Determining the output resistance R_{out} of linear circuits (for example, amplifiers) can often be done by manipulation of the circuit model  often the smallsignal equivalent circuit  to highlight those components connected to the output terminals, and by evaluation of the networks that result.
This simple approach falls down in circuits where there is negative feedback  for example in source follower circuits, and in OpAmp circuits where negative feedback is almost invariably in use. The inspection method gives incorrect and misleading answers in these cases.
A different method is needed for such circuits. The method described below is quite general, and can be used perfectly satisfactorily, even in circuits where no negative feedback is applied, and will it lead to the same result as the method of inspection in these simple circuits..
We start with a suitable circuit, typically the smallsignal equivalent circuit representation. For OpAmp circuits this is typically the schematic circuit drawn using the nonideal OpAmp symbol. Then we proceed as follows. We:
We must evaluate the value v/i, which is the output resistance R_{out}.
Whatever the circuit of our amplifier, be it JFET, MOSFET, OpAmp or some other technology, we can represent it using the generic amplifier form, shown in the pecked rectangle below. Our aim is to show that if we use the approach presented above, the quantity v/i represents the output resistance of the circuit, which we denote R_{out}.

If we shortcircuit the input to earth, the input signal is by definition 0, and the output from the amplifier stage is also 0. If we represent the circuit by means of our generalpurpose amplifier model (see Lecture 15), the voltage generated by its internal voltage generator, Av', is then also zero. If voltage source Av' is a zero voltage source we can regard it equivalently as a shortcircuit. Replacing that generator with a shortcircuit, and with the external source v connected, we see that the current that must flow, i, is just v/R_{out}. Hence R_{out} = v/i. 