In the proof of this theorem a test current source is attached to the terminals of a network called $N$. We want to know the equivalent of network $N$. Then we calculate the potential at this terminal which is:

$$\Delta V = V_\text{th} + R_\text{th} I_\text{external} \, .$$

$V_\text{th}$ is the potential due to the network and $R_\text{th} I_\text{external}$ is the potential due to the external current source and equivalent resistance of the network.

The key point in the proof is that in order to calculate the potential of the network we do not consider the external current source and open circuit it. This is understandable because we are using superposition principle.

However, what if we connect the network $N$ to another external network called $E$. Then in order to calculate the potential at the terminal we cannot open circuit the terminal because when we ignore the sources in the external network $E$ it does not mean that the terminal is open now. Maybe still current goes to the external network $E$. How can we prove the theorem in this case?

Another issue is that when instead of current source we use another voltage source. Then the potential at terminal should be the same as that of the voltage source. How can we get the equivalent circuit when we have a voltage source at the terminals.


Here is the Thevenin theorem form Wikipedia:

  1. Any linear electrical network with voltage and current sources and only resistances can be replaced at terminals A-B by an equivalent voltage source $V_\text{th}$ in series connection with an equivalent resistance $R_\text{th}$.

  2. The equivalent voltage $V_\text{th}$ is the voltage obtained at terminals A-B of the network with terminals A-B open circuited.

  3. The equivalent resistance $R_\text{th}$ is the resistance that the circuit between terminals A and B would have if all ideal voltage sources in the circuit were replaced by a short circuit and all ideal current sources were replaced by an open circuit.

  4. If terminals A and B are connected to one another, the current flowing from A to B will be $V_\text{th}$/$R_\text{th}$. This means that $R_\text{th}$ could alternatively be calculated as $V_\text{th}$ divided by the short-circuit current between A and B when they are connected together.

  • 4
    $\begingroup$ I think it would be easier to answer this question if you clearly and explicitly state what "Thevenin theorem" is in your mind :-) $\endgroup$ – DanielSank Jan 8 '17 at 3:19
  • $\begingroup$ I still don't get it. Why should the $V_\text{th}$ be the voltage of open circuit terminal? $\endgroup$ – MOON Jan 8 '17 at 13:52

I wrote up a proof of Thevenin's Theorem going through each step in detail, inspired by another question. One of the answer's referenced a text by Kendall Su "Fundamentals of Circuits, Electronics, and Signal Analysis", Appendix A.1, page 568).

For the general form of Thévenin's theorem we allow unlimited resistors and unlimited independent voltage or current sources. (Dependent sources add a small complication not covered here. They basically get grouped in with the resistors.)

Illustration of Thevenin's theorem

Select any two nodes inside the circuit and bring them out to a port. The goal is to prove the voltage and current can be related like this: $v = \text V_\text T + \bf Ri$, where $\bf R$ stands for some expression composed of resistance values.

Assume a linear network composed of,

  • any number of resistors,
  • $N$ voltage sources: $\text V1 \ldots \text V_N$,
  • $M$ current sources: $\text I1 \ldots \text I_{M-1}$ internal current sources plus an external current source $\text I_M$ connected to the port.

Linear circuit with external current source connected to the port

Now apply the principle of superposition. The original circuit can be decomposed into $N+M$ sub-circuits, one for each source. Each sub-circuit contains a single source and some arbitrary network of resistors. (All the other sources have been suppressed.)

For the $N$ sub-circuits with a voltage source, the voltage at the port is always a scaled version of the internal voltage source. The scale factor is a dimensionless ratio of resistances, $\bf A$,

$v_n = \bf A_n \text V_n, \quad$ where $n$ ranges from $1 \ldots \text N$

For $M-1$ sub-circuits with an internal current source the voltage at the port is always the Ohm's Law product of the current source $\text I_m$ times some resistance expression $\bf R$.

$v_m = \text I_m \bf R_m, \quad$ where $m$ ranges from $1 \ldots \text M-1$

The last current source is the external $\text I_M$, which produces this voltage,

$v_M = \text I_M \bf R_M$

where $\bf R_M$ is the equivalent resistance looking into the port when all the internal sources are suppressed.

We've found all the separate contributions to the port voltage. Now superimpose (add them all up),

$\displaystyle v = \sum_{n=1}^N v_n + \sum_{m = 1}^{M-1} v_m + v_M$

$\displaystyle v = \sum_{n=1}^N A_n \text V_n + \sum_{m=1}^{M-1} \text I_m \bf R_m + \text I_M \bf R_M$

The first two summation terms produce voltage values based on the internals of the circuit, with nothing connected to the port. For every one of these sub-circuits the external current source was suppressed. The combination of these two summation terms gets a special name, $v_{oc}$, which stands for open-circuit voltage,

$v_{oc} = \displaystyle \sum_{n=1}^N A_n \text V_n + \sum_{m=1}^{M-1} \text I_m \bf R_m$

$v_{oc}$ is the voltage that appears at the port when the port is left open.

With this new variable name we rewrite the superposition equation,

$v = v_{oc} + \text I_M \bf R_M$

The whole arbitrary circuit boils down to this equation. It's exactly what we want. This voltage-current relationship allows us to construct a Thévenin equivalent circuit,

Thevenin equivalent voltage source and resistor

$\text R_\text T = \bf R_M$

$\text V_\text T = v_{oc}$

$v = \text V_\text T + i\,\text R_\text T$

Both the original complex circuit and the little Thévenin equivalent obey the same $i$-$v$ equation. Done! We proved any circuit composed of resistors, voltage sources, and current sources can be reduced to a single voltage source and a single resistor.

  • $\begingroup$ Welcome to Physics Stack Exchange! Please do not post links without at least a basic summary of what they contain and how that content answers the question, since link-only answers become useless if the link rots away. I'm deleting this since link-only answers are not considered answers here. If you edit this to contain the relevant information, you can flag it for moderator attention and we'll review it for undeletion. $\endgroup$ – David Z Apr 8 '18 at 5:20

The above answer is OK but needs some more details. (1) The answer seems to be valid for resistive networks containing only independent voltage and current sources. That is inclusion controlled sources have not been considered. (2) I quote from the above answer "The last current source is the external $I_M$, which produces this voltage,


where $R_M$ is the equivalent resistance looking into the port when all the internal sources are suppressed" I think this needs justification in a proof.

All these can be addressed if we use modified node equation technique to represent an electric network. The technique may be described as follows.

Modified Node Equation Technique

Assume that the circuit has n+1 nodes of which node #0 is labeled as reference node. The circuit contains LRCM passive elements, independent voltage and current sources and linear controlled sources. The following algorithm would augment the circuit so that node equation technique may be used for its analysis. The underlying concept is to replace all the sources other than the independent current sources by unknown current sources. With these replacements included one can easily obtain n node equation associated with n nodes. To solve the circuit, however, one would require n+m number of equations, where, m is the total number of active sources that are replaced by unknown current generators in the augmented circuit. The extra m equations are readily obtained from the specifications of the replaced sources.

Augmentation Algorithm:

• Replace all independent voltage sources, controlled current sources and controlled voltage sources by unknown current sources. • Obtain the set of n node equations corresponding to n nodes of the circuit. • For an independent voltage source equate the output voltage with the difference between the terminal node voltages to get an extra equation. • For a controlled source express the control law in terms of the node voltages and the circuit parameters to get the necessary extra equation.

We may now state Thevenin's theorem as follows. Consider a one port network A that is terminated to a load network B.

Network A may contain linear passive elements and linear independent and controlled sources. On the other hand there is no restriction of linearity for network B. However, following constraint exist 1. There is no electro-magnetic coupling between network A and B. 2. There exists no source at network A that is controlled by any voltage or current in network B. Under this condition Thevenin’s theorem states that the network A may be replaced by a voltage source VTH connected in series with an impedance ZTH where VTH is the open circuit voltage measured at the port of A and ZTH is the driving point impedance of A with all the independent sources in A reduced to zero.

Since there exists no coupling between the circuits A and B we can consider network A separately with an out put current I1 flowing away from its port. The port current I1 and the port voltage V1 will be the solution of the pair of equations relating V1 and I1 in A and in B respectively. Thevenin’s theorem aims at deriving the relationship valid for the first network that is network A.

Following the modified node equation technique one can find the general representation scheme for an arbitrary A network. (for detail description you are referred to https://www.researchgate.net/publication/262764661_Thevenin%27s_Theorem_An_Easy_Proof_Suitable_for_Undergraduate_Teaching

Using this representation one can very easily express the voltage at the port of A similar to the statement of Thevenin's theorem. This is well explained in the above link.


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.