# Euler's Equation and KVL

1. Feb 13, 2013

### iRaid

In one of my engineering classes we discussed these two topics and I have two questions about this stuff.

First question is how does euler's equation work exactly..
$$e^{j\varphi}=cos\varphi+jsin\varphi$$

Second question is how do you solve this:
$$V_{M}cos\omega t=Ri(t)+L\frac{di(t)}{dt}$$

Thanks for the help..
(These aren't homework problems, just notes I do not understand)

2. Feb 13, 2013

### Whovian

I, unfortunately, don't understand the question. What do you mean by "how?" (Note that I think you mean i, not j.)

And, to your second, what are Ri(t) and i(t)? Or $V_M$? I might be being an idiot and missing something.

3. Feb 13, 2013

### mathman

Euler's equation is an identity. What do have in mind with the question "How does it work?"?

4. Feb 13, 2013

### I like Serena

Hi iRaid!

Euler's equation is more than an equation - it's an identitiy.
It is always true.
It's not so easy to explain why it is true, but there are proofs lying around.
For instance here.

For your 2nd question, it's an inhomogeneous first order ODE.

The way to solve it is one of:
- method of undetermined coefficients
- with a standard formula for it
- variation of parameters

@Whovian: in electrical engineering, the function $i(t)$ is often used to define an input signal, usually a current.
That makes the use of the symbol $i$ for the imaginary constant ambiguous.
That's why the symbol $j$ is used for the imaginary constant.
The expression $R i(t)$ is Ohm's law ($\Delta V = I \times R$).

Last edited: Feb 13, 2013
5. Feb 13, 2013

### iRaid

Ah I remember euler's formula now, the j is what got me..
Well I don't know ODE's so this sucks lol.

6. Feb 13, 2013

### I like Serena

Ah well, let's just say that in your case the general solution is of the form (method of undetermined coefficients):
$$i(t) = C_0 e^{-\frac R L t} + K \cos ωt + M \sin ωt$$

Can you find the constants $C_0, K, M$?

7. Feb 13, 2013

### I like Serena

Oh, and since we're not in electrical engineering here, let me add that KVL is the abbreviation for Kirchhoff's Voltage Law, which states that the sum of voltages in any electrical loop must be zero. The 2nd equation sort of represents this for a specific situation.

8. Feb 13, 2013

### SteamKing

Staff Emeritus
Remember, electrical engineers use j = sqrt(-1) in place of i to avoid confusion with current.

9. Feb 13, 2013

### iRaid

If they are given I can lol. I'm not sure what you mean.

10. Feb 13, 2013

### joeblow

The imaginary unit is also symbolized as j in physics. I believe that they denote it that way because it is consistent with the quaternions.

You can take the expedient yet historically dishonest path of defining exp(j*x)= cos x + j sin x. If you want to go the more honest path, consider the movement about the unit circle with a constant arc length speed and do some diff eq. with the IC that exp(0)=1.

11. Feb 13, 2013

### iRaid

Yeah I see what you're saying. I knew that j was the imaginary unit, but when I saw that equation, I forgot that or something, I'm not sure lol it just screwed me up. And as stated above the j is used as i in EE so that it isn't confused with current as current is used im many equations.

12. Feb 13, 2013

### I like Serena

When solving a differential equation, we're usually left with a general solution that has 1 or more as yet unknown constants.
When you fill in the general solution in the differential equation, you should find that some of those constants will have to have a specific value to satisfy the equation (for all values of t).
The question would then become: what do you get if you fill in this general solution in your equation?
(Only if you're interested of course.)

13. Feb 13, 2013

### iRaid

Doing di(t)/dt it would be:
$$C_{0}=\frac{-R}{L}\\K=-\omega\\M=\omega$$

I believe?

14. Feb 13, 2013

### I like Serena

Let's see...
$${di(t) \over dt} = - C_0 \frac R L e^{- \frac R L t} - Kω \cos ωt + Mω \sin ωt$$

So
$$V_M\cos ωt = Ri(t)+L{di(t) \over dt}$$
$$\qquad = R(C_0 e^{- \frac R L t} +K \cos ωt + M \sin ωt) + L(- C_0 \frac R L e^{- \frac R L t} - Kω \sin ωt + Mω \cos ωt)$$
$$\qquad = (RK +LMω) \cos ωt + (RM - KLω) \sin ωt$$

Since it has to be true for any t, we find:
$$\left\{ \begin{array}{l} RK +LMω = V_M \\ RM - KLω = 0 \end{array}\right.$$

From this you can solve K and M.
You still have a free choice for $C_0$, but that choice follows if you know $i(0)=i_0$.

Last edited: Feb 13, 2013
15. Feb 13, 2013

### iRaid

I see thank you for the clarification.