If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

RC step response solve (2 of 3)

We find the RC step response using the method of Natural + Forced response = Total response. Created by Willy McAllister.

Want to join the conversation?

  • male robot hal style avatar for user Aaditya Phadnis
    Why not use just regular methods for solving ODE rather than going through all that mathematical acrobatics?
    (8 votes)
    Default Khan Academy avatar avatar for user
    • primosaur ultimate style avatar for user Raul Lara
      Please don't be selfish, and let others know what methods you would have used to solve the ODE? This type of math isn't really geared towards mathematicians, the same is seen in physics lectures where they often just "guess" the solution.

      The reason for the "acrobatics" is to gain an understanding of different concepts as you go through the problem.
      (8 votes)
  • blobby green style avatar for user CalColson
    I posted this first as a tip, but think it will be seen more here... is this helpful?

    I was initially confused by this lesson, because the instructor is purporting to solve the total response through a summation of the natural and forced responses. However, the terminology used here is a bit loose. Each response consists of two parts, a transient portion and a permanent portion.

    For the forced response, the transient portion is the charging of the capacitor until the voltage across the capacitor is equal to Vs. The permanent portion of the forced response is the steady input voltage of Vs that is applied to the circuit.

    For the natural response, the transient portion is the drop in voltage of the capacitor as it discharges to 0 volts. The permanent portion of the natural response is nullified, because its steady-state is 0.

    Now, what the instructor refers to as the "natural response" portion of the total response is actually a sum of the two transient portions of both the natural and forced responses. What he refers to as the "forced response" portion of the total response is actually just the permanent portion of the forced response.

    However, the instructor's solution will always work out correctly, because he is still adding up all the pieces, just not in the intuitive order described. Also, the transient portions of the responses have the same form of A*e^(-t/RC), so when we solve for that constant 'A', we should get the correct value for the total response solution, as long as we do this solving under the constraints of the total response, and not just those of the natural response (which would give us a value of A = V_0).

    This link should help clarify some of the confusion on this lesson (read the top-rated answer by Felipe_Ribas): https://electronics.stackexchange.com/questions/93061/difference-between-natural-response-and-forced-response
    (7 votes)
    Default Khan Academy avatar avatar for user
  • blobby green style avatar for user soulimane.mammar
    Hi,
    I think the guessing that the solution of the equation
    dv/dt + v/RC = Vs/RC is a constant was not very successful
    The question of Chambers Wong was legitimate, the natural response is as you said in a previous video V0 e^-t/RC
    and there is no reason to pick an other constant because a time t=0 the voltage across the capacitor is V0
    The question of Learner was also legitimate the voltage across the capacitor can't be Vs (at least at the beginning).
    All this inconstancy is because the guess was not successful.
    Consider the mathematical solution for dv/dt + v/RC = Vs/RC which is Vs(1-e^-t/RC) [you can verify that's a valid solution].
    This time if we consider superposition we get:
    1) V0e^-t/RC
    2) Vs(1-e^-t/RC)
    and adding these give us the solution (V0-Vs)e^-t/RC + Vs
    (4 votes)
    Default Khan Academy avatar avatar for user
  • piceratops seedling style avatar for user Chambers Wong
    In previous video, it says natural response equation has a fixed constant ratio Vo. and in this video, you replace it with Kn. Since we already have presumed the initial i and v in natural response condition, then why not a Vo? thank you for your help.
    (2 votes)
    Default Khan Academy avatar avatar for user
    • spunky sam orange style avatar for user Willy McAllister
      In the previous RC Natural Response video, we made a guess at a solution. This is where we introduced a K as a placeholder for the amplitude term in our guess. In that problem, it turned out K = Vo.

      In this video on RC Step Response we start with a different circuit that includes a step voltage source. There is no justification for bringing forward the solution from the previous just-RC circuit. In this step response problem we use the same guessing technique to solve a different differential equation. Again, we introduce K as a placeholder for the amplitude term. K turns out to be Vo - VS in this case.
      (1 vote)
  • leaf green style avatar for user Daniel T
    For step response, is it valid to derive the step response formula from the natural response formula by changing the reference point so that Vs becomes ground (0)?

    The step response graph looked really similar to a natural response graph that was flipped over the x-axis and translated up.

    Natural response equation: V(t)=(V0)e^(-t/RC)
    Transform the graph so that V becomes V - Vs,
    and Vs becomes Vs - Vs = 0,
    and V0 becomes V0 - Vs,
    and V(t) becomes V(t) - Vs,

    so that we get the same formula without any calculus:
    V(t)=(V0)e^(-t/RC) becomes V(t)-Vs=(V0-Vs)e^(-t/RC)
    and we end up with the same V(t)=(V0-Vs)e^(-t/Rc)+Vs

    This looks too simple to be true. Is there anything wrong with this reasoning?
    (1 vote)
    Default Khan Academy avatar avatar for user
    • spunky sam orange style avatar for user Willy McAllister
      There's nothing wrong with your reasoning, but I think perhaps you are taking full advantage of hindsight since you know what the solution should be. I don't know if you could take this path if you started from scratch with just the natural response solution. It might be a leap to justify Vs --> 0.

      But you are correct, the step response is an 'upside down' version of the natural response. It makes intuitive sense at the end. For me the surprise is how twisty the math is compared to how intuitively obvious the answer comes out.
      (2 votes)
  • blobby green style avatar for user 😊
    I have a conceptual question: the forced voltage on the capacitor is equal to the voltage from the voltage source according to the video (the yellow box Vf = Vs at ). Why isn't there a voltage drop across the resistor? In other words, I do not understand why putting a capacitor after a resistor changes what a resistor does.
    (1 vote)
    Default Khan Academy avatar avatar for user
    • spunky sam orange style avatar for user Willy McAllister
      At , the forced response v_f turned out to be equal to the step voltage, V_S. It may seem strange that the forced response is a constant for all time. But remember, the forced response is not the whole answer, it's only a portion of the answer. We have to add the forced response to the natural response to get the total (the actual) response. So all by itself, the forced response is only part of the whole story.

      This method of forced + natural is a mathematical technique we use to break a hard problem into two easier problems. While the math works great, the intermediate result (the forced response) may seem a little strange.

      One way to interpret the forced response is to say it is the solution to the circuit after a long time has passed (this is true because any natural response of any circuit always dies out to zero, so the only thing happening after a long time is the forced response). In this step response circuit, the step happens and the capacitor charges up. After a long time, the capacitor voltage grows until it matches the source voltage, which means the voltage difference across the resistor becomes 0, and current stops flowing.
      (1 vote)
  • blobby green style avatar for user Gaofeng Fan
    I have two question. Question 1. At , why dVs/dt=0, can it be dVs/dt=Dirac(t)?
    Question 2, how if input Vs=Diract(t)+V0. I want to calculate the impulse response, but based on the above analysis in the video, V=Kn*exp(-t/RC), I cannot solve Kn.
    (1 vote)
    Default Khan Academy avatar avatar for user
    • spunky sam orange style avatar for user Willy McAllister
      Q1: Why is dVs/dt = 0 as opposed to dVs/dt = impulse.
      A1: We are solving the step response using the method of natural+forced response. The point is to break the problem into two simpler parts rather than trying to solve everything in a single analysis. We can separate the circuit problem like this because the principle of Superposition holds, which allows us to add natural + forced = total response.

      As the video says, we figure out the forced response after t=0, not for all time. For t > 0 the input voltage is constant at Vs, so the derivative with respect to time is 0.

      It might seem a little fishy that the forced response ignores what happened prior to t = 0. The forced response only deals with after t = 0. Dealing with before t = 0 is the job of the natural response part of the problem. When we add them together, the entire answer emerges.
      (1 vote)
  • blobby green style avatar for user olijah501
    How do we get vsubn = Ksubn e^t/RC from s=-1/RC?
    (1 vote)
    Default Khan Academy avatar avatar for user
  • eggleston blue style avatar for user dena escot
    @ at t=0 , Vt=Vo , is Vt=Vo before t equal to zero ?
    (1 vote)
    Default Khan Academy avatar avatar for user
    • old spice man green style avatar for user Willy McAllister
      Yes. Vt = V_0 before the step happens. Prior to the input voltage step the value of the voltage source is V_0. Let's let V_0=1V just to have something to talk about. Assuming it has been at V_0 for a "long time", what is the state of the circuit?

      The capacitor will be charged up to 1V. That makes the voltage across resistor R to be 1V - 1V = 0. With 0V across the resistor you know the current is 0, (I = V/R = 0/R = 0). So some charge is sitting on C sufficient to make the capacitor voltage = 1V. The circuit is just sitting there waiting for the step to come along. The step is the 'forcing function" that makes the circuit change state.
      (1 vote)
  • male robot donald style avatar for user Samir Mukshe
    At , I saw you move the screen for space.How do you do that?
    (1 vote)
    Default Khan Academy avatar avatar for user
    • spunky sam orange style avatar for user Willy McAllister
      It's pretty simple. I draw in Autodesk Sketchbook Pro. I make the page size really tall, like 2 screens high. The video is captured using Camtasia, which does a continuous screen grab of a rectangular section of my monitor. To move the drawing up it's just a matter of scrolling up the Sketchbook window while Camtasia is running.
      (1 vote)

Video transcript

- [Voiceover] In the last video on step response, we set up the differential equation that describes our circuit, and we found that it was a non-homogeneous equation, and now we're gonna follow through on the strategy of solving it with a forced response plus a natural response. So here's two copies of our circuit, and we're gonna, on the top one here, we're gonna solve for the natural response, and on the bottom one, we'll use that one for the forced response. Alright, let's go to work on the natural. Now, to do a natural response, what we do is we set the initial conditions to whatever they were for the original circuit, so that involves some q and a v zero, a v naught on the capacitor. So let's just draw in quick, let's draw in our q. There's some q here, that'll be a plus or minus v naught on the capacitor. Now the other thing we do, is for the inputs, we set the inputs to zero to do the natural response. And how you do that, our input here is a voltage source, and we're gonna suppress the voltage source, that's how we do it with super position. When you suppress a voltage source, what you do is you set the voltage to zero. And that means that's the same as creating a short circuit for this. So, for the natural response we're gonna take out our voltage source and replace it with a short circuit, like this. So, this circuit now is identical to the circuit that we used for the natural response. So, if you haven't seen that video on how to solve for the natural response, this would be a good time to go look at that. What I'm gonna do now is just write down the answer from that, this will be plus or minus, we'll call this v natural, and that equals some k natural times e to the minus t over r c. This is the natural response of an r c circuit, so I'll put a square around that so we can remember it. And I've left the constant in here, and we'll work out what this constant is a little bit later. Okay, so now let's move on and do the forced response for this circuit. And for the forced response, we remember the initial conditions are set to zero. So that means that q equals zero here, and that means that initial v is zero. I'll just write in zero. So that's what it means to set the initial conditions to zero. And the inputs are equal to, we use to inputs this time, the input is equal to v s. And in particular, the input is equal to v s, we're gonna solve the forced response after time is equal to zero, so that means that the input is v s, capital v s. And as a reminder, what we're trying to solve here is the differential equation from the previous screen, and that is c times d v d t, I'll put a forced in here, plus one over r, times v forced equals one over r times v s, and I can plug in capital v s here because we're trying to solve this for this initial condition. So this is our differential equation for the forced response, and we're gonna take a strategy here that's like we did with the natural response, is we're gonna guess at an answer for v f, and then plug it into this differential equation, this non-homogeneous differential equation, and we're gonna see if it works. And a good guess here is to guess something that looks like the input. So a good guess is gonna be some function that looks like v s. And v s looks like what? V s looks like, over here, v s looks like a constant. So we're gonna make a guess here that the forced response looks like some constant, we'll call that k f. The way we test our guess is to plug it back into here. Plug it back into the differential equation. So I'm gonna do that, alright so we get c times the derivative of k f, with respect to time, plus one over r, k sub f, equals one over r times v s. Alright, now here's the interesting thing that happens next. What's the derivative of a constant with respect to time? That's That goes to zero. So this first leading term here of our differential equation goes to zero, and now I'm left with k f times one over r equals v s times one over r. So that makes k f equal to v s. So I'll tuck this in here, our forced response, v f equals v s. So now I have our natural response right here, and in this square we have our forced response, the forced response is just a constant v s. So now we're ready to come up with our total response, and we'll call that v capital t, the total response is equal to the natural response plus the forced response. So now we're using our principal of super position. Okay, v t equals k some constant times e to the minus t over r c. And let's add the forced response, and the forced response is right here, it's v s, and we're getting close, the only thing we have left is to figure out this value now, we have to figure out the gain factor in front of the exponential term of the natural response. Now the way we go after that, is we would know this if we knew what v s is, and we do, it's called v s. If we knew v t at some time, we could plug in two values here. We could plug in a v t and a t, and one of the most convenient times to know this, if we set t equal to zero, and we know that the total at time equals zero is what? It's basically, let me roll it up here a little bit, let's go backwards a little bit. Let me go back here and use this diagram again, this was the forced response, so we're not gonna use that right now. We're gonna use, this is now the total circuit all assembled together, and we have to figure out what is the actual initial charge on here, and if we recall, the initial charge on here was v naught, it was this value here, just before time equals zero, this was the value of the voltage on that capacitor. So I can fill that in here, v naught. Alright, let's go back to our total solution and plug in these two values, time is zero and v t equals v naught. So that looks like v naught equals natural constant times e to the minus zero over r c plus v s. And let's solve for k n k n equals, this term is e to the zero is one, so this term goes to one, that says that it's k n plus v s, is this side, and I can write down here v naught minus v s. So now I've solved for k, and we can finish our total response, we can say the total response is v t equals k sub n, which is v naught minus v s, e to the minus t over r c and plus v s. And that is the total response of our circuit, and we saw that in two steps, first we did the natural and then we did the forced and we added them together and worked out the last constant, whatever the constant was, and there's our answer. So in the next video we'll do an explicit example with values for r and c in the step and we'll see what it actually looks like.