RnFR wrote:Findeton wrote:KMG wrote:Base stage of AMT LA series
la_stage.jpg
Source and drain capacitors differ from stage to stage (may not exists)
Interstage circuit is also different for different stages
I get the idea but I differ on the implementation. The diode is used to emulate grid current happening on a triode when the grid voltage becomes positive. But the diode has an exponential current function whereas the grid current is a power function. That's why I think adding another fet between gate and drain to emulate the grid current would make more sense. What do you think?
it's a great idea, can it be used for lower voltages as well? these preamps are freakin amazing, but i would really like know if it can be adapted to lower voltages, simply for ease of use and circuit simplification for lazy slobs like me.
I also discussed this on
thispost. A low voltage implementation of this solid state triode could be this one:
Which has similiar current graphs than a triode, just with lower voltages. This is for Plate current. The R7/R5 ratio controls the distance between the current lines for different values of Vgk, and R6 controls the plate current gain:
But that graph is nothing new, the thing is that we achieved a grid current function like this one:
RDISTORTION controls the gain of the grid current. A diode between the gate and source of a FET would also produce something similar to the "grid current" of a triode and that's what the circuit of KMG does, but that would be an exponential function, something like:
Ig=e^(Vgk/const1 - const2)
when the grid current is a power function:
Ig=k*[ (Vgk)^3 ]*[ (1-Vpk/500)^3 ]
I won't enter into details here, but in practice, because of using a load line and because of grid current, the term (1-Vpk/500)^3 is practically constant for a given stage (therefore we can control that constant with RDISTORTION, which value should be adjusted for each stage). The FET I added to the design (M24 in the schematic) has a transfer function similar to this one:
Ig=k*(Vgk-0.4)^2
Which is a power function, instead of exponential. So what do you people think of this other approach? I guess it's time to test the sound.
Thanks 
)
