Recipe
Compute a Laplace transform directly from the definition
Use when: you need \mathcal{L}\{f(t)\}(s) for an f(t) not yet in a lookup table, or you're deriving or verifying a table entry
Goal: a closed-form F(s), together with the values of s for which the improper integral actually converges
Taught in Definition and motivation
Quick recipe
-
1
Write down \int_0^\infty f(t)e^{-st}\,dt with the given f(t) plugged in, and combine any exponentials sharing base e into a single exponent.
Example
For f(t)=e^{3t}: \int_0^\infty e^{3t}e^{-st}\,dt = \int_0^\infty e^{-(s-3)t}\,dt — combine the exponents by adding them, 3t-st=-(s-3)t.
Why: this integral IS the transform, by definition — every later step is just mechanics for evaluating it.
-
2
Replace the upper bound \infty with a variable T and wrap the whole integral in \lim_{T\to\infty}, since \int_0^\infty is shorthand for that limit, not a bound you can plug a number into directly.
Example
\int_0^\infty e^{-(s-3)t}\,dt = \lim_{T\to\infty}\int_0^T e^{-(s-3)t}\,dt
Why: skipping straight to "evaluate at \infty" hides exactly the question this step makes explicit — whether that limit even exists, which is where the transform's domain of validity comes from.
-
3
Find the antiderivative in t (treating s as a constant), apply the fundamental theorem of calculus over [0,T], then let T\to\infty and see what survives.
Example
\int_0^T e^{-(s-3)t}\,dt = \left[-\dfrac{1}{s-3}e^{-(s-3)t}\right]_0^T = -\dfrac{1}{s-3}e^{-(s-3)T}+\dfrac{1}{s-3}. Taking T\to\infty: if s>3, the exponent -(s-3)T\to-\infty, so e^{-(s-3)T}\to0, leaving \dfrac{1}{s-3}.
- the coefficient of T in the exponent (after Step 1's combining) is negative — e^{(\cdots)T}\to0 as T\to\infty and the limit exists — this is the "s>" condition reported alongside the answer.
- that coefficient is zero or positive — the exponential term stays at 1 or grows without bound, the limit fails to exist, and the transform simply doesn't exist for that s.
Why: this is exactly where every transform's "s>\text{something}" restriction comes from — it isn't decoration, it's the actual convergence requirement uncovered by taking the limit honestly instead of skipping it.
Common mistakes
- Treating e^{-\infty} as an intuitive shortcut instead of actually checking the sign of the exponent's coefficient on T.
- A sign slip combining e^{at}e^{-st} into e^{-(s-a)t}, which flips the reported condition from s>a to s>-a.
- Reporting F(s) with no restriction on s at all, as if the formula held everywhere.
- Forgetting to evaluate the antiderivative at the lower limit t=0, not just at T.