Derivative on the process variable
Taking derivative action on the measurement rather than the error, so a setpoint change does not produce a derivative kick.
What it means
Derivative on the process variable means the derivative term acts on the measurement instead of on the error. The two agree while the setpoint sits still, and they diverge the moment it moves.
Derivative on error produces a derivative kick. A step in setpoint is an abrupt change in error, so the derivative term fires a spike of output the loop never asked for; derivative on the measurement reacts only to the process's own movement, which is the information derivative action is actually for.
Most platforms let you choose which signal the derivative sees, and the choice is worth checking before concluding that a loop is mis-tuned. A loop that jumps at every setpoint change is often not too aggressive — it is simply differentiating the wrong signal.
Related
Where this comes up
Last reviewed 2026-09-18.