Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual 205

Chapter 23. Fourier series

Exercises 23.2
1. (a) f (t) = 3 sin 100πt − 4 sin 200πt + 0.7 sin 300πt. The fundamental has frequency 50 Hz and amplitude 3.
The second and third harmonics have frequencies 100 Hz and 150 Hz and amplitudes 4 and 0.7 respectively.
(b) f (t) = sin 40t − 0.5 cos 120t + 0.3 cos 240t. The fundamental has frequency 20/π Hz and amplitude 1. The
second, fourth and fifth harmonics are missing. The third has frequency 60/π Hz and amplitude 0.5, and the
sixth has frequency 120/π Hz and amplitude 0.3.
2. (a)
f (t) = 2 cos t − 3 sin t = R cos(t + α) = R cos t cos α − R sin t sin α
Therefore
R cos α = 2 and R sin α = 3
√ √
Hence R = √ 13 and tan α = 3/2, i.e. α = 0.983. Hence f (t) = 2 cos t − 3 sin t = 13 cos(t + 0.983). Here the
amplitude is 13 and the phase is 0.983.

(b)
f (t) = 0.5 cos t + 3.2 sin t = R cos(t − α) = R cos t cos α + R sin t sin α
√
Hence R cos α = 0.5 and R sin α = 3.2. Hence R = 0.52 + 3.22 = 3.24 and tan α = 3.2/0.5 = 6.4, i.e. α = 1.42.
Hence f (t) = 0.5 cos t + 3.2 sin t = 3.24 cos(t − 1.42). Here the amplitude is 3.24 and the phase is −1.42.
Alternatively we could write f (t) = 3.24 cos(t + 2π − 1.42) = 3.24 cos(t + 4.867).

(c) f (t) = 3 cos 3t has amplitude 3 and zero phase.

(d)
f (t) = 2 cos 2t + 3 sin 2t = R cos(2t − α) = R cos 2t cos α + R sin 2t sin α
√
Hence R cos α = 2√and R sin α = 3 so that R = 13 and √tan α = 3/2, i.e. α = 0.983. Hence f (t) =
2 cos 2t + 3 sin 2t = 13 cos(2t − 0.983). Here the amplitude is 13 and the phase is −0.98.

3.
(a)
1

t
-5 -3 -1 1 3 5
(b)
f (t)

π π t
−π − π
2 2

Pearson
c Education Limited 2001
206 Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual

(c)
2

t
-3 -1 1 3

4.
(a)


1 0≤t≤1
f (t) = Period 2

0 1<t<2

(b)

 2t 0 ≤ t ≤ 1/2
f (t) = Period 1

2 − 2t 1/2 ≤ t ≤ 1

(c)


 0 0≤t≤1
f (t) = Period 3

0.5t − 0.5 1≤t<3

Exercises 23.3
1. (a) odd, (b) neither, (c) odd, (d) even.

2. (a) even, (b) odd, (c) even, (d) odd, (e) neither.
5
3. (a) t3 is odd. Hence −5
t3 dt = 0.
5
(b) t3 cos 3t is odd. Hence −5 t3 cos 3t dt = 0.
Similarly in parts (c) and (d) the integrands are odd and the integrals are therefore 0.
1
(e) |t| is even. Hence we require 2 0 t dt = 1.
(f) t|t| is odd and the required integral is zero.

Exercises 23.5
1. Here T = 10.

Pearson
c Education Limited 2001
Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual 207

 5
1 nπt
an = dt cos
0 5 5
 5
1 sin nπt/5
=
5 nπ/5 0
= 0

 5
1 1 5
a0 = 1 dt = [t] = 1
5 0 5 0

 5
1 nπt
bn = sin dt
5 0 5
 5
1 − cos nπt/5
=
5 nπ/5 0
1
= (− cos nπ + 1)
nπ
Hence b1 = 2/π, b2 = 0, b3 = 2/3π, etc.
1 2 πt 2 3πt
f (t) = + sin + sin + ...
2 π 5 3π 5
2. Here T = 2π.

 0
1
an = (−t) cos nt dt
π −π
 0  0

1 −t sin nt sin nt
= + dt
π n −π −π n
 0
1 cos nt
= − 2
π n −π
 
1 1 cos nπ
= − 2+
π n n2

− πn2 2 , n odd
=
0, n even

 0  2 0
1 1 t π
a0 = −t dt = − =
π −π π 2 −π 2

 0
1
bn = −t sin nt dt
π −π
 0 

1 −t(− cos nt) 0

cos nt
= − dt
π n −π −π n

Pearson
c Education Limited 2001
208 Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual

cos nπ
=
n
(−1)n
=
n
Then a1 = −2/π, a2 = 0, a3 = −2/9π, . . ., b1 = −1, b2 = 1/2, b3 = −1/3, . . ..

π 2 1 2 1
f (t) = − cos t − sin t + sin 2t − cos 3t − sin 3t + . . .
4 π 2 9π 3
3. Here T = 2π.

 π 
1 2
an = (t + πt) cos nt dt
π −π
 π  π

1 sin nt sin nt
= 2
(t + πt) − (2t + π) dt
π n −π −π n
 π  π

1 cos nt 2 cos nt
= (2t + π) 2 − dt
π n −π −π n2
 
1 3π cos nπ π cos nπ
= +
π n2 n2
4 cos nπ
=
n2

 π  π
1 1 t3 πt2 2π 2
a0 = (t2 + πt)dt = + =
π −π π 3 2 −π 3

 π 
1
bn = (t2 + πt) sin nt dt
π −π
 π  π

1 cos nt cos nt
= −(t + πt)
2
+ (2t + π) dt
π n −π −π n
 π  π

1 −2π 2 cos nπ sin nt sin nt

= + (2t + π) − 2 dt
π n n −π −π n
 π

1 −2π 2 cos nπ cos nt

= +2
π n n2 −π
−2π cos nπ
=
n

Hence a1 = −4, a2 = 1, a3 = −4/9, . . ., b1 = 2π, b2 = −π, b3 = 2π/3, . . ..

π2 4 2π
f (t) = − 4 cos t + 2π sin t + cos 2t − π sin 2t − cos 3t + sin 3t . . .
3 9 3
4. This function is odd and hence an = 0. Here T = 2π.
 0  π 
1
bn = −4 sin nt dt + 4 sin nt dt
π −π 0

Pearson
c Education Limited 2001
Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual 209

 0  π 
1 4 cos nt 4 cos nt
= −
π n −π n 0
8 8 cos nπ
= −
nπ nπ
b1 = 16/π, b2 = 0, b3 = 16/3π, . . ..

16 16
f (t) = sin t + sin 3t + . . .
π 3π
5. Here T = 2.  0  0
a0 = 2 + 2t dt = 2t + t2 −1 = 1
−1

 0
an = (2 + 2t) cos nπt dt
−1
 0  0
(2 + 2t) sin nπt 2 sin nπt
= − dt
nπ −1 −1 nπ
 0
2 cos nπt
=
n2 π 2 −1
2 2
= − 2 2 cos nπ
n2 π 2 n π

 0
bn = (2 + 2t) sin nπt dt
−1
 0 
−(2 + 2t) cos nπt 0
2 cos nπt
= + dt
nπ −1 −1 nπ
2
= −
nπ

a1 = 4/π 2 , a2 = 0, a3 = 4/9π 2 , . . .. b1 = −2/π, b2 = −1/π, b3 = −2/3π, . . ..

1 4 2 1
f (t) = + cos πt − sin πt − sin 2πt + . . .
2 π2 π π
6.

π2
a0 =
3

2 cos nπ
an =
n2

2 (n2 π 2 − 2)
bn = − 3
− cos nπ
πn n3 π

π2 (π 2 − 4) sin t π sin 2t cos 2t (9π 2 − 4) sin 3t 2 cos 3t

f (t) = + − 2 cos t − + + − + ...
6 π 2 2 27π 9
7. f (t) = 2 sin t.

Pearson
c Education Limited 2001
210 Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual

Exercises 23.6
1. Here T = 2π.
f (t)

3π

−π π t

The periodic extension is even, and hence bn = 0.

  π
2 π 2 3t2
a0 = 3t dt = = 3π
π 0 π 2 0


2 π
an = 3t cos nt dt
π 0
 π  π 
6 t sin nt sin nt
= − dt
π n 0 0 n
 π
6 cos nt
=
π n2 0
6
= (cos nπ − 1)
n2 π

a1 = −12/π, a2 = 0, a3 = −12/9π, . . .. Hence

3π 12 12
f (t) = − cos t − cos 3t . . .
2 π 9π
2. Here T = 2π.
The periodic extension is odd and hence an = 0.


2 π
bn = t sin nt dt
π 0
 π  π 
2 −t cos nt cos nt
= + dt
π n 0 n
0
2 −π cos nπ
=
π n
2 cos nπ
= −
n

b1 = 2, b2 = −1, b3 = 2/3, . . .. Hence

2
f (t) = 2 sin t − sin 2t + sin 3t . . .
3

Pearson
c Education Limited 2001
Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual 211

3. Here T = 2π and the periodic extension is even; bn = 0.


2 π 2
a0 = sin t dt = [− cos t]π0 = 4/π
π 0 π

 π
2
an = sin t cos nt dt
π 0
Let
 π
I= sin t cos nt dt
0

Then

 π  π
sin nt sin nt
I = sin t − cos t dt
n 0 n
 0π
cos t cos nt 1
= 2
+ 2I
n 0 n

Then

1 − cos nπ 1
I 1− 2 = 2
− 2
n n n
i.e.
1
I= (− cos nπ − 1)
n2 − 1
Thus
−2
an = (1 + cos nπ) (n = 1)
(n2 − 1)π

a2 = −4/3π, a3 = 0, a4 = −4/15π . . .. It can be shown that a1 = 0. Hence

2 4 4
f (t) = − cos 2t − cos 4t . . .
π 3π 15π

4. f (t) = et , 0 < t < 1. Here T = 2, and bn = 0.

t
-1 1
 1
a0 = 2 et dt = 2[et ]10 = 2(e − 1)
0

Pearson
c Education Limited 2001
212 Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual

 1
an = 2 et cos nπt dt
0

Let  1
I= et cos nπt dt
0

 1
I = et cos nπt dt
0
 1  1
et sin nπt sin nπt t
= − e dt
nπ 0 0 nπ
 t  1 
−e cos nπt − cos nπt t
= − − e dt
n2 π 2 0 n2 π 2
 t 1
e cos nπt 1
= 2 2
− 2 2I
n π 0 n π

Hence
1
I= (e cos nπ − 1)
1 + n2 π 2
and
2
an = (e cos nπ − 1)
1 + n2 π 2
Finally
∞
 2
f (t) = (e − 1) + (e cos nπ − 1) cos nπt
1
1 + n2 π 2

5. Here T = 4. For the odd periodic extension an = 0.

 2
nπt
bn = (2 − t) sin dt
0 2
 2 
−(2 − t) cos nπt/2 2
cos nπt/2
= − dt
nπ/2 0 0 nπ/2
= 4/nπ

Hence b1 = 4/π, b2 = 2/π, b3 = 4/3π . . .. Finally

4 πt 2 4 3πt
f (t) = sin + sin πt + sin ...
π 2 π 3π 2

Exercises 23.8
1. (a) T = 4.

 2
1
cn = e−jnπt/2 dt
4 0
 2
1 e−jnπt/2
=
4 −jnπ/2 0
j
= (e−jnπ − 1)
2nπ

Pearson
c Education Limited 2001
Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual 213

 2
1 1 2 1
c0 = dt = [t]0 =
4 0 4 2
(b) T = 2.

 1
1
cn = et e−jnπt dt
2 −1
 1
1 e(1−jnπ)t
=
2 1 − jnπ −1
1  
= e cos nπ − e−1 ejnπ
2(1 − jnπ)
(1 + jnπ) cos nπ(e − e−1 )
=
2(1 + n2 π 2 )

(c) T = 2π/ω.
 π/ω
ω
cn = A sin ωt e−jnωt dt
2π 0
 π/ω
ωA
= e(jω−jnω)t − e(−jω−jnω)t dt
4πj 0
 π/ω
ωA e(jω−jnω)t e(−jω−jnω)t
= −
4πj jω − jnω −jω − jnω 0
 
−A cos(1 − n)π 1 cos(1 + n)π 1
= − + −
4π 1−n 1−n 1+n 1+n
 
A 2 cos(1 − n)π 2
= −
4π (n + 1)(n − 1) (n + 1)(n − 1)
−A cos nπ + 1
=
2π (n + 1)(n − 1)

2.
∞

f (t) = cn ej2nπt/T
−∞

 T  ∞
T 
−j2mπt/T
e f (t) dt = cn ej2πt(n−m)/T dt
0 0 −∞
∞
  T
= cn ej2πt(n−m)/T dt
−∞ 0

Now if n = m the right-hand side reduces to zero because n − m = k for some k ∈ Z, and ej2πk = 1. The only
contributing term in the sum is that arising when n = m in which case we have
 T  T
−j2mπt/T
e f (t) dt = cm 1 dt = cm T
0 0
so that  T
1
cm = e−j2mπt/T dt
T 0
as required.

Pearson
c Education Limited 2001
214 Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual

Review Exercises 23
1. T = 2π, an = 0.


2 π
bn = t sin t sin nt dt
π 0

1 π
= t(cos(n − 1)t − cos(n + 1)t)dt
π 0
 π  π 
1 sin(n − 1)t sin(n + 1)t sin(n − 1)t sin(n + 1)t
= t − − − dt
π n−1 n+1 0 0 n−1 n+1
 π
1 cos(n − 1)t cos(n + 1)t
= −
π (n − 1)2 (n + 1)2 0
 
1 cos(n − 1)π 1 cos(n + 1)π 1
= − − +
π (n − 1)2 (n − 1)2 (n + 1)2 (n + 1)2
 
1 − cos nπ cos nπ 1 1
= + − +
π (n − 1)2 (n + 1)2 (n − 1)2 (n + 1)2
−4n cos nπ − 4n
=
π(n − 1)2 (n + 1)2
Hence
∞
 4n(cos nπ + 1)
f (t) = − sin nt
1
π(n − 1)2 (n + 1)2

2.

 π
2
bn = cos 2t sin nt dt
π
0 π
1
= sin(n + 2)t + sin(n − 2)t dt
π 0
 π
1 cos(n + 2)t cos(n − 2)t
= − −
π n+2 n−2
 0

1 cos(n + 2)π − 1 cos(n − 2)π − 1
= − −
π n+2 n−2
 
1 cos nπ − 1 cos nπ − 1
= − −
π n+2 n−2
1 2n cos nπ − 2n
= −
π (n + 2)(n − 2)
2n(cos nπ − 1)
= −
π(n + 2)(n − 2)

b1 = −4/3π, b2 = 0, b3 = 12/5π . . ..

3. (a) an = 0, T = 2τ .

  

τ /4 τ
4 4t nπt 4 t nπt
bn = sin dt + 1− sin dt
2τ 0 τ τ τ /4 3 τ τ

Pearson
c Education Limited 2001
Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual 215

 τ /4  τ /4
2 4t cos nπt/τ cos nπt/τ 4
= − + dt
τ τ nπ/τ 0 0 nπ/τ τ
  τ  τ 

4 t cos nπt/τ cos nπt/τ 4

+ − 1− + − dt
3 τ nπ/τ τ /4 τ /4 nπ/τ 3τ
 τ /4
2 cos nπ/4 4τ
= − + 2 2 sin nπt/τ
τ nπ/τ n π 0
 τ

cos nπ/4 4τ sin nπt/τ

+ + −
nπ/τ 3 n2 π 2 τ /4
32 nπ
= sin
3n2 π 2 4

Then
∞
 32 sin nπ/4 nπt
f (t) = sin
1
3n2 π 2 τ

(b) bn = 0, T = 2τ .

  τ 

τ /4
2 4t 4 t
a0 = dt + 1− dt
τ 0 τ τ /4 3 τ
 τ /4  τ

2 2t2 4 t2
= + t−
τ τ 0 3 2τ τ /4
 
2 τ 2τ 28τ
= + −
τ 8 3 96
= 1

  τ 

τ /4
2 4t nπt 4 t nπt
an = cos dt + 1− cos dt
τ 0 τ τ τ /4 3 τ τ
 τ /4  τ /4
2 4t sin nπt/τ 4 sin nπt/τ
= − dt
τ τ nπ/τ 0 0 τ nπ/τ
  τ  τ 

4 t sin nπt/τ 4 sin nπt/τ 1

+ 1− − − dt
3 τ nπ/τ τ /4 τ /4 3 nπ/τ τ
 τ /4  τ

2 sin nπ/4 4τ cos nπt/τ sin nπ/4 4τ cos nπt/τ

= + − −
τ nπ/4 n2 π 2 0 nπ/τ 3 n2 π 2 τ /4
 
2 16τ cos nπ/4 4τ 4τ cos nπ
= − 2 2−
τ 3n2 π 2 n π 3n2 π 2
32 cos nπ/4 8 8 cos nπ
= − 2 2−
3n2 π 2 n π 3n2 π 2

∞
1  nπt
f (t) = + an cos
2 1
τ

Pearson
c Education Limited 2001
216 Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual

4. The function is even. bn = 0, period = T .

 T /6
2V
a0 = dt
T −T /6
2V T /6
= [t]−T /6
T
2V
=
3

 T /6
2V 2nπt
an = cos dt
T −T /6 T
 T /6
V 2nπt
= sin
nπ T −T /6
2V nπ
= sin
nπ 3

 2V ∞
V nπ 2nπt
f (t) = + sin cos
3 1
nπ 3 T

5. The function is neither odd nor even. Period T = 2π/ω.

 T /2
2
a0 = I sin ωt dt
T 0
 T /2
2I cos ωt
= −
T ω 0
4I
=
ωT
2I
=
π

 T /2
2I
bn = sin ωt sin nωt dt
T 0
= 0 n = 1

If n = 1,
 π/ω
ωI I
b1 = sin2 ωt dt =
π 0 2

 T /2
2I
an = sin ωt cos nωt dt
T 0
 π/ω
ωI
= sin ωt cos nωt dt
π 0
I(cos nπ + 1)
= − n = 1
π(n + 1)(n − 1)

Pearson
c Education Limited 2001
Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual 217

If n = 1 a1 is easily shown to equal 0. Thus

∞
 I(cos nπ + 1)
I I
f (t) = + sin ωt − cos nωt
π 2 2
π(n + 1)(n − 1)

6.
 0.01
1
cn = V e−2nπtj/0.02 dt
0.02 0
 −jnπt/0.01 0.01
V e
=
0.02 −jnπ/0.01 0
V  
= 1 − e−jnπ
2jnπ
V
= (1 − cos nπ)
2jnπ
Thus
∞
 V
f (t) = (1 − cos nπ)e100jnπt
−∞
2jnπ

7. Here f is even and hence bn = 0.

 8
2
a0 = f (t)dt
8 0
 4  8
1 1
= 2 − t dt + t − 6 dt
4 0 4 4
 4  8
1 t2 1 t2
= 2t − + − 6t
4 2 0 4 2 4
1 1
= [0] − [32 − 48 − (8 − 24)]
4 4
= 0

 4 
1 nπt 1 8 nπt
an = (2 − t) cos dt + (t − 6) cos dt
4 4 4 4 4
 0
4  4

1 nπt 4 4 nπt
= (2 − t) sin · + sin dt
4 4 nπ 0 0 nπ 4
 8  8

1 nπt 4 4 nπt
+ (t − 6) sin · − sin dt
4 4 nπ 4 4 nπ 4
 4  8

1 16 nπt 16 nπt
= − 2 2 cos + 2 2 cos
4 n π 4 0 n π 4 4
 
1 16 16 16 16
= − 2 2 cos nπ + 2 2 + 2 2 − 2 2 cos nπ
4 n π n π n π n π
 
1 32 32
= − 2 2 cos nπ
4 n2 π 2 n π
8
= [1 − cos nπ]
n π2
2

Pearson
c Education Limited 2001
218 Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual

8. This result follows immediately from Parseval’s theorem.

9.
∞ 
a0  2nπt 2nπt
f (t) = + an cos + bn sin
2 n=1
T T

Hence
  ∞
  T

T T
a0 f (t)  T
2nπt 2nπt
2
(f (t)) dt = dt + an f (t) cos dt + bn f (t) sin dt
0 0 2 n=1 0 T 0 T
∞  
T a20  T an T bn
= + an + bn
4 n=1
2 2
∞
T a20 T  2
= + a + b2n
4 2 n=1 n

and so
 ∞
2 T
a20  2
(f (t))2 dt = + an + b2n
T 0 2 n=1

as required.
10. Half-range cosine series:
Here bn = 0.

 1
4
a0 = sinh πtdt
2 0
 1
cosh πt
= 2
π 0
2 2
= cos π −
π π
a0 1
= [cosh π − 1]
2 π

To find an consider the integral I defined by


I = sinh πt cos nπtdt

sin nπt sin nπt
= sinh πt − · π cosh πtdt
nπ nπ
  
sin nπt cos nπt cos nπt
= sinh πt − − cosh πt 2 + · π sinh πtdt
nπ n π n2 π
sin nπt cos nπt 1
= sinh πt + cosh πt 2 − 2I
 nπ n π n
1 sin nπt cos nπt
I 1+ 2 = sinh πt + cosh πt 2
n nπ n π
2
 
n sin nπt cos nπt
I = sinh πt + cosh πt 2
1 + n2 nπ n π

Pearson
c Education Limited 2001
Croft,Davison and Hargreaves,Engineering Mathematics,Solutions Manual 219

Then
 1
2n2 sin nπt cos nπt
an = sinh πt + cosh πt
1 + n2 nπ n2 π 0

2n2 cosh π cos nπ 1
= − 2
1 + n2 n2 π n π
2
= ((−1)n cosh π − 1)
π(1 + n2 )

∞
1 2
f (t) = (cosh π − 1) + 2)
((−1)n cosh π − 1) cos nπt
π n=1
π(1 + n

Half-range sine series:

1
an = 0. bn = 2 0 sinh πt sin nπt dt. Consider

I = sinh πt sin nπtdt

cos nπt cos nπt
= − sinh πt + · π cosh πtdt
nπ nπ

cos nπt sin nπt sin nπt
= − sinh πt + cosh πt 2 − · π sinh πt dt
nπ n π n2 π
cos nπt sin nπt 1
= − sinh πt + cosh πt 2 − 2I
 nπ n π n
1 cos nπt sin nπt
I 1+ = − sinh πt + cosh πt 2
n2 nπ n π
2
 
n cos nπt sin nπt
I = − sinh πt + cosh πt
n2 + 1 nπ n2 π

Then
 1
2n2 cos nπt sin nπt
bn = − sinh πt + cosh πt
n2 + 1 nπ n2 π 0

2n2 − sinh π cos nπ
= 2
n +1 nπ
2n sinh π(−1)n
= − 2
n +1 π

∞
 2n
f (t) = − 2)
(−1)n sinh π sin nπt
n=1
π(1 + n

Pearson
c Education Limited 2001