ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

previous: Don Symes Since that section is still 'coming... -- 1132630145

Re: Mosfet PI

11/22/2005 11:33 AM
R.G.	Re: Mosfet PI My apologies - I have a stack of backlog projects and work that's literally a decade or so deep, so the interval between things is longer than with other people, I think. A split-load phase inverter is simple with MOSFETs and meets the criteria for "no gain" applications. You put a MOSFET with equal source and drain resistors and bias the gate at about 1/4 of the available B+ through a largish resistor, 1M being typical. The gate being at 1/4 of the B+ forces the source to be at the same point minus the negligible Vgs needed to make the current flow. Since there is no gate current, the same current flows in the drain resistor and you get the desired phase splitter action. The high feedback from an unbypassed source resistor forces the gate-source capacitance to look tiny, no gain artifacts appear because the gain is unity. You take the signals off source and drain to the power tubes. You protect the MOSFET against overvoltage on the gate-source by tying a 12V zener to the gate/source, cathode to gate. I suppose you could fake current flow on positive grid voltages by using a lower voltage zener there with a 5.1K resistor in series with it. The differential style phase inverter needs some work I have not had time to do yet. It is not a unity-gain stage, so the gate-source capacitance could be a problem It might need followers in front of the differential to avoid problems driving the Cgs.

11/22/2005 11:33 AM

R.G.

Re: Mosfet PI

My apologies - I have a stack of backlog projects and work that's literally a decade or so deep, so the interval between things is longer than with other people, I think.

A split-load phase inverter is simple with MOSFETs and meets the criteria for "no gain" applications. You put a MOSFET with equal source and drain resistors and bias the gate at about 1/4 of the available B+ through a largish resistor, 1M being typical. The gate being at 1/4 of the B+ forces the source to be at the same point minus the negligible Vgs needed to make the current flow. Since there is no gate current, the same current flows in the drain resistor and you get the desired phase splitter action.

The high feedback from an unbypassed source resistor forces the gate-source capacitance to look tiny, no gain artifacts appear because the gain is unity. You take the signals off source and drain to the power tubes.

You protect the MOSFET against overvoltage on the gate-source by tying a 12V zener to the gate/source, cathode to gate. I suppose you could fake current flow on positive grid voltages by using a lower voltage zener there with a 5.1K resistor in series with it.

The differential style phase inverter needs some work I have not had time to do yet. It is not a unity-gain stage, so the gate-source capacitance could be a problem It might need followers in front of the differential to avoid problems driving the Cgs.