What Is the Laplace Transform?
The Laplace transform converts a function of time f(t) into a function of a new variable s, defined by the improper integral
F(s) = L{f(t)} = ∫0∞ f(t) e−st dt
Named for Pierre-Simon Laplace, the transform is the workhorse of differential equations, control theory, and circuit analysis because it turns calculus into algebra: differentiation in the time domain becomes multiplication by s in the s-domain. A linear ODE with initial conditions becomes an algebraic equation for F(s), which you solve and invert. This calculator applies the standard transform table term by term, states the rule used for each piece, and then verifies its own answer by computing the defining integral numerically.
How to Use the Laplace Transform Calculator
Type any function of t built from the classic table entries: constants, powers tn, exponentials eat, the oscillations sin(bt) and cos(bt), the hyperbolic pair sinh(bt) and cosh(bt), the square root √t, and products such as tneat, eatsin(bt), eatcos(bt), t·sin(bt), and t·cos(bt). Sums and constant multiples of these are handled automatically through linearity. The calculator splits your input into terms, matches each against the table, applies the first shifting theorem where an exponential factor appears, and assembles F(s). A numeric check evaluates ∫ f(t)e−st dt at a sample s and compares it against the symbolic result, so you can trust what you copy down.
Worked Example
Take f(t) = t² + e−3t + sin(2t), the default input. Linearity lets each term go through the table on its own:
L{t²} = 2!/s³ = 2/s³, L{e−3t} = 1/(s + 3), L{sin(2t)} = 2/(s² + 4)
so F(s) = 2/s³ + 1/(s + 3) + 2/(s² + 4), valid for s > 0. The graph shows f(t) rising in the time domain and F(s) decaying in the s-domain, a shape typical of transforms: large s weights only the earliest moments of f, so F(s) → 0 as s → ∞ for every ordinary function.
The Table Rules and the Shifting Theorem
Every entry the calculator uses comes from integrating the definition once and reusing the result. The power rule L{tn} = n!/sn+1 follows by repeated integration by parts. The exponential rule L{eat} = 1/(s − a) is a single substitution, convergent when s > a. Sine and cosine give b/(s² + b²) and s/(s² + b²), while their hyperbolic cousins swap the plus for a minus. The first shifting theorem ties them together: multiplying f(t) by eat simply replaces s with s − a in F(s). That is why L{eatcos(bt)} = (s − a)/((s − a)² + b²): the calculator shows this substitution explicitly whenever a term carries an exponential factor. Multiplication by t corresponds to −dF/ds, which produces the t·sin(bt) and t·cos(bt) entries used for resonance problems.
Why the s-Domain Matters
The transform's real power appears when derivatives enter: L{f′(t)} = sF(s) − f(0) and L{f″(t)} = s²F(s) − s f(0) − f′(0). Initial conditions are absorbed automatically instead of being fitted at the end, which is why engineers solve initial value problems this way rather than hunting for homogeneous and particular solutions separately. The ratio of output transform to input transform of a linear system is its transfer function, the central object of control engineering: poles of F(s), the s-values where the transform blows up, encode decay rates and oscillation frequencies of the time-domain signal. A pole at s = −3 means a e−3t component; a pole pair at ±2i means a sustained oscillation at frequency 2.
Common Mistakes to Avoid
- Forgetting the factorial in the power rule. L{t³} is 6/s⁴, not 1/s⁴. The n! comes from n rounds of integration by parts and is the most commonly dropped factor.
- Shifting the wrong way. e−3t replaces s by s + 3, not s − 3. The rule is s → s − a with a = −3, and the sign error flips a decaying solution into a growing one.
- Mixing up sine and cosine numerators. Sine carries the constant b on top, cosine carries s. Remember that cos(0) = 1 forces the s-like behavior 1/s at large s for cosine.
- Ignoring the region of convergence. The integral defining L{e2t} only converges for s > 2. The formula 1/(s − 2) is meaningless at s = 1, and the graph here only draws F where it is valid.
- Treating the transform as multiplicative. L{f·g} is not L{f}·L{g}; products in time correspond to convolution in s. The product entries in the table are special cases, not a general rule.
Real-World Applications
Circuit analysis is the classic home of the Laplace transform: capacitors and inductors, whose behavior is differential in time, become simple impedances 1/(sC) and sL in the s-domain, so an RLC network reduces to Ohm's-law algebra. Mechanical engineers use the identical mathematics for mass-spring-damper systems, and control engineers design stabilizing feedback by placing the poles of a transfer function in the left half of the s-plane. Signal processing, pharmacokinetics (drug concentration decay), heat conduction, and seismology all lean on the same idea: move to the s-domain, work with algebra, and return. Paired with the initial value problem calculator and the second-order ODE solver, this tool covers the standard workflow of a differential equations course.
Frequently Asked Questions
What does the Laplace transform actually do to a function?
It reweights f(t) by the decaying exponential e−st and integrates over all positive time, producing a new function of s. Each value F(s) is a weighted average of f's entire history, with larger s emphasizing early times. The result trades time-domain calculus for s-domain algebra.
Which functions can this calculator transform?
Any finite sum of table forms: constants, tn, eat, sin(bt), cos(bt), sinh(bt), cosh(bt), √t, and the products tneat, eatsin(bt), eatcos(bt), t·sin(bt), t·cos(bt), each with any constant coefficient. That covers essentially every forcing function in a standard ODE course.
Why does the transform use e^(−st) instead of some other weight?
Exponentials are eigenfunctions of differentiation: d/dt e−st = −s e−st. Integrating against them converts d/dt into multiplication by s, which is exactly the property that turns differential equations into algebraic ones. Fourier analysis makes the same choice with imaginary exponents.
What is the region of convergence and why should I care?
It is the set of s where the defining integral is finite, typically s greater than the fastest growth rate in f. Formulas like 1/(s − a) are only meaningful there. When transforms are inverted or combined, all manipulations must happen inside a common region of convergence.
How does the calculator verify its symbolic answer?
It picks a sample s safely inside the region of convergence, computes ∫0T f(t)e−st dt numerically with Simpson's rule for a large T, and compares that number with the symbolic F(s) evaluated at the same s. Agreement to several decimals confirms the table lookup.
What is the first shifting theorem in plain terms?
Multiplying a signal by eat slides its transform sideways: L{eatf(t)} = F(s − a). Growth or decay in time becomes pure translation in s. It is the fastest route to transforms of damped oscillations like e−2tsin(5t), which model every real vibrating system with friction.
Why do initial conditions appear in the derivative rule?
Integrating L{f′} by parts produces a boundary term −f(0) alongside sF(s). This is a feature: when you transform an ODE, the initial data enters the algebra immediately, so the solution you invert already satisfies the initial conditions with no constants left to fit.
Can the Laplace transform handle discontinuous forcing functions?
Yes, and that is a major reason engineers prefer it. Step and impulse inputs, switches flipping and hammers striking, transform cleanly via e−cs/s and e−cs respectively, cases where classical undetermined coefficients struggle. This tool focuses on the smooth table entries; step functions follow the same logic with the second shifting theorem.
What is the difference between Laplace and Fourier transforms?
The Fourier transform uses a purely oscillating weight e−iωt over all time and analyzes steady-state frequency content; the Laplace transform adds a decay envelope and starts at t = 0, making it ideal for transients and initial value problems. Setting s = iω recovers Fourier from Laplace when both exist.
Do the poles of F(s) really describe the behavior of f(t)?
Directly. A simple real pole at s = −a corresponds to an e−at term; a complex pair at −a ± bi corresponds to e−at oscillation at frequency b; a repeated pole adds a factor of t. Reading pole locations is how engineers predict stability without ever inverting the transform.