Waveform Differentiation

A waveform is a function that repeats itself or cycles after a specified period of time. This time, denoted T, is called the period of the waveform.

The diagram to the right shows an electronic circuit that takes as input a voltage waveform and gives as output an approximation of the derivative of that
waveform. The approximation can be made quite good by using small enough values of R and C or restricting ourselves to slowly varying input voltages.

To calculate the derivative of a waveform it is sufficient to calculate the derivative of just the first cycle. This is because the derivative of a waveform is
another waveform. The derivative of all the other cycles will look the same as that of the first cycle. If f (t) is the formula for the first cycle then f ′ (t), its
derivative, is the formula for the derivative of the first cycle. If required, the formula for the derivative of any other cycle can be found by simply applying a
shift to the formula for the derivative of the first cycle.

Problem
Calculate and sketch the derivative of the waveform shown to the right, whose period is 2 seconds and whose first cycle is defined as:
Solution
We begin by calculating the derivative, F′(t), of the above formula:

This is the formula for the first cycle of the derivative. Any other cycle looks the same as the first one. A sketch of this derivative is shown below.

Maximum Power Transfer in a Circuit

Problem
A power supply can be represented by an ideal voltage source of E volts in series with an internal resistance Rint. If a load is connected to the power supply,
show that the maximum power that can be supplied to the load is achieved when the resistance R of the load is chosen to equal the internal resistance Rint of the
power supply.

Solution
We will express the power P dissipated in the resistor R as a function of R and then find where this function has its maximum. The power P dissipated in a
resistor R is given by the formula P = I 2 R, where I is the current flowing through the resistor. Expressing the current I as a function of R (using Ohm’s law)
gives:

Now we can express the power P as a function of R:

Notice that this function is zero when R is zero, is positive for all positive values of R, and decreases like 1/R when R is very large. Clearly P must have a
maximum. Now calculate the derivative dP/dR:
This derivative is equal to zero when R = Rint. From the above discussion it is clear that the derivative is zero here because this is the value of the resistance R at
which the power P is a maximum (as opposed to say a minimum).

Background

The average or mean of a set of numbers is found by adding up the numbers and dividing by the number of numbers. The mean value of a function is the
generalization of this idea. The mean value of any function f(x) is defined as the mean height of the function. It can be found by calculating the area under the
function and dividing by the width of the function. We will denote the mean value of a function f using angle brackets like this:

Then using calculus notation the above statement can be written as:

(We are assuming that the function runs from a to b, so integrating from a to b gives the area under the function and dividing by b - a divides by the width of the
function.)

If the function is periodic (i.e. if the function repeats itself exactly after a certain amount of time, called the period, T) then we can replace the above formula by this
formula:

In other words we only have to find the area for one cycle and divide by the width of one cycle.
The RMS value of a function is related to the idea of the standard deviation of a set of numbers. The RMS value of a function is defined as the square root of the
mean value of the square of a function. In other words to get the RMS value of a function we follow these three steps:

Square the function (i.e. draw a new function whose height everywhere is the square of the original function),
calculate the Mean of this new function (This can be done using the above formula, namely, finding the area under the new function, and dividing by the width of
the function),
taking the square Root
We will denote the RMS value of a function f like this:

Then using calculus notation the above procedure can be written as:

(Again we are assuming that the function runs from a to b, so integrating from a to b gives the area under the squared function and dividing by b - a gives the
mean of the squared function.)

If the function is periodic with period T then we can replace the above formula by this formula:

Problem

Find the mean value and RMS value of the triangular waveform shown to the right. It represents the current i flowing in a circuit as a function of the time, t.
Solution

The mean value of the current. Use the formula:

The period is T = 4. The integral from 0 to 4 must be broken into 3 separate integrals because the function has a different form in each section. The result is:
Thus the mean value of the current is zero. We could have gotten this result much more easily using symmetry rather than integration. The picture to the right
shows the "area under the curve" shaded in grey. It is clear that the area from 0 to 2 is positive and the area from 2 to 4 is negative and that the two areas cancel
out.

The RMS value of the current. Use the formula:

The lower picture shows the area under the i2 curve shaded in grey. Notice that the i2 curve has the period T = 2. In fact, notice that the area from 0 to 1 and the
area from 1 to 2 are equal. Thus we will save work if we find the area from 0 to 1 and double our answer.

The picture to the right shows the region from t=0 to t=1 enlarged. Notice that in this region i = 10t, so i2 = 100t2
The result is:

Notice that the final answer is the peak value of the current (namely 10) divided by the square root of 3. In fact the RMS value for triangular waveforms with
triangles of any shape turns out to always be the peak value divided by the square root of 3.

Problem:

Find the current in the circuit to the right after time t=0. The switch is closed at t=0, and the initial current in the circuit is zero.

Solution:

Using Kirchoff’s Voltage Law:

vL + vR = E
Express the voltage equation in terms of current.

Put the above equation in differential form.

L di = (E-Ri) dt

Separate the variables.

Integrate both sides of the equation.

Solve for the current i. First isolate the logarithm.

Express in exponential form.

Substitute the initial conditions into this equation to evaluate the constant A. The initial conditions are i=0 at t=0.

Thus the equation becomes

Evaluate the expression for the values given in the circuit diagram.

A graph of this function is shown below. Note that if the inductor were absent the current would rise instantly to 5 amperes, the value given by ohms law, i=E/R.
The presence of the inductor causes the current to rise exponentially to 5 amperes, with the characteristic time being 1/2000 s.