I have slightly adjusted the scheme. This reduced the distortion.

 

Ian0,
I'm fulfilling my promise. I have overloaded the amplifier so that the mos transistor closes, but Q1 does not close.

 

Ian0,
I'm fulfilling my promise. I have overloaded the amplifier so that the mos transistor closes, but Q1 does not close.
View attachment 350964

Almost, but doesn't quite shut it down, but there is still an 80% reduction in the LTP current.
If the LTP were resistor-loaded it would be in trouble.
By the way, there's a lot of voltage on M1 gate!

 

Almost, but doesn't quite shut it down, but there is still an 80% reduction in the LTP current.
If the LTP were resistor-loaded it would be in trouble.
By the way, there's a lot of voltage on M1 gate!

And what voltage should be on the gate at an input voltage of -2.87 V? You yourself forced me to apply excessive voltage to the input in order to close the transistor M1, and then you are surprised at the voltages. At normal input voltages, the gate voltages are normal.