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

Sketching exponentials - examples

Sketch exponentials with different time constants. Created by Willy McAllister.

Want to join the conversation?

  • eggleston blue style avatar for user dena escot
    @ "that this waveform is at risk of not making it
    Current transcript segment all the way down to the final value. Before it's asked to turn up and go around the other way" why? due to the time constant?
    (2 votes)
    Default Khan Academy avatar avatar for user
  • blobby green style avatar for user Suad
    I didn't understand at minutes of the video. What do you mean by saying, "Before it's asked to turn up and go around the other way"
    (1 vote)
    Default Khan Academy avatar avatar for user
    • old spice man green style avatar for user Willy McAllister
      What we see here is the aqua output signal, v_c, rising toward its potential final value, 1V, but not making it. The orange input signal, v_s, transitions from 1V to 0V before the output voltage completes its exponential rise. That causes the rising waveform to reverse direction and become a falling waveform.

      So what I meant by "it's asked to turn around" is that the input v_s switched states from high to low prior to the output v_c reaching its final state.
      (2 votes)
  • blobby green style avatar for user aandras
    In the second example, what would happen if we draw out more periods?

    What I mean is: at t=0.02 s, the voltage did not make it "all the way up". Thus, since now it starts a little lower than at t=0s, does it mean that in the next period it would make it "all the way down", until zero at t=0.03s?

    I'm not sure whether it would make it all the way down to zero, but my intuition says, it would for sure go a bit lower than it went at t=0.01s.

    In that case, at the next peak (t=0.04s) it would be even a bit lower than it was at t=0.02s. And so on, the peaks would be gradually lower.

    I guess there would be some limit value (maybe zero, maybe something else) to which the peaks at k*0.02s "converge"?

    Or would, at some point the voltage "bounce" somehow and get back to 1.0 V and the cycle start all over again?
    (1 vote)
    Default Khan Academy avatar avatar for user
  • female robot grace style avatar for user Agneta Rosenheck
    Sorry, again is the same video as the previous, under a different title?
    (0 votes)
    Default Khan Academy avatar avatar for user

Video transcript

- [Voiceover] Now we're gonna take the ideas from the last video and learn how to sketch in these exponentials really rapidly. Now I gonna move this up and we'll do a couple of examples. Here's an example circuit I have already set up. It's an RC circuit, this is 1000 ohms, and this is two microfarad capacitor. And this voltage source provides a step voltage to us. It's gonna start out at one volt, and then step rapidly down to zero volts at times equals zero. And then again at ten milliseconds, it'll step up again to one volt and continue that way. This can be a clock in a circuit, in a digital circuit, or it can be any kind of a digital signal. What we want to know, is we want to find this voltage right here. What does that look like? We're gonna use what we know about RC waveforms to tell us what's gonna happen here, really quickly. We don't need to use a computer to simulate this. So, first thing I need to know, is what's the RC time constants? RC equals one thousand, one K, times two microfarads, and that equals two K times, K is plus three, micro is minus six, so this is two times ten to the minus three or two milliseconds. So I can actually go over here to my chart and I can put a spot right here at two milliseconds. Right there, there's five milliseconds, there's five, there's about two and rapidly what I know, is that that line is gonna look roughly, like this. It's gonna start, of course, it's gonna start right here and that line's gonna go right through here. So, let's see if I can sketch that in. That's about what that's gonna look like. Now, on the other side, when it goes up, it's gonna start here and it's gonna go up to where? It's gonna go two more milliseconds. One, two. It's about two more milliseconds. That's the final value of that voltage. And I can draw another line here like this, I can sketch that in. And the same thing's gonna happen over here. As soon as that goes down, I go two milliseconds over. It's gonna go right through about there. Start here. And down I go. Now, let me get a little more exact. What I want to do now is do a little better job sketching in this exponential curve. And we know that it's about 37 percent, right about there. Here's about 37 percent of the starting voltage, right there. So that now I just use my sketching skills and I just smooth in some kind of curve that looks like that. And on the other side, I come down 37 percent, which is roughly right about there. And I can draw, again, I can just sketch this in. And that'll come up to some value there, finishing value. And the same thing here, I go 37 percent up, the exponential's gonna go through that. So it's gonna come down and then curve off like this. And that's gonna be my estimate. Now, I didn't use a computer to do that. How do we know if we got it right? Well, I did use a computer once, so here's what it's gonna look like. Let me turn that on. And we'll move it up a little bit. And we can compare what we got. I simulated this in Excel, just using the equations to get a really precise answer. And what you notice, is our sketch looks pretty good, it looks pretty close to that. So what we're comparing, is compare this, to that. And that's pretty similar, that's not bad. And that tells me that our output signal, our voltage on the capacitor, is following the digital signal, fairly well. And it's reaching its high level. And it's rapidly coming down as fast as it can. So, let me do one more, let me do one more problem and we'll change the capacitor value. Now, the change here is four microferads. This is still one K. This is four microferads. So, RC, let's do it again real fast. RC equals one K times four microfarads and that equals four milliseconds. So now I know that the time constant is four milliseconds. That means that I know that straight line sketch, here's five milliseconds, it goes through right about here. I start up here. I can sketch in a straight line, like that. It goes roughly like that. Same thing over here. I can go over another four milliseconds. Up like that. And sketch in my slope. And we'll do the third one over here. Here's four milliseconds after the change, is right about there. And I sketch the line in. And now we'll go back and do our 37 percent trick. Which is 37 percent, is right about there. And we'll come up, it goes through right there. And if I sketch it in now, what's gonna happen is, it's gonna go through that and something over there's gonna happen. I don't know if it's gonna finish. It might, it might not. This one's gonna go up, same way. Let's do 37 percent down. It's about there. Accurate enough. And we'll sketch in this line. And then we'll do the same thing over here, up 37 percent. And down we go. Now, let me show you the answer that I calculated with the computer. We'll see how that looks. Let's move it up. And we did pretty well. Notice how down here, notice how down in this area here, we sort of got it, that's pretty accurate. I missed a little bit out here. That's okay, that's okay. But with that time constant, we can notice right away that this waveform is at risk of not making it all the way down to the final value. Before it's asked to turn up and go around the other way. So that kind of intuition, that kind of understanding of an exponential curve, really makes you very quick. You can do this manually. And in your head, you can sketch out what these exponentials look like. The time point here, this line right here, that's RC seconds after it drops. That's the basic thing. And the other one is this point right here. That's about 37 percent of where it started from. Thirty seven percent to go. So that's just a nice little skill to have when you're looking at exponentials, which you do all the time if you're designing computers. Or any other kind of circuit that has sudden changes in voltage like we had here.