Why some vibrations grow out of control — and why the fix is almost never just "make it stiffer."
Every mechanical system — a beam, a shaft, a floor, a bridge deck, a bracket — has a frequency it "wants" to vibrate at if you disturb it and let it swing freely. Nudge most structures and they'll oscillate at that one frequency and quietly decay. But apply a repeating force atthat same frequency, and the response stops being quiet. Each push arrives in sync with the structure's own motion, adding energy on top of energy already stored, and the amplitude can grow to many times what the same force would produce at any other frequency. That phenomenon is resonance, and it is one of the most consequential ideas in mechanical and structural design.
A system's natural frequency, ωₙ = √(k/m), is set entirely by its mass (m) and stiffness (k) — it is the frequency the system oscillates at with no external forcing, once disturbed. Resonance occurs when an external periodic force drives the system at or near ωₙ, causing the amplitude to grow dramatically — theoretically without bound if there were zero damping. Damping (friction, material hysteresis, dashpots, dampers) is the piece that actually exists in every real system: it dissipates energy, caps the peak resonant amplitude at some finite value, and determines how quickly vibration dies out once the driving force stops. All three pieces — natural frequency, forcing frequency, and damping — have to be considered together; none of them alone tells the whole story.
A force applied off-resonance spends part of every cycle fighting the system's own motion — pushing against it as often as it pushes with it — so energy comes and goes without ever building up much. A force applied exactly in sync with the system's natural oscillation always pushes in the same direction the system is already moving, adding energy on every single cycle with nothing working against it. With zero damping, that energy has nowhere to go and the amplitude grows without bound. Every real system has some damping — friction, internal material hysteresis, a dashpot, aerodynamic drag — which converts a fraction of that added energy to heat every cycle. The damping ratio ζ = c / (2√(km)) determines how much: a lightly damped system (ζ near 0) still reaches a very high resonant peak before losses catch up; a heavily damped system reaches equilibrium at a far lower amplitude, and its free-vibration response after a disturbance falls into one of three regimes — underdamped (ζ < 1, oscillates while decaying), critically damped (ζ = 1, the fastest possible return to rest with no overshoot), or overdamped (ζ > 1, a slower, sluggish return with no oscillation).
False, or at best dangerously incomplete. Stiffness changes the natural frequency — ωₙ = √(k/m) — it does not eliminate resonance risk. Making a structure stiffer raises ωₙ, which can just as easily move the natural frequency into the range of an operating disturbance (a motor's running speed, wind-induced vortex shedding, footfall frequency from pedestrians) that it was safely far away from before. A structure that was "too flexible" and got stiffened to fix that can end up resonating with a forcing frequency it never used to be near — a well-documented, real cause of structural and machinery failures. Strength resists a static overload; it says nothing about whether the natural frequency now lines up with a forcing frequency the structure will actually see. The classic illustrative case is a pedestrian or wind-induced bridge or floor resonance: the structure is easily strong enough to carry the static weight involved, and fails (or becomes unusably lively) anyway, because its natural frequency ended up close to the forcing frequency of the crowd or the wind. Managing resonance risk requires checking ωₙ against every relevant forcing frequency and, where needed, changing mass, stiffness, or damping deliberately — not just adding strength.
Explains why resonance occurs when a periodic driving force matches a system's natural frequency (ωₙ = √(k/m)), why amplitude grows dramatically without damping to limit it, and why the three damping regimes — underdamped, critically damped, overdamped — determine how a disturbed system actually settles back to rest.
It's intuitive to assume that making something stiffer or stronger automatically makes it less prone to vibration problems. In reality, stiffness is one of two variables (along with mass) that set the natural frequency, ωₙ = √(k/m) — changing it shifts where resonance occurs, but does not remove the possibility of resonance. A structure can be stiffened specifically to reduce deflection or increase strength and inadvertently move its natural frequency into the exact range of an operating disturbance it previously avoided.
A single-degree-of-freedom system is governed by m·x″ + c·x′ + k·x = F(t). Its natural frequency, ωₙ = √(k/m), is the frequency of free vibration with no forcing or damping. When a periodic force F₀cos(ωt) is applied at or near ω = ωₙ, the amplitude of the steady-state response grows sharply, approaching infinity as damping approaches zero, because the force stays in phase with the system's own velocity and continuously adds energy. The damping ratio, ζ = c / (2√(km)), governs both the finite peak amplitude actually reached at resonance in a real system and the character of the free-vibration response after a disturbance: underdamped (ζ < 1, oscillates while decaying), critically damped (ζ = 1, fastest non-oscillatory return to rest), or overdamped (ζ > 1, a slower non-oscillatory return).
Resonance analysis is a required check in rotating machinery design (shaft critical speeds), structural engineering (floor vibration from footfall, bridge response to wind-induced vortex shedding or pedestrian loading), automotive and aerospace design (engine-mount and panel resonances), and any system with a periodic forcing source. The well-known Tacoma Narrows Bridge and various footbridge and floor vibration incidents are commonly cited real-world illustrations of resonance-driven structural response, and modern design codes require explicit natural-frequency and damping checks against expected operating and environmental forcing frequencies for exactly this reason.
For a simple single-degree-of-freedom system, the natural frequency is ωₙ = √(k/m), set entirely by stiffness (k) and mass (m). Increasing stiffness raises the natural frequency; increasing mass lowers it. Real structures have multiple natural frequencies (vibration modes), each tied to a different deformation shape, but the same k/m relationship governs each mode.
At resonance, the driving force stays in phase with the system's velocity throughout the cycle, so it does positive work on the system continuously rather than alternately adding and removing energy. With little or no damping to dissipate that added energy, it keeps accumulating cycle after cycle, driving the amplitude to a value far larger than the same force would produce at any other frequency — theoretically unbounded with zero damping.
They describe how a system disturbed from rest returns to equilibrium with no further forcing, based on the damping ratio ζ. Underdamped (ζ < 1) oscillates back and forth with decaying amplitude before settling. Critically damped (ζ = 1) returns to rest in the shortest possible time with no oscillation and no overshoot. Overdamped (ζ > 1) also returns without oscillating, but more slowly than the critically damped case.
Not necessarily, and it can make things worse. Stiffening a part raises its natural frequency (ωₙ = √(k/m)), which shifts where resonance occurs — it doesn't eliminate the possibility of resonance. If the new, higher natural frequency happens to land on or near an operating disturbance frequency (motor speed, rotor imbalance frequency, etc.), the 'fix' can introduce a resonance problem that didn't exist before.
Common approaches include shifting the natural frequency away from expected forcing frequencies (by changing mass or stiffness deliberately, with the shift verified by calculation or testing), adding damping (dashpots, tuned mass dampers, viscoelastic materials) to cap peak amplitude and speed decay, and avoiding sustained operation at or near a known resonant speed (e.g., rapidly passing through a rotor's critical speed rather than running continuously at it).
Try our Mechanical Studio
More calculators, simulators, and guides for this discipline.