User:John R. Brews/Sample

In electronics, the mode of an electrical device refers to its steady-state bias condition or operating point in the absence of signals. In analog circuits the so-called active mode of the device is chosen by the circuit designer to allow adequate signal amplitude and adequate voltage or current gain, along with acceptable signal distortion. In digital circuits, devices toggle between the off-mode (or cutoff mode) and the on-mode (or saturation mode in bipolar transistors, or ohmic mode for MOSFET's), and visits the active mode only briefly while switching between the on and off modes.

For historical reasons, the saturation mode of the MOSFET refers to its active mode, while the saturation mode of the bipolar transistor invariably refers to its on mode. This confusion of terminology does nothing to clarify discussion.

In the bipolar device, the emitter is designed for efficient injection, while the collector is designed to collect with low capacitance between collector and base. Thus, the bipolar device is inherently asymmetrical, and a distinction between forward and reverse modes of operation makes sense. In the MOSFET, the source and drain are interchangeable, so reversing polarity simply exchanges the source for the drain. An exception is the power MOSFET, which like the bipolar transistor, has source and drain separately optimized for their particular function.

In the MOSFET, the threshold at source (or drain) is altered by the source-to-body (or drain-to-body) voltage, requiring a larger gate-to-source (gate-to-drain) voltage the larger the reverse bias. Hence, the two threshold voltages VT(S) and VT(D) are different if the two reverse biases differ.

The table separates modes based upon the presence or absence of an inversion layer, or channel, at one or both ends of the channel region, concepts that retain meaning even for modern MOSFETs with very small dimensions. On the other hand, the obsolete Shichman-Hodges model that short-circuits the source to the body to obtain a three-terminal representation, bases the distinction between modes upon voltages: VDS < ( VGS – VT ) or VDS > ( VGS – VT ), an approach limited to this outdated framework.