TABLE OF CONTENTS A convenient technique to implement the current dividers with MOS transistors instead of resistors is described in [26] and [27]. The principle is based on the symmetrical behavior of MOS transistors and the particular structure of their drain current equation:
Where I D is the drain current of the transistor, W and L its width and length respectively, and V S and V D the source and drain voltages. f(V G, V) is a function depending only on the gate voltage V G and the voltage through the channel V, which is the integration variable. It is simply the inversion charge density in function of the channel voltage [28]:
where ? is the electron/hole mobility. The equation for Q i depends on the inversion mode of the transistor. In strong inversion (the inversion factor I F is much larger than 1):
where C ox is the gate oxide capacitance (per unit area), n the slope factor, and V P the pinch-off voltage defined by:
V T0 is the threshold voltage. In weak inversion, (I F much smaller than 1):
where U T is the thermodynamic potential, ? 0 the surface potential and ? F the bulk Fermi potential.
Figure 30 shows a graphical representation of the function f(V G, V) = ?Q i. From the plot, I D can...
