Choosing central bias

[Merlinb]

A lighter load means its ratings will go up? How can I practically measure that?

You don't really need to, it makes little difference to the operation of the preamp. It's not like a high voltage means you suddenly have to build a Soldano or anything!

If the transformer is 190Vac, then the max DC voltage will be 190*root 2, or 266V. The more you load the transformer the more this will fall, and according to the manufacturer you expect 250V at full load. Therefore you can be pretty sure that whatever you do, your HT voltage will be between 250V and 266V, which is a mere 6% variation! Don't worry about it.

[snowballfight]

In response to MWJB earlier; "Well, you won't really be able to avoid going lower unless you use a very high value cathode resistor (>2.7K?) & a low value plate resistor"

Whats wrong with having a steep load line on the 12AU7, say with 47k to get to the bias point with a higher voltage? It would mean a cathode resistor of, I think i had this at one point, 4k? Would this actually damage/affect the tube?

[MWJB]

No, it won't damage the tube, it might sound harsh though? Try it & see.

Normally preamps are designed around cathode & plate values that give the desired character, then tweaked as far as plate/B+ supply voltage, bias current typically ends up as what it is after the other perameters are decided.

[Wilder Amplification]

In playing with the 12AU7 load line that's about what I got too. They're not a very high gain tube.

In this load line I drew up for one, at a supply voltage of 250V and a plate resistor of 25K it gets me a gain of about 12.5. An 820R resistor will bias you right up to where you wanna be a -4V in this scenario, which is as close to "center bias" as you're gonna get.

[snowballfight]

With a gain of only 12.5, am I pretty much good to couple that directly with the second stage? through a 22uF, perhaps a little lower (10uF), and relatively low kilo-ohm resistor.

[Wilder Amplification]

Not sure why you'd wanna do that, but that all depends on what your actual output signal amplitude from that stage is. The gain just determines the "input sensitivity" of the stage (i.e. how much input signal it takes to drive the output signal to a given amplitude).

Using a low value leak resistor on the proceeding stage will load the driving stage's output somewhat.

[jimboyogi]

I've just built a 12AU7 first and second stage cascaded pre-amp. ~ 250V B+ (depends on what else I run from the power supply as I play around with other channels), 47K anode resistors. For the first stage I am biasing it with a blue LED, giving 2.8V bias, and a voltage gain of ~12. Second stage has Rk = 3.3K, 10uF bypass cap, giving 5.9V bias, and voltage gain of ~11 (as this pushes into the non-linear area of the grid curves). I chose this bias for symmetry of clipping/grid current, and tweaked the bias to get this balance using a CRO and signal generator. The slight compression from this grid curve bunching sounds pretty good IMO. I put a 1M pot between the stages, using it as the second stage grid leak. I designed this pre-amp to be a 'clean' channel. I'm really happy with the sound, and it would really suit a bass pre-amp I think.