Wave Equation

Kalpana Mahalingam
March 2020

Consider the wave equation : Let u(x, t) denote the vertical dis-
placement of a string from the x axis at position x and time t. The
string has length l. Its left and right hand ends are held fixed at height
zero and we are told its initial configuration and speed. For notational
convenience, choose a coordinate system so that the left hand end of
the string is at x = 0 and the right hand end of the string is at x = l.
We assume that the string is undergoing small amplitude transverse
vibrations so that u(x, t) obeys the wave equation
∂ 2u 2
2∂ u
=c , 0 < x < l, t > 0 (1)
∂t2 ∂x2
• Boundary conditions : The condition that the left and the
right end of the string is held at height 0, is given by
u(0, t) = u(l, t) = 0, t > 0

• Initial Conditions : The position and speed of the string at

time 0, (The functions f (x) and g(x) are given )
u(x, 0) = f (x) , 0 < x < l
ut (x, 0) = g(x) , 0 < x < l

1
K.Mahalingam Wave Equation

Problem : : Determine u(x, t) satisfying the above conditions for all

x and t.

Methods :

1. Separation of Variables

2. D’Alembert’s Method (Using characteristics)

0.1 Separation of Variables

We solve the wave equation (4) subject to the initial and boundary
conditions.

• We find all solutions of the form F (x)G(t), where F depends

only on x and G depends only on t.
P
• If Fi (x)Gi (t) is a solution, then i ai Fi (x)Gi (t) is also a solution
for ai being constants.

• We find constants ai such that the boundary and the initial con-
ditions are satisfied.

Solution :
Step 1 : Finding a general solution : Let u(x, t) = F (x)G(t) be
a solution of the wave equation (4). Then, it satisfies the equation and
we have,
00 00
F (x)G (t) = c2 G(t)F (x)
00 00
F (x) 1 G (t)
=⇒ = 2
F (x) c G(t)
Note that the left hand side is independent of t and the right hand
side is independent of x which means that both ratios are equal to a
K.Mahalingam Wave Equation

constant say k. Thus, we get,

00 00
F (x) 1 G (t)
= 2 =k
F (x) c G(t)
00
F (x)
=⇒ =k
F (x)
and 00
1 G (t)
=k
c2 G(t)
So we get two ordinary ODE’s
00 00
F (x) − kF (x) = 0 , G (t) − c2 kG(t) = 0
We have two homogeneous ODE’s with constant co-efficients. Solving
them we get, √ √
F (x) = A1 e kx + A2 e− kx
and √ √
c kt −c kt
G(t) = A3 e + A4 e
where A1 , A2 , A3 , A4 are all arbitrary constants. If k = 0, then F (x) =
A1 x + A2 and G(t) = A3 t + A4 .
Thus the solution of the wave equation (4) is given by :
√ √ √ √
kx − kx c kt −c kt
u(x, t) = (A1 e + A2 e )(A3 e + A4 e ), f or k 6= 0
and
u(x, t) = (A1 x + A2 )(A3 t + A4 ), k = 0
Step 2 : Applying the boundary conditions The boundary
conditions u(0, t) = u(l, t) = 0 implies that,
F (0) = F (l) = 0
We now check the solutions in Step 1 and consider only the solutions
that satisfy this boundary conditions.
K.Mahalingam Wave Equation

• If k = 0, then u(x, t) = (A1 x + A2 )(A3 t + A4 ) and F (x) =

A1 x + A2 . Applying the condition we get,

F (0) = A2 = 0

F (l) = A1 l + A − 2 = 0
which implies that both A1 and A2 are zero in this case and hence
u(x, t) = 0 which is trivial and hence we omit the value of u(x, t)
when k = 0.
• If k 6= 0, then,
√ √ √ √
kx − kx c kt −c kt
u(x, t) = (A1 e + A2 e )(A3 e + A4 e )
√ √
kx − kx
with F (x) = A1 e + A2 e and applying the boundary con-
ditions we get,

F (0) = A1 + A2 = 0, =⇒ A1 = −A2
√ √
kl − kl
F (l) = A1 e + A2 e =0
Since A1 = −A2 , the above equation becomes,
√ √
A1 (e kl
− e− kl
)=0

If A1 were zero, then F (x) would again be identically zero and

hence useless. So instead, we consider only all k that satisfies the
following:
√ √ √ √ √
e kl
− e− kl
=0 ⇔e kl
= e− kl
⇔ e2 kl
=1

Since k 6= 0, we have
√ √ nπi n2 π 2
2 kl = 2nπi ⇔ k = ⇔k=− 2
l l
K.Mahalingam Wave Equation

2 2
Thus, we have A1 = −A2 and k = − n l2π . SUbstituting these
values we obtain,
nπi nπi cnπi cnπi
u(x, t) = F (x)G(t) = A1 (e l x − e− l x )(A3 e l t + A4 e− l t )
 nπ h  cnπ   cnπ i
= 2iA1 sin x (A3 +A4 ) cos t +i(A3 −A4 ) sin t
l l l
 nπ h  cnπ   cnπ i
= sin x an cos t + bn sin t
l l l
where, an = 2iA1 (A3 + A4 ) and bn = 2A1 (A4 − A3 )

Step 3 :Taking linear combination of solutions from Step 2.

Consider the solutions obtained from Step 2 as
 nπ h  cnπ   cnπ i
un (x, t) = sin x an cos t + bn sin t
l l l
Then, a linear combination of the above is also a solution and hence,
∞
X ∞
X  nπ h  cnπ   cnπ i
∗
u(x, t) = un (x, t) = sin x Bn cos t +Bn sin t
n=1 n=1
l l l

Step 4 : Applying the initial conditions and use Fourier series

expansion
We now know that u(x, t) given in Step 3 satisfies the wave equation
(4) and the initial conditions. We need to find Bn and B∗n such that
u(x, t) satisfies the initial conditions. Thus,
∞
X  nπ 
f (x) = u(x, 0) = Bn sin x (2)
n=1
l

∞
X cnπ  nπ 
g(x) = ut (x, 0) = Bn∗ sin x (3)
n=1
l l
K.Mahalingam Wave Equation

We choose Bn and B∗n such that both u(x, 0) = f (x) and ut (x, 0) =
g(x) (Equations (2) and (3)) satisfy the Fourier sine series. (Refer
Example 17, Sturm-Liouville lecture notes). Hence,
Z l
2 l
Z
1  nπ   nπ 
Bn =   f (x) sin x dx = f (x) sin x dx
|| sin nπ x ||2 0 l l 0 l
l

where, Z l
 nπ 
2 nπ l
 
2
|| sin x || = sin x dx =
l 0 l 2
Similarly,
Z l
cnπ ∗ 1  nπ 
Bn =   f (x) sin x dx
l nπ
|| sin l x ||2 0 l

which implies, Z l
∗ 2  nπ 
B = g(x) sin x dx
cnπ 0 l
Step 5 : General Solution
We now have the general solution of the wave equation (4) satisfy-
ing both the initial as well boundary conditions using separation of
variables method.
X∞  nπ h  cnπ   cnπ i
∗
u(x, t) = sin x Bn cos t + Bn sin t
n=1
l l l

with Z l
2  nπ 
Bn = f (x) sin x dx
l 0 l
and Z l
2  nπ 
Bn∗ = g(x) sin x dx
cnπ 0 l
K.Mahalingam Wave Equation

Step 6 : Special Case : g(x) = 0

Usually, in this method we mainly consider the case when ut (x, 0) =
g(x) = 0 which implies that Bn∗ = 0 and hence, the solution becomes,
∞
X  nπ   cnπ 
u(x, t) = Bn sin x cos t
n=1
l l

with Z l
2  nπ 
Bn = f (x) sin x dx
l 0 l
Example 1 Solve

∂ 2u 2
2∂ u
=c for 0 < x < 1, t > 0
∂t2 ∂x2
such that,
u(0, t) = u(1, t) = 0, for t > 0
u(x, 0) = x(1 − x), for 0 < x < 1
ut (x, 0) = 0, for 0 < x < 1
Solution : Note that, l = 1, f (x) = x(1 − x) and g(x) = 0. Thus,
the solution is
∞
X
u(x, t) = Bn sin(nπx) cos(cnπt)
n=1

with, Z 1
Bn = 2 x(1 − x) sin(nπx) dx
0
8

n3 π 3 n odd
=
0 n even
K.Mahalingam Wave Equation

Thus, the solution is,

∞
X 8
u(x, t) = sin(nπx) cos(cnπt)
n3 π 3
n=1, n−odd

0.2 D’Alembert’s Method

We solve the wave equation

∂ 2u 2
2∂ u
= c , 0 < x < l, t > 0 (4)
∂t2 ∂x2
with the given initial conditions u(x, 0) = f (x) and ut (x, 0) = g(x).
We use the characteristics and convert the equation to its canonical
form. We then use the canonical form to find u(x, t).

Step 1 : Finding the characteristic:

Note that the wave equation can be written as:
2
2∂ u ∂ 2u
c − 2 =0
∂x2 ∂t
Here, A = c2 , B = 0, C = −1. Hence the equation of the characteristic
is:
d2 t dt 2
2d t
A 2 −B + Ct = 0 =⇒ c −t=0
dx dx dx2
which gives the auxiliary,
1
c2 m2 − 1 = 0 =⇒ m = ±
c
Hence,
dt 1
=±
dx c
K.Mahalingam Wave Equation

Integrating the above we get two characteristics,

v = x − ct and w = x + ct

. Step 2 : Converting to its canonical form:

We now convert the given wave equation in terms of v and w. We

have,
vx = 1, vt = −c, wx = 1, wt = c
ux = uv vx + uw wx = uv + uw
uxx = uvv + 2uvw + uww
ut = uv vt + uw wt = −cuv + cuw
utt = c2 uvv − c2 uvw − c2 uwv + c2 uww = c2 (uvv − 2uvw + uww
Thus, the wave equation becomes,

c2 uxx − utt = c2 (uvv + 2uvw + uww ) − c2 (uvv − 2uvw + uww ) = 0

=⇒ uvw = 0
which is the canonical form of the given wave equation.
Step 3 : Solving for u in terms of v and w
We use the canonical form of the wave equation obtained in Step 2 and
integrate it with respect to v and w.
Z
uvw dv = 0 =⇒ uw = φ1 (w)
Z Z
=⇒ uw dw = φ1 (w)dw + ψ(v)
R
Taking φ(w) = φ1 (w)dw, we have,

u(v, w) = φ(w) + ψ(v)

K.Mahalingam Wave Equation

=⇒ u(x, t) = φ(x + ct) + ψ(x − ct)

where φ and ψ are arbitrary functions.
Step 4 : Applying initial conditions

We first apply the condition u(x, 0) = f (x) which gives us

u(x, 0) = f (x) = φ(x) + ψ(x) (5)
Applying the other boundary condition ut (x, 0) = g(x), we get,
ut (x, t) = cφ0 (x + ct) − cψ 0 (x − ct)
where φ0 and ψ 0 represent derivatives with respect to their arguments
x + ct and x − ct respectively. Hence,
ut (x, 0) = g(x) = cφ0 (x) − cψ 0 (x)
g(x)
=⇒ φ0 (x) − ψ 0 (x) =
c
Integrating the previous expression with respect to x from x0 to x and
by Fundamental theorem of Calculus, we have
1 x
Z
φ(x) − ψ(x) = g(s) ds + k(x0 ) (6)
c x0
where k0 = φ(x0 ) − ψ(x0 ) is the constant of integration. Now adding
Equations (5) and (6) we get,
1 x
Z
f (x) k(x0 )
φ(x) = + g(s) ds + (7)
2 2c x0 2
Now, Subtracting Equation (6) from equation (5) we get,
1 x
Z
f (x) k(x0 )
ψ(x) = − g(s) ds − (8)
2 2c x0 2
K.Mahalingam Wave Equation

As, u(x, t) = φ(x + ct) + ψ(x − ct) we replace x with x + ct in Equation

(7) and x with x − ct in Equation (8), we get,

1 x+ct
Z
f (x + ct) k(x0 )
φ(x + ct) = + g(s) ds +
2 2c x0 2

f (x − ct) 1 x−ct
Z
k(x0 )
ψ(x − ct) = − g(s) ds −
2 2c x0 2
Z x0
f (x − ct) 1 k(x0 )
= + g(s) ds −
2 2c x−ct 2
Thus, adding the above two equations we get,

1 x+ct
Z
1h i
u(x, t) = f (x + ct) + f (x − ct) + g(s) ds
2 2c x−ct

Special Case : If ut (x, 0) = g(x) = 0, then

1h i
u(x, t) = f (x + ct) + f (x − ct)
2