1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Find the Thevenin Equivalent Circuit

  1. Sep 22, 2016 #1
    1. The problem statement, all variables and given/known data
    upload_2016-9-22_17-37-4.png

    2. Relevant equations
    Node Voltage Method
    V=IR
    upload_2016-9-22_17-43-0.png


    3. The attempt at a solution
    KCL at Node 1:
    upload_2016-9-22_17-34-30.png

    KCL at Node 2:
    upload_2016-9-22_17-35-29.png
    i is the current moving towards "a"

    I've simplified the two equations down but the problem I run into is that normally I'm able to write a third equation to define i0. In this case there is no resistance. i0=v1/R but since R=0 this would be undefined. Where do I move from here?
     

    Attached Files:

  2. jcsd
  3. Sep 22, 2016 #2

    gneill

    User Avatar

    Staff: Mentor

    Get creative :smile:

    Insert a "temporary" resistor in the ##i_0## path, call it ##R_x##. Solve the problem keeping ##R_x## as a variable. Then let ##R_x → 0~Ω##.

    By the way, your KCL for node 2 doesn't make sense with the v1 and v2 being summed as currents. I think you missed dividing by a resistance value...
     
    Last edited: Sep 22, 2016
  4. Sep 22, 2016 #3
    I need to eliminate v1 when I'm solving the system. How can I do this with having v1/Rx now being involved? I can't eliminate this term or combine it with the other v1. If I find a common denominator to combine it I can't eliminate that whole thing.
     
  5. Sep 22, 2016 #4
    If I were to solve this using KVL and I made 3 loops the first loop would give me a value i1. That value i1 would be the current across the 3kohm resistor. i2 would be the current across the 2kohm resistor. Can I consider that part of the wire with i0 to be i1-i2? Or for the sake of my KCL equations (v1-15)/3k - (v2-v1)/2k = i0?
     
  6. Sep 22, 2016 #5

    gneill

    User Avatar

    Staff: Mentor

    Solve your v1 node equation for v1. Yes, it'll have both Rx and v2 variables in it. You'll substitute the whole lot into your v2 node equation to replace 'v1' there. Only after substitution will you let Rx go to zero.
     
  7. Sep 22, 2016 #6

    gneill

    User Avatar

    Staff: Mentor

    Yes, if you're using mesh currents. Mesh currents sum through shared paths (i.e. on the "borders" of the loops.
     
  8. Sep 22, 2016 #7
    I decided just to solve using mesh currents. My professor's supposed easy way to solve it without having to solve for Vth then Isc apparently wasn't all that much easier for this problem.
     
  9. Sep 22, 2016 #8

    gneill

    User Avatar

    Staff: Mentor

    I think that any way you approach this problem the algebra will not be pretty. The nodal analysis approach that I suggested works, but again the algebra will be a bit time consuming. If you show your complete solution I'll offer mine :smile:
     
  10. Sep 25, 2016 #9
    upload_2016-9-25_14-0-12.png

    My teacher mentioned in class the note I put in the top right corner which would make it easier to solve the whole thing in one go.
     
  11. Sep 25, 2016 #10

    The Electrician

    User Avatar
    Gold Member

    Your final result schematic showing a voltage Vth in series with Rth appears to have the sign of Vth reversed.

    Using linear algebra techniques the problem can be solved in a compact manner using nodal analysis. Here's how it would be done if the wire carrying i0 is temporarily replaced with a resistor Rx. After the node voltages are found Rx is allowed to go to zero:

    M1.png

    The problem can also be solved with mesh analysis. The wire carrying i0 can be treated like a current source, so the 3 meshes form a supermesh:

    M2.png
     
  12. Sep 25, 2016 #11

    gneill

    User Avatar

    Staff: Mentor

    Okay, everyone's showing a solution and I did promise to show mine if the OP showed his. So here we go.

    Insert ##R_x## in the ##i_o## path so that we can write node equation for node 1. Later we'll let this resistance go to zero. Also stick a load resistor ##R_L## on the output (across a-b). You'll see why in a bit.

    upload_2016-9-25_17-43-57.png

    Write the node equations:

    Node 1:
    ##\frac{v1 - 15}{3000} + \frac{v1}{R_x} + \frac{v1 - v2}{2000} = 0##

    and solving for v1: ##~~~~v1 = \frac{3}{5} (10 + v2) \frac{R_x}{R_x + 1200}##

    Node 2:
    ##\frac{v2 - v1}{2000} + 18\frac{v1}{R_x} + \frac{v2}{2000} + \frac{v2}{R_L} = 0##

    Factor out v1, then substitute for v1 from the node 1 equation. Hit the whole thing with the algebra hammer until v2 is isolated. When the smoke clears:

    ##v2 = \frac{30 (R_x - 36000) R_L}{(7 R_x R_L + 120000 R_L + 10000 R_x + 12000000)}##

    Now it's time to let ##R_x## go to zero:

    ##v2 = \frac{-1080000 R_L}{12000000 + 120000 R_L}##

    Now make this look like a voltage divider equation. For a Thevenin model with a load it will resemble:

    ##V_{out} = V_{th} \frac{R_L}{R_L + R_{th}}##

    Thus we have:

    ##v2 = \frac{-1080000}{120000} \frac{R_L}{R_L + \frac{12000000}{120000}}##

    ##v2 = -9 \frac{R_L}{R_L + 100}##

    So the Thevenin voltage is -9 V and the Thevenin resistance is 100 Ω.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Find the Thevenin Equivalent Circuit
Loading...