Physics 5B

Winter 2009

How to add sine functions of dierent amplitude and phase

In these notes, I will show you how to add two sinusoidal waves, each of dierent amplitude and phase, to get a third sinusoidal wave. That is, we wish to show that given E1 = E10 sin t , E2 = E20 sin(t + ) , the sum E E1 + E2 can be written in the form: E = E10 sin t + E20 sin(t + ) = E0 sin(t + ) (3) (1) (2)

where the amplitude E0 and phase are determined in terms of E10 , E20 and . In these notes, we shall derive that the amplitude E0 is given by E0 =
2 2 E10 + E20 + 2E10 E20 cos

and the phase is determined modulo 2 by1 sin = E20 sin , E0 cos = E10 + E20 cos . E0

By denition, the amplitudes E10 and E20 are positive numbers. If we divide the last two equations, then can be determined modulo from: tan = E20 sin . E10 + E20 cos

To determine modulo 2, we need to supplement the result for tan with sign(sin ) = sign(sin ) , sin = 0 ,

where sign(sin ) literally means the sign (i.e. either +1 or 1) of the quantity sin . Finally, if sin = 0 then cos = 1, and can be xed modulo 2 by 1 , cos = 1 , cos = 1 , cos = 1 , and E10 > E20 , 1 , cos = 1 , and E10 < E20 .

Note that in the case of cos = 1 and E10 = E20 , we have E = 0 in which case E0 = 0 and is no longer meaningful. In fact, this is the only circumstance in which E0 can vanish. I shall provide two dierent derivations of the above formulae. Finally, in an appendix, I will provide a mathematically more advanced derivation that makes use of complex numbers. If you are unfamiliar with complex numbers, you can skip the appendix for now and return to this last derivation later after you take Physics 116A or an equivalent course. This last method also provides the real motivation for the method of phasors introduced in Section 2 below.

1 The phrase, is determined modulo 2, means that is determined up to an additive integer multiple of 2. This is all that is needed, since adding a multiple of 2 to the phase angle does not change the value of sin(t + ).

1. Algebraic method First, set t = 0 in eq. (3) to obtain E20 sin = E0 sin . Solving for sin yields: sin = E20 sin . E0

Next, set t = /2 in eq.(3). Noting that sin( + /2) = cos , it follows that E10 + E20 cos = E0 cos . Solving for cos yields: cos = E10 + E20 cos . E0

Finally, using cos2 + sin2 = 1, and inserting the expressions for cos and sin just obtained, one nds:
2 2 E0 = (E10 + E20 cos )2 + E20 sin2 2 2 = E10 + 2E10 E20 cos + E20 (cos2 + sin2 ) 2 2 = E10 + E20 + 2E10 E20 cos .

By denition, E0 is a non-negative number.2 Thus, we take the positive square root to obtain E0 = This completes our derivation.
2 2 E10 + E20 + 2E10 E20 cos .

2. Geometric methodthe method of phasors Consider a ctitious vector in a two-dimensional space, whose length is E0 , which makes an angle with respect to the x-axis as shown below: y

E0 sin

E0

If we project this vector onto the y-axis, then its projected length is E0 sin as shown above. This vector is called a phasor and represents a quantity with an amplitude E0 and an angle . The utility of such a representation is that we can perform the sum of eq. (3) by considering the phasors corresponding to each sine term in the sum, and then adding the phasors vectorially! The projection of the vector sum of the two phasors onto the y-axis is just the sum of the two sine functions that we wish to compute. This vector sum can be carried out geometrically, and provides a second method for evaluating E0 and .
2 Warning!

This is a matter of convention, which Giancoli chooses not to follow (without warning you).

To see how this works, consider the computation of eq. (3) by the method of phasors. We represent E1 and E2 [cf. eqs. (1) and (2)] as shown in the gure below.

Then, the phasor representation of E is just the vector sum shown above. We identify E10 , E20 and E0 , as the lengths of the phasors representing E1 , E2 and E , respectively. To evaluate E0 and , we focus on the triangle in the gure above. First, using the law of cosines,
2 2 2 E0 = E10 + E20 2E10 E20 cos( ) ,

since is the angle between the phasors representing E1 and E2 . Using cos( ) = cos , we end up with E0 = as before. Next, using the law of sines, sin( ) sin = . E20 E0 Using sin( ) = sin , we can solve for sin . We nd that sin = E20 sin , E0
2 2 E10 + E20 + 2E10 E20 cos ,

which again agrees with our previous result. This equation only xes modulo . In order to x modulo 2, we employ the law of sines again (noting that the angle in the triangle between the phasors representing E2 and E is given by ): sin( ) sin = . E20 E10 This equation can be rearranged in the following form: E10 sin( ) = E20 E10 = = sin cos cos sin sin sin cos . tan 3

One can solve this equation easily for tan to obtain tan = E20 sin , E10 + E20 cos

in agreement with our previous result. Finally, we can use our results above for sin and tan to compute cos as follows cos = sin = tan E20 sin E0 E10 + E20 cos E20 sin = E10 + E20 cos , E0

which completes the derivation.

3. The limit of equal amplitudes As a check, consider the case of equal amplitudes, E10 = E20 E0 . Then, using the above results, E0 = 2E0 (1 + cos ) . Recalling the trigonometric identity, cos2 (/2) = 1 (1 + cos ), we end up with: 2 E0 = 2E0 | cos(/2)| . Note the absolute value sign, since by denition the amplitude E0 is dened to be non-negative, which means we must take the positive square root: cos2 (/2) = | cos(/2)|. If cos = 1, then E0 = 0 and the angle is undened. Otherwise, we may use the results derived above for sin and cos to obtain. sin 2 sin(/2) cos(/2) sin(/2) , if cos(/2) > 0 , sin = = = sin(/2) , if cos(/2) < 0 , 2| cos(/2)| 2| cos(/2)| cos = cos2 (/2) 1 + cos = = | cos(/2)| , 2| cos(/2)| | cos(/2)|
1 2

after using the trigonometric identity sin = 2 sin(/2) cos(/2). Note that cos(/2) = (1 + cos ) = 0 , for cos = 1 ,

which must hold if E0 = 0. The results above for sin and cos imply that: = /2 , if cos(/2) > 0 , + /2 , if cos(/2) < 0 .

In Section 34-4 of Giancoli, a convention is chosen in which E0 can be of either sign, so in order to compare Giancolis results with ours, one must be careful if cos(/2) < 0. Nevertheless, independently of this convention, we can write: E = E0 sin(t + ) = 2E0 | cos(/2)| sin(t + ) = 2E0 | cos(/2)|(sin t cos + cos t sin ) cos2 (/2) sin(/2) cos(/2) = 2E0 | cos(/2)| sin t + cos t | cos(/2)| | cos(/2)| = 2E0 cos(/2) (sin t cos(/2) + cos t sin(/2)) = 2E0 cos(/2) sin(t + /2) , which agrees with Eq. 34-5c of Giancoli (on page 907). 4

Appendix: Adding two sine functions of dierent amplitude and phase using complex numbers To perform the sum: E = E10 sin t + E20 sin(t + ) = E0 sin(t + ) , we note the famous Euler formula: ei = cos + i sin . In particular, sin is the imaginary part of ei . Thus, if we consider the equation: E10 eit + E20 ei(t+) = E0 ei(t+) , (5) (4)

then the imaginary part of this equation coincides with eq. (4). By the way, I can view a complex number x + iy as a vector in a two-dimensional space (called the complex plane) that points from the origin to the point (x, y). This vector is precisely the phasor that we employed in Section 2 of these notes. In particular, in this language, eq. (5) describes the sum of two complex numbers, which is depicted by the sum of the phasors in the gure shown in Section 2. The projected lengths of the phasors on the y-axis simply correspond to the imaginary parts of the corresponding complex numbers. Thus, to solve for E0 and , we can simply start from eq. (5). If I multiply this equation by eit , I obtain: E10 + E20 ei = E0 ei (6) where E10 , E20 and E0 are non-negative (and real) by denition. To compute E0 , I simply take the complex absolute value of both sides of this equation. For any complex number z = x + iy, the complex absolute value is given by |z| = x2 + y 2 . Hence,
2 E0 = |E10 + E20 ei |2 = |E10 + E20 cos + iE20 sin |2 2 = (E10 + E20 cos )2 + E20 sin2 2 2 = E10 + 2E10 E20 cos + E20 ,

after using sin2 + cos2 = 1. Taking the positive square roots yields E0 =
2 2 E10 + E20 + 2E10 E20 cos .

To determine , I simply take the real and imaginary parts of eq. (6): E0 cos = E10 + E20 cos , E0 sin = E20 sin ,

Solving for cos and sin , respectively, we immediately nd: cos = sin = E10 + E20 cos , E0 E20 sin . E0

Thus, we have successfully reproduced the main results obtained previously in these notes. I think you must agree that this last approach is by far the simplest from a computational point of view. Once you learn how use and manipulate complex numbers, many tasks in mathematical physics become much simpler!