I'm studying for my circuit analysis exam and I stumbled upon a problem I can't understand. Here it is:

enter image description here

The voltage source is 0V for t<0s and 10V for t>0s, and I have to find the Vout voltage formula for t>0s.

I can't understand what happens as soon as the voltage source gets turned on. I calculated the final Vout voltage using KVL (Vout = -10V, assuming that the voltage in all the capacitors is the same and no current flows through the resistor after a while), so my result would be:

$$ Vout = 10(e^{\frac{-t}{τ}}-1) [V] $$

I know the result is correct but can anyone explain the logic behind this circuit? Also how do I calculate the time constant τ ?


  • 1
    \$\begingroup\$ Start by assuming the input RC is tied to ground. What is the input current? (hint, it's an inverse exponential, just like the capacitor voltage - and for the same reason). Once you know the current into the - input, you can calculate the output voltage by integration (which is what the feedback cap will provide). \$\endgroup\$ – WhatRoughBeast Jan 4 '17 at 2:00
  • 1
    \$\begingroup\$ It's not a real life circuit - there are vital components missing that would make it a practical circuit. \$\endgroup\$ – Andy aka Jan 4 '17 at 8:32
  • \$\begingroup\$ Do you need to solve it in the time domain, or can you use Laplace transform techniques? \$\endgroup\$ – Matt L. Jan 4 '17 at 11:16
  • \$\begingroup\$ I need to solve it in the time domain, we haven't learned Laplace techniques yet. I tried with @WhatRoughBeast method but I get a wrong result :( \$\endgroup\$ – squeck Jan 4 '17 at 12:29
  • \$\begingroup\$ So, what exactly did you get for the current, and what exactly did you get for the output voltage? "I get a wrong result" is not remotely informative. Edit your answer to show what you did and why. \$\endgroup\$ – WhatRoughBeast Jan 4 '17 at 13:35

Note that you have an ideal op-amp with zero current into the inputs and with zero voltage between the inputs. Assuming a current \$i(t)\$ through the feedback capacitor (from right to left), we get

$$i(t)=C\frac{dv_o(t)}{dt}\quad\text{and}\quad v_o(t)=\frac{1}{C}\int i(t)dt\tag{1}$$

Since no current flows in or out of the inverting input of the op-amp, the same current also flows through the series connection of \$R\$ and \$C\$ at the input:

$$v_i(t)+Ri(t)+\frac{1}{C}\int i(t)dt=0\tag{2}$$

Plugging \$(1)\$ into \$(2)\$ we get


Using the time constant \$\tau=RC\$, \$(3)\$ can be rewritten as


The homogeneous solution of \$(4)\$ is

$$v_{o,h}(t)=Ae^{-t/\tau},\quad t>0\tag{5}$$

(with some constant \$A\$ to be determined) and, since \$v_i(t)\$ is constant for \$t>0\$ (equal to \$v_i(0^+)=10V\$), a particular solution is

$$v_{o,p}(t)=-v_i(0^+),\quad t>0\tag{6}$$

The total solution is the sum of the homogeneous and the particular solution:

$$v_o(t)=Ae^{-t/\tau}-v_i(0^+),\quad t>0\tag{7}$$

The constant \$A\$ must be determined from the initial condition \$v_o(0^+)=0\$:

$$v_o(0^+)=A-v_i(0^+)=0\quad\Longrightarrow\quad A=v_i(0^+)\tag{8}$$

which finally gives us the complete solution

$$v_o(t)=v_i(0^+)(e^{-t/\tau}-1),\quad t>0\tag{9}$$

  • \$\begingroup\$ Thanks a lot! This is very helpful :) However, there's a thing I don't understand. Since the inverting input of the op-amp is 0V, I would assume the voltage drop across the resistor and the input capacitor to be opposite in sign than Vout and the voltage drop across the feedback capacitor. I would have written formula (2) like this $$v_i(t)-Ri(t)-\frac{1}{C}\int i(t)dt=0\tag{2}$$ but this looks strange because the voltage drops don't follow the sign of the i(t) current. Drawing here: i.imgur.com/SdcxO9i.jpg Can you explain what's going on? Thank you so much. \$\endgroup\$ – squeck Jan 4 '17 at 17:12
  • \$\begingroup\$ @squeck: I defined the current to flow from right to left (just like in your drawing), but then the voltages are in the same direction as the input voltage. Your + and - signs for the voltages across the resistor and the capacitor at the input are the wrong way around. \$\endgroup\$ – Matt L. Jan 4 '17 at 20:29

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.