Damped free vibration: viscous damping and logarithmic decrement
Viscous damping, critical damping and damping ratio, under-, critically and overdamped response, damped natural frequency and measuring damping by logarithmic decrement.
Drafted with Aria, reviewed by the AiCanCode.org team. Spotted an error? Use Give Feedback at the bottom of the page.
Why it matters
Damping decides how quickly a disturbed structure settles, how large a resonance can grow and, in aeroelasticity, whether a wing mode stays stable as speed rises: flutter is the speed at which a mode's total damping falls to zero. Damping cannot be calculated reliably from drawings, so it is measured, most often from the decay of free vibration through the logarithmic decrement. This topic gives you the model and the measurement.
Key ideas
Viscous damping model. A dashpot produces a force opposing the velocity, F_d = −c·ẋ, with c the damping coefficient (N·s/m). The equation of motion of a damped SDOF system in free vibration is
m·ẍ + c·ẋ + k·x = 0
Viscous damping is linear, which makes the equation easy to solve and lets modal analysis carry over. Real structural damping comes from material hysteresis, friction in joints and fasteners, and air; for light damping it is represented by an equivalent viscous damping ratio.
Characteristic equation. Trying x = e^(s·t) gives m·s² + c·s + k = 0, so s = −ζ·ω_n ± ω_n·√(ζ² − 1). The nature of the motion depends on the damping ratio ζ = c/c_c, where c_c = 2·√(k·m) = 2·m·ω_n is the critical damping coefficient.
- ζ < 1, underdamped: the roots are complex, the motion oscillates at the damped natural frequency ω_d = ω_n·√(1 − ζ²) inside an exponentially decaying envelope e^(−ζ·ω_n·t). Almost all aircraft structures fall here, with ζ typically 0.01–0.05.
- ζ = 1, critically damped: two equal real roots; the system returns to equilibrium fastest without overshooting more than once.
- ζ > 1, overdamped: two real negative roots; slow, non-oscillatory return.
Underdamped response. x(t) = e^(−ζ·ω_n·t)·[x₀·cos(ω_d·t) + ((v₀ + ζ·ω_n·x₀)/ω_d)·sin(ω_d·t)]. Note that ω_d < ω_n, but for ζ = 0.05 the difference is only 0.13 %.
Logarithmic decrement. The ratio of any two successive peaks (one damped period T_d = 2π/ω_d apart) is constant: x_i/x_(i+1) = e^(ζ·ω_n·T_d). Its natural logarithm is the logarithmic decrement δ = 2π·ζ/√(1 − ζ²). For small damping δ ≈ 2π·ζ. Using peaks n cycles apart averages out measurement noise: δ = (1/n)·ln(x₀/x_n). Count cycles carefully: x₀ to x_n spans n periods.
Energy view. In one cycle a lightly damped system loses a fraction ≈ 2δ of its energy, since energy goes as amplitude squared. A useful rule: the number of cycles for the amplitude to halve is ln 2/δ ≈ 0.11/ζ.
Other damping types (for awareness). Coulomb (dry friction) damping gives a linear, not exponential, decay of amplitude (each half cycle the amplitude drops by 2·F_f/k). Structural (hysteretic) damping is written as a complex stiffness k·(1 + i·g); at resonance g ≈ 2ζ. If a measured decay is a straight line rather than an exponential, the damping is not viscous and the log decrement is not constant.
Link to aeroelasticity. In flight, aerodynamic forces add their own damping (positive or negative) to each mode. The V–g method and flight flutter testing track the damping of each mode as speed increases; the flutter speed is where it crosses zero.
Formulas
m·ẍ + c·ẋ + k·x = 0 — damped free vibration; c damping coefficient (N·s/m).
c_c = 2·√(k·m) = 2·m·ω_n — critical damping coefficient (N·s/m).
ζ = c/c_c = c/(2·m·ω_n) — damping ratio (dimensionless).
ω_d = ω_n·√(1 − ζ²) — damped natural frequency (rad/s), underdamped only (ζ < 1). T_d = 2π/ω_d (s).
x(t) = X·e^(−ζ·ω_n·t)·sin(ω_d·t + φ) — underdamped free response; X and φ from initial conditions.
δ = ln(x_i/x_(i+1)) = (1/n)·ln(x₀/x_n) — logarithmic decrement (dimensionless), from peaks one or n cycles apart, same side of equilibrium.
δ = 2π·ζ/√(1 − ζ²) and inversely ζ = δ/√(4π² + δ²); for ζ < about 0.1, δ ≈ 2π·ζ.
n_half = ln 2/δ — cycles for the amplitude to halve.
Worked examples
Example 1 (standard). m = 10 kg, k = 4000 N/m, c = 40 N·s/m. Find ω_n, ζ, ω_d, δ and the amplitude ratio after 5 cycles.
ω_n = √(k/m)= √(4000/10) = 20 rad/s.c_c = 2·√(k·m)= 2·√(40 000) = 400 N·s/m, soζ = c/c_c= 40/400 = 0.10.ω_d = ω_n·√(1 − ζ²)= 20·√0.99 = 19.90 rad/s.δ = 2π·ζ/√(1 − ζ²)= 0.6283/0.99499 = 0.6315.- After 5 cycles: x₅/x₀ = e^(−5δ) = e^(−3.157) = 0.0425.
Answer: ω_n = 20 rad/s, ζ = 0.10, ω_d = 19.90 rad/s, δ = 0.631; the amplitude falls to 4.25 % in 5 cycles.
Example 2 (GATE level). In a ground test a 5 kg lumped-mass model is plucked. Its peak displacement falls from 10 mm to 2.5 mm in 6 cycles and the measured period is 0.20 s. Find ζ, ω_n, the damping coefficient and the stiffness.
δ = (1/n)·ln(x₀/x_n)= (1/6)·ln(10/2.5) = (1/6)·ln 4 = 1.3863/6 = 0.2310.ζ = δ/√(4π² + δ²)= 0.2310/√(39.478 + 0.0534) = 0.2310/6.2874 = 0.0367.ω_d = 2π/T_d= 2π/0.20 = 31.42 rad/s;ω_n = ω_d/√(1 − ζ²)= 31.42/0.99932 = 31.44 rad/s.c = 2·ζ·m·ω_n= 2 × 0.0367 × 5 × 31.44 = 11.55 N·s/m.k = m·ω_n²= 5 × 31.44² = 4941 N/m.
Answer: ζ ≈ 0.037, ω_n ≈ 31.4 rad/s, c ≈ 11.6 N·s/m, k ≈ 4.94 kN/m. The amplitude halves every ln 2/δ = 3.0 cycles, consistent with 10 → 5 → 2.5 mm in 6 cycles.
Common mistakes
- Using ln(x₀/x_n) without dividing by n when the peaks are several cycles apart, or counting peaks instead of intervals.
- Taking one peak above and the next below equilibrium; that is half a cycle and gives δ/2.
- Writing ζ = c/(2·√(k/m)); the critical coefficient is 2·√(k·m) = 2·m·ω_n.
- Applying ω_d = ω_n·√(1 − ζ²) to an overdamped system, which does not oscillate.
- Assuming damping noticeably lowers the oscillation frequency of a lightly damped aircraft structure; it mainly controls decay and resonant amplitude.
- Treating friction (linear) decay as viscous; the log decrement then varies from cycle to cycle.
For GATE AE
Typical questions: find ζ or c from the decay of two peaks; find ω_d from ω_n and ζ; identify under-, critically or overdamped behaviour from m, c and k; compute the number of cycles for the amplitude to fall to a given fraction; compare the exact and small-damping forms of δ. Practise both directions (δ → ζ and ζ → δ) and keep track of whether a given frequency is ω_n or ω_d.
Quick check
- m = 2 kg, k = 800 N/m. What is the critical damping coefficient?
- ζ = 0.1. What is δ (exact)?
- Two successive peaks are 8 mm and 6 mm. Estimate ζ.
- Is a system with m = 1 kg, k = 100 N/m and c = 25 N·s/m underdamped?
- For ζ = 0.05 and ω_n = 10 rad/s, what is ω_d?
Answers: 1. 2·√(800 × 2) = 80 N·s/m; 2. 0.631; 3. δ = ln(8/6) = 0.288, ζ ≈ 0.0458; 4. No, c_c = 20 N·s/m so ζ = 1.25, overdamped; 5. 9.987 rad/s.
Interview questions
All Structural Dynamics and Aeroelasticity interview questionsTry answering each one aloud before you open it.
1.What is viscous damping in the context of structural dynamics?Concept
Viscous damping refers to a type of damping in which the resistance force is proportional to the velocity of the moving object. It is commonly modeled as a linear damping force, where the damping force F_d is given by F_d = -c·v, with c being the damping coefficient and v the velocity. This type of damping is often used to represent energy dissipation in mechanical systems due to fluid resistance or internal friction.
2.Explain the concept of logarithmic decrement in damped free vibration.Concept
Logarithmic decrement is a measure of the rate of decay of oscillations in a damped system. It is defined as the natural logarithm of the ratio of the amplitudes of two successive peaks in the same direction. Mathematically, it is expressed as δ = ln(A_n / A_(n+1)), where A_n and A_(n+1) are the amplitudes of successive peaks. This concept is useful for determining the damping ratio of a system when the damping is small.
3.How does viscous damping affect the natural frequency of a system?Concept
Viscous damping reduces the amplitude of oscillations over time but does not significantly affect the natural frequency of the system. The damped natural frequency is slightly lower than the undamped natural frequency, but for lightly damped systems, this difference is negligible. The damped natural frequency ω_d is given by ω_d = ω_n√(1 - ζ²), where ω_n is the undamped natural frequency and ζ is the damping ratio.
4.Why is viscous damping commonly used in modeling mechanical systems?Application
Viscous damping is commonly used because it provides a simple and mathematically convenient way to model energy dissipation in mechanical systems. It is linear, which makes the equations of motion easier to solve analytically. Additionally, many real-world damping mechanisms, such as air resistance and internal material friction, can be approximated as viscous damping, making it a practical choice for engineers.
5.What happens to the amplitude of oscillations in a system with high viscous damping?Application
In a system with high viscous damping, the amplitude of oscillations decreases rapidly over time. The system may reach a critically damped or overdamped state, where it returns to equilibrium without oscillating. High damping leads to a faster decay of motion, which can be desirable in applications where quick stabilization is needed, such as in automotive shock absorbers.
6.How can you experimentally determine the damping ratio using logarithmic decrement?Application
To determine the damping ratio experimentally using logarithmic decrement, measure the amplitudes of successive peaks in a damped oscillation. Calculate the logarithmic decrement δ using δ = ln(A_n / A_(n+1)). The damping ratio ζ can then be found using the relation ζ = δ / (2π√(1 + (δ / (2π))²)). This method is particularly useful for lightly damped systems.
7.If a system has a damping ratio of 0.1, what is the logarithmic decrement?Numerical
The exact relation is δ = 2πζ/√(1 − ζ²). With ζ = 0.1, δ = 0.6283/√0.99 = 0.6283/0.995 ≈ 0.631. The small-damping approximation δ ≈ 2πζ gives 0.628, only about 0.5 % lower, so for ζ up to about 0.1 the approximation is acceptable. It means each peak is e^(−0.631) ≈ 0.53 times the previous one.
8.Calculate the damped natural frequency of a system with an undamped natural frequency of 10 rad/s and a damping ratio of 0.05.Numerical
The damped natural frequency ω_d is given by ω_d = ω_n√(1 - ζ²). Substituting the given values, ω_n = 10 rad/s and ζ = 0.05, we get ω_d = 10√(1 - 0.05²) = 10√(0.9975) ≈ 9.987 rad/s. This shows that the damped natural frequency is slightly less than the undamped natural frequency.
9.What is the significance of the damping ratio in the context of aeroelasticity?Application
In flight each structural mode has a total damping that is the sum of structural damping and aerodynamic damping, and the aerodynamic part changes with airspeed. Flutter occurs at the speed where the total damping of a mode falls to zero; beyond it the damping is negative and oscillations grow. Flutter analysis (for example the V–g method) and flight flutter tests therefore track modal damping ratio against speed, and certification requires adequate positive damping with margin up to beyond the design dive speed.
10.Explain how the concept of logarithmic decrement can be used to improve the design of aerospace structures.Application
Damping cannot be predicted reliably from drawings, so it is measured: in a ground vibration test or a flight flutter test a mode is excited, the excitation is stopped, and the decay of successive peaks gives δ = (1/n)·ln(x₀/x_n) and hence ζ = δ/√(4π² + δ²). The measured ζ replaces assumed damping in the dynamic and flutter models, which sets realistic resonant amplitudes and fatigue loads. If a mode is too lightly damped, designers add damping treatments (viscoelastic layers, dampers) or change stiffness and mass distribution, and in flight testing a falling δ with speed warns of approaching flutter.
Finished this topic? Mark it so your progress, study plan and readiness keep up.
Stuck on something here?