I understand this has been asked before, but I've never been able to find a satisfactory answer to the question I have, which is: in general, how do we treat the dependent sources in the small signal models when calculating the short-circuit time constants?

Case in point: I ended up with the following circuit (in part) when doing the low-frequency analysis of a CS-CB circuit, for one of the capacitors:

Part of the Circuit

From the example given in this question, I guessed that the equivalent resistance would be R1||R2. Or, as my lecturer said about a similar circuit, the right-half of the circuit is "completely separate" from the left-half.

However, I thought the way to compute equivalent resistance was to replace the capacitor with a voltage source, and to calculate the current? In that case, G would have a non-zero potential, and then wouldn't the dependent current source add current from the right-half of the circuit to the left-half...?

Am I understanding "equivalent resistance" incorrectly? When can we "ignore" the dependent source when finding the resistance "seen" by the capacitor?

  • \$\begingroup\$ The left and right halves are only connected at one point. There's no return path for anything to flow from one side to the other. \$\endgroup\$ – Roger Rowland Sep 18 '16 at 6:30
  • \$\begingroup\$ @RogerRowland Thanks for the comment Mr. Rowland. In general, though, is it correct, as I described, to replace the capacitor with a test voltage and to find the resulting current? \$\endgroup\$ – user3109672 Sep 18 '16 at 6:40

Yes generally speaking using an indipendent voltage generator as a stimulus to find equivalent resistance is more than correct, it is applying its definition.

Just a few add-ons:

It doesen't have to be a voltage generator, can be either voltage or current, very often one of the two choices gives much simpler maths.
So before starting reckless calculations it is definitevely worthwile having a short survey to find out which is best in that case.

When we say parallel resistance being \$R_\text{p}=R_1R_2/(R_1+R_2)\$ we have actually connected a voltage generator, done the maths and finally remember the result by heart.
This is usefull since it's such a common problem, it's no use re-invent the wheel every time.

The same applies to series resistance (but with a current generator though).

A few other configurations may be worthly learned by heart, e.g. common collector basic output resitance is \$1/g_\text{m}\$ parallel something else, basic BJT input resitance with emitter degeneration is \$(1+g_\text{m}r_\pi) R_\text{E}\$ series something else, and so on depending or what you need/ are studying.

Recapping, use equivalent resistance defintion, hence voltage or current stimulus, is always correct but usually longer, so should be reserved to cases not covered above.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.