Bias resistor for source follower driver

[SpreadSpectrum]

I'm using fqpf2n80's as source follower drivers (direct coupled to KT88 grids). I apply bias to the gates of the 2n80's which are capacitively coupled to the previous stage. I'm currently using 500k resistors to bias the gates of the 2n80's. I'm considering increasing this resistor.

My question is, how high is too high?

[SpreadSpectrum]

Anyone? I want to apply bias through a 1meg resistor to the FET gate. Anyone see anything wrong with that? How much would be too much?

[wa2ise]

I don't know enough about that FET, but if you are concerned about losing too much bass because of a lower resistance value, you can use a larger coupling cap instead. Other than that, I can't really help.

[ErikdeBest]

The question is interesting, so I keep an eye on the thread, but unfortunately I can't help...

[Fuling]

1M should be perfectly fine, I´ve seen circuits with several Megaohms in that position.
A side note: Perhaps you just want to get away with small coupling caps but if it is your driver stage that cannot drive 500k you might sface some problems driving the input capacitances in the mosfets.

BR
/Daniel

[SpreadSpectrum]

if you are concerned about losing too much bass because of a lower resistance value

Actually, I want to increase the resistance to use a lower value cap (or improve the bass performance with the current cap). I'm just concerned that there is something that is going to get me if the value is too high. Is my only consideration the leakage through the coupling cap to the previous stage? Do I consider the FET to have a truly infinite input impedance?

[SpreadSpectrum]

Previous stage is 6SN7 LTP. 500k sounds fine, I just thought that 1M might sound a tiny bit better.

[Eli Duttman]

The FQPF2N80 appears to be a useful part. Unlike tubes, FETs are immune to contact biasing effects. Even a 10 MOhm gate to ground resistor rates to be safe. Don't forget to use a Carbon comp. gate stopper right at the FET's gate terminal.

[SpreadSpectrum]

The FQPF2n80 sounds good. I've got the gate stopper. I'll try out higher valued resistors today.

[gingertube]

I used 1M5 without any problems which allowed the use of a 10nF coupling cap.

As a few folk have indicated above it is best to choose the FET for low input capacitance. Because you are using it as a source follower Crss is NOT multiplied (no Miller Effect) and Ciss is effectively "anti-bootstrapped" (source follows gate AC wise), as is the capacitance of the protection zener which MUST be used between gate and source. I would choose the MOSFET for low Crss. Whenever the power dissipation rating is adequate I like to use ZVN0545A.

The power rail to the MOSFET drain must be clean and tight and you need to make sure that on the most positive signal swing at the source you still maintain a minimum of say 30V from Drain to Source. This will help minimise effects of the Crss being modulated by the audio signal.

Cheers,
Ian