For a unity feedback system with open-loop $G(s)$, the error is $E(s) = R(s)/(1 + G(s))$, and
$$e_{ss} = \lim_{s\to0}\frac{sR(s)}{1 + G(s)}$$
The type of the system is the number of open-loop poles at the origin ($1/s^N$ factors in $G$).
Static error constants
$$K_p = \lim_{s\to0}G(s),\qquad K_v = \lim_{s\to0}sG(s),\qquad K_a = \lim_{s\to0}s^2G(s)$$
| Type | Step $A/s$ | Ramp $A/s^2$ | Parabola $A/s^3$ |
|---|---|---|---|
| 0 | $A/(1 + K_p)$ | $\infty$ | $\infty$ |
| 1 | 0 | $A/K_v$ | $\infty$ |
| 2 | 0 | 0 | $A/K_a$ |
Raising the type removes error for that input, but each integrator adds 90° of phase lag and makes the loop harder to stabilise.