# PID Loop Simulator: Derivative on Process Value (PV)

Description| How it works| Derivative on PV| Process| Robustness| PID TuningUsing using the derivative on error causes a spike in the output when the setpoint is changed. Due to this, the derivative is often calculated using the Process Value (PV) instead of the error. The screenshots below show the difference between using the derivative on the error and the derivative on the PV. In these screenshots, the same process is controlled by 2 controllers, one with derivative on error and the other with derivative on PV. The process values are:

Process Gain = 1, Time Constant = 20 secs, Delay = 5 seconds

The tuning constants are the same in both cases (Gain = 1, Integral = 20 secs, Derivative = 1). Note the large spike at t = 1 second caused by the derivative on error.

The spike caused by the step change in set point is not present in the controller where the derivative is calculated on the PV.

A simulation of the controller with the derivative on the PV can be downloaded here (.xls file, 142 KB)