Physics · dynamics

Mass-Spring-Damper

The textbook second-order system — watch it move, and read why it moves that way.
m·x″ + c·x′ + k·x = 0
—

System

tune it
1.0 kg
100 N/m
2.0 N·s/m
1.00 m
0.0 m/s

Live response

— t = 0.0 s
speed
Characteristics
Response detail
How it works

Reading a second-order response

Undamped (ζ = 0)

No damping at all: oscillates forever at its natural frequency. A useful ideal; real systems always have some.

Underdamped (0 < ζ < 1)

The common case: overshoots and oscillates, each swing smaller than the last. Bells, suspensions and tuned loops live here.

Critically damped (ζ = 1)

Returns to rest as fast as possible without overshooting: door closers, instrument needles.

Overdamped (ζ > 1)

So much damping the system crawls back, never oscillating: slower than critical. Past the sweet spot, more damping hurts.

Worked example. With m = 1 kg and k = 100 N/m the natural frequency is ωn = √(k/m) = 10 rad/s (about 1.6 Hz) and critical damping is ccrit = 2√(km) = 20 N·s/m. At c = 2 the damping ratio is ζ = 0.1: lightly damped, it overshoots and rings for several cycles before settling (roughly 4 seconds to stay within 2%). Raise c to 20 and ζ = 1: it returns as fast as possible without overshooting. Bars show the damping coefficient c.
Field notes

The four damping regimes

ζ = 0.707

A common design target

The fastest response that keeps overshoot small (about 4%). Below it you get faster but bouncier; above, smoother but more sluggish. At ζ = 1 the two roots coincide and the response decays without crossing zero: the boundary between the oscillating and the sluggish worlds.

Does more damping always settle faster?

No: that is the surprise. Past critical, extra damping makes the system slower to return because it is fighting its own motion. Fastest settling is right around critical damping.

The same maths everywhere

A car suspension, a building swaying in wind, an RLC circuit, the closed loop of a tuned controller. Two numbers decide the character: the natural frequency ωn (how fast it wants to oscillate) and the damping ratio ζ (how quickly the oscillations die away).

What changes with two masses?

Two masses in series make a genuine 4-state coupled system with two natural frequencies (modes) instead of one. Pluck mass 1 and energy sloshes into mass 2: a simple model for a car’s body on suspension on tyre, or two floors of a building joined by a soft joint.

Where the model breaks down

It assumes ideal viscous damping, a massless linear spring and small motion. Real systems add friction, nonlinear stiffness and more: treat it as the clean ideal, not the whole story.

Questions

What people ask

What's special about ζ = 0.707?

It's a common design target — the fastest response that keeps overshoot small (about 4%). Below it you get faster but bouncier; above it, smoother but more sluggish.

Why does critical damping not oscillate?

At ζ = 1 the two roots of the system coincide and the response decays without crossing zero. It's the boundary between the oscillating (underdamped) and sluggish (overdamped) worlds.

Does more damping always settle faster?

No — that's the surprise. Past critical, extra damping makes the system slower to return, because it's now fighting its own motion. Fastest settling is right around critical damping.

Where does this model break down?

It assumes ideal viscous damping, a massless linear spring, and small motion. Real systems add friction, nonlinear stiffness, and other effects — treat this as the clean ideal, not the whole story.

What changes with 2 masses?

Two masses in series creates a genuine 4-state coupled system with two natural frequencies (modes) instead of one. Pluck mass 1 and energy sloshes back and forth into mass 2 — a simple model for anything from a car's body-on-suspension-on-tyre to two floors of a building connected by a soft joint.

Results are for reference only. An idealised second-order model — real systems add nonlinearity and effects this doesn't capture. Check against other sources.
Version history · 1 release
  1. v0.902026-07-28Springs drawn as coils rather than zigzags; showcase pendulums centred in their panel

Releases in which this page changed, newest last. Derived from the archived copy of every release, not from notes written afterwards — so it reflects what actually shipped. Site-wide passes are left out; they are in the full changelog.

Rules of thumb are approximations, offered to get you to the right order of magnitude fast. Check anything load-bearing against its source or the exact tool. Privacy · Terms

© 2026 Rules of Thumb · Runs entirely in your browser