow the PI works great so far. Amp sounds a little stiff, i had a 100u on the old tube PI, maybe I can drop that for the mosfet, or even run it off of the preamp node.

[Doug H]

So what should I expect altering the source and drain resistor values? There is the RC output filter, but aside from that, higher values will use less current, at some point will the PI not be able to drive the PT grids and I'll start getting that ugly harmonic content I stuck the mosfet in there to avoid?

I can't seem to find many examples or discussions regarding this. At the low voltage I run this at, around 240v, it "seems" fine so far in the 47k 56k range.