High Speed, High Current Op-amp

Bi-FET reefers to amplifiers that use both bipolar and field-effect transistors in the same circuit.  Early on, it was often a primitive mosfet driving a bipolar Q.  This yielded a composite device that had the high Z-in of the mosfet and the ruggedness of the bipolar.  Connecting the drain of the mosfet to the collector of the bipolar was often possible, this made the mosfet current flow through the load as well as the bipolar.  This kept things relatively efficient.  In practice mosfet control terminals usually look like capacitors, which bipolars can be very good at slewing.

Nowadays, since mosfets have gotten better, a mosfet is often the final.  A bipolar in common-emmiter mode is an inverter, providing 180 degrees of phase.  A mosfet in common source, driven by the switching bipolar also imposed 180 degrees of phase.  So the overall amplifier would have 360 degrees, equal to zero degrees (with a little delay) of phase.  The bipolar amplifier, on the left side, has an input resistor, typically, to convert the control potential into a current.  If we attach a large resistor from the final (mosfet) output all the way to the (bipolar) input, when the power switch we are constructing starts to turn on, load current (due to Q2) will make it turn on harder.  The on-to-off threshold has been biased away from the off-to-on threshold, thereby creating a Schmidt trigger.  This output side would not be good for audio, unless we are doing class D.  It makes for a very crisp pulse amp in terms of rise and fall times because those characteristics are not set by the driving pulse.  In a sense this output stage 'cleans up after' the controlling pulse.

Bi-FET reefers to amplifiers that use both bipolar and field-effect transistors in the same circuit.  Early on, it was often a primitive mosfet driving a bipolar Q.  This yielded a composite device that had the high Z-in of the mosfet and the ruggedness of the bipolar.  Connecting the drain of the mosfet to the collector of the bipolar was often possible, this made the mosfet current flow through the load as well as the bipolar.  This kept things relatively efficient.  In practice mosfet control terminals usually look like capacitors, which bipolars can be very good at slewing.

Nowadays, since mosfets have gotten better, a mosfet is often the final.  A bipolar in common-emmiter mode is an inverter, providing 180 degrees of phase.  A mosfet in common source, driven by the switching bipolar also imposed 180 degrees of phase.  So the overall amplifier would have 360 degrees, equal to zero degrees (with a little delay) of phase.  The bipolar amplifier, on the left side, has an input resistor, typically, to convert the control potential into a current.  If we attach a large resistor from the final (mosfet) output all the way to the (bipolar) input, when the power switch we are constructing starts to turn on, load current (due to Q2) will make it turn on harder.  The on-to-off threshold has been biased away from the off-to-on threshold, thereby creating a Schmidt trigger.  This output side would not be good for audio, unless we are doing class D.  It makes for a very crisp pulse amp in terms of rise and fall times because those characteristics are not set by the driving pulse.  In a sense this output stage 'cleans up after' the controlling pulse.