Dependent vs. Indepenent sources?

[JLR]

Dependent vs. Indepenent sources!??...

Can anyone give any insight into why you can't treat dependent sources as open circuits (dependent current sources), and why you can't treat dependent voltage sources as short circuits (dependent voltage sources)? I understand why it is the case for independent sources, but I just don't see why the same thinking isn't extended to dependent sources. For example, I have an example in my circuits II textbook where the Thevenin voltage is 0V at the terminals measured. The goal is to find the Thevenin resistance, but for whatever unexplained reason a test current of 1A is sent into the circuit to excite the dependent source's controlling component... why can't you read "through" them like any other source when they are deactivated; that is, when there controlling parameters are not excited. To put another way just to attempt to be more clear, if there is no input to the circuit; effectively "zeroing" the dependent sources, why can't you still read the equivalent resistance through the dependent sources like you do for ideal sources? I have not read any justification for this anywhere for this circuit analysis technique.

 

[MrSparkle]

its pretty much by definition. An independent source is like an independent variable. You can set it to whatever you want. By setting it to zero, you make the circuit much simpler. So you calculate what the circuit when only 1 independent source is turned on at any given time, then you can add them all up according to the superposition theorem. But dependent sources, by their very definition, are essentially reactive elements that you can't just turn off.

 

[donpacino]

Dependent sources are typically used as parts of a model for complicated elements such as transistors. You cannot simply set part of a model to zero, as that would break the model.

 

[JLR]

its pretty much by definition. An independent source is like an independent variable. You can set it to whatever you want. By setting it to zero, you make the circuit much simpler. So you calculate what the circuit when only 1 independent source is turned on at any given time, then you can add them all up according to the superposition theorem. But dependent sources, by their very definition, are essentially reactive elements that you can't just turn off.

So then will it be appropriate to say then that since dependent sources are by definition reactive components, that even if there controlling elements have a zero input, by trying to measure the circuits resistance at that moment, you can't predict how the reactive components will respond by trying to measure the circuits resistance at that point?

 

[MrSparkle]

I think you almost have it. I'm trying to be precise with my language. This is a technique meant to overall simply the circuit. So saying you 'can't' predict them is inaccurate because we are trying to quantify them. The central problem is that dependent sources are dependent on all the sources, even if they look like they are only dependent on one.