Damping Ratio from the Damping Coefficient

A dashpot's rating in newton-seconds per metre means nothing on its own, because whether that is a lot of damping depends entirely on what it is damping. The reference is the critical damping coefficient $c_c = 2\sqrt{km}$: the exact amount that lets a displaced system return to rest in the shortest possible time without overshooting even once. The damping ratio is the honest measure, $\zeta = c/c_c$, and it is dimensionless by construction.

Three regimes follow. Below $\zeta = 1$ the system is underdamped: it overshoots and rings its way back, and essentially every machine and structure lives here. At exactly 1 it is critically damped, the fastest non-oscillating return there is. Above 1 it is overdamped: no overshoot, but slower than critical, because now the dashpot is fighting the return as much as it damped the oscillation. Instrument designers aim slightly below critical — a galvanometer or a bathroom scale settles fastest with a single small overshoot rather than a lazy creep.

The numbers found in real structures are much smaller than people expect. Welded steel frames sit near $\zeta = 0.005$ to 0.02, bolted assemblies and reinforced concrete somewhat higher because friction at the joints dissipates energy, elastomeric mounts around 0.05, and reaching 0.2 takes a purpose-built viscous damper. If a measurement comes back above about 0.3 with nothing in the system that could produce it, suspect the measurement: energy leaving through a foundation, a pipe connection or an instrument cable reads as damping in the machine.

Two honest limits. Viscous damping — force proportional to velocity — is a mathematical convenience more than a description of reality. Real energy loss in structures comes mostly from hysteresis in the material and friction at joints, neither of which is proportional to velocity, and the viscous model is used because it makes the equations linear and solvable. The ratio it produces is an equivalent, fitted to give the same energy loss per cycle. And a practical note for this page: the catalog has no unit type for a damping coefficient yet, so $c$ is entered directly in N·s/m and nothing converts it.