Skip to content
LoveToCalc

Definition — what does that mean?

Second-order linear differential equation

The roots of r² + ar + b = 0 decide what the solution looks like. Complex roots α ± βi give an oscillation, e^(αx) times a cosine and a sine.

In Swedish: Linjär differentialekvation av andra ordningen

y″(x)+ay′(x)+by(x)=0

General solution

y=C₁er₁x+C₂er₂x,a²−4b>0eαx(C₁ cos(βx)+C₂ sin(βx)),a²−4b<0(C₁+C₂x)erx,otherwise

where

  • r₁=−a+a²−4b2
  • r₂=−a−a²−4b2
  • r=−a2
  • α=−a2
  • β=4b−a²2

C₁ and C₂ are arbitrary constants.

x
The variable
a
The constant coefficient of y′
b
The constant coefficient of y

The equation and the three cases

Try y = e^(rx). That gives the characteristic equation r² + ar + b = 0, and the sign of its discriminant a² − 4b picks the row of the general solution above: two real roots r₁ ≠ r₂, complex roots α ± βi, or one double root r.

Example

y″ + 2y′ + 5y = 0: r² + 2r + 5 = 0 gives r = −1 ± 2i, so α = −1 and β = 2:

y = e^(−x)(C₁ cos 2x + C₂ sin 2x)

y″ − y′ − 6y = 0: r² − r − 6 = 0 gives r = 3 and r = −2, so y = C₁e^(3x) + C₂e^(−2x).

y″ + 4y′ + 4y = 0: r² + 4r + 4 = 0 gives the double root r = −2, so y = (C₁ + C₂x)e^(−2x).

The initial conditions fix the constants

y″ + 2y′ + 5y = 0 with y(0) = 1 and y′(0) = 1. From the solution above, y(0) = C₁ = 1, and y′ = e^(−x)((−C₁ + 2C₂) cos 2x + (−C₂ − 2C₁) sin 2x) gives y′(0) = −1 + 2C₂ = 1, so C₂ = 1: y = e^(−x)(cos 2x + sin 2x). A damped oscillation: the factor e^(−x) shrinks while the sine-cosine part swings.

Where next

Common mistakes

Common questions

Why does the guess y = e^(rx) work? Because the derivative of e^(rx) is r · e^(rx): every term of the equation becomes a constant times the same e^(rx), and what is left is r² + ar + b = 0.

Where do the cosine and sine come from? From e^((α + βi)x) = e^(αx)(cos βx + i sin βx), Euler’s formula. The real and imaginary parts are each a real solution, and the two together span all of them.

Laplace transform table

Table

Laplace transform table
f(t)F(s)Valid whenName
Transform pairs
11ss > 0Constant
tnn!sn+1s > 0Power (n = 0, 1, 2, …)
eat1s−as > aExponential
sin(ωt)ωs²+ω²s > 0Sine
cos(ωt)ss²+ω²s > 0Cosine
eat sin(ωt)ω(s−a)²+ω²s > aDamped sine
eat cos(ωt)s−a(s−a)²+ω²s > aDamped cosine
teat1(s−a)²s > aLinear times exponential
tneatn!(s−a)n+1s > aPower times exponential
eat−e−at2as²−a²s > |a|Hyperbolic sine
eat+e−at2ss²−a²s > |a|Hyperbolic cosine
1,t≥a0,otherwisee−asss > 0Unit step u(t − a), a ≥ 0
δ(t−a)e−asall sUnit impulse (Dirac delta), a ≥ 0
Rules
αf(t)+βg(t)αF(s)+βG(s)where F and G both existLinearity
eatf(t)F(s−a)where F(s − a) existsShift in s
u(t−a)f(t−a)e−asF(s)where F existsShift in t, a ≥ 0
f′(t)sF(s)−f(0)where F existsFirst derivative
f″(t)s²F(s)−sf(0)−f′(0)where F existsSecond derivative
tf(t)−F′(s)where F existsMultiplication by t
∫0tf(τ)dτF(s)swhere F exists and s > 0Integration
∫0tf(τ)g(t−τ)dτF(s)G(s)where F and G both existConvolution

Last reviewed October 1, 2026 ·How pages are reviewed

CalculusAll formulas