Hanna Instruments HI 8000 Series - Manual (Page 13)

13
opening and closing of a valve through which a neutral solution is circulated.
The amount of neutral solution added is relative to the opening of the valve,
which is directly proportional to the magnitude of difference between the mixing
tank pH and the set point.
Control of pH implies a complex adjustment. The relation between the added
reagent and the process pH is logarithmic. There is the possibility to introduce
large errors in the process due to over dosing or under dosing reagent creating
an oscillation effect. PID control can be used to reduce the possibility of over-
shoot and large oscillations in the process by creating control output propor-
tional to the magnitude of deviation from the set point (P), time integral of error
(I), and rate of change of the measurement (D). Proportional, Integral, and
Derivative (PID) control can be used individually (typically Proportional control
only) or in combination such as PI, PD or PID. How these control actions are
used depends upon the requirements of the process.
Set point (reference value) is the desired value of the measurement. The error is
defined as the difference between the set point and measurement:
Error = Setpoint – Measurement
The descriptions and definitions of the individual control actions are as follows:
PROPORTIONAL ACTION (P):
The simplest continuous control mode is proportional control, so called because
the controller output is proportional to the magnitude of error. However, the
proportional control is subject to one major limitation, steady state offset (steady
deviation from the set point). Increasing the sensitivity of the controller (control-
ler gain) can reduce the steady state offset but only with slow processes. For this
reason the proportional control by itself it is used primarily for slow, consistent
processes that can tolerate high controller gain, which minimizes the steady state
offset. Consequently, high gain control action can throw the process into oscil-
lation if the process variable becomes unstable and begins to change rapidly.
INTEGRAL ACTION (I):
To eliminate offset droop and tighten the control of the process, the integral
action is introduced in conjunction with proportional control (PI). Integral con-
trol produces control action proportional to the time integral of the error. As
long as the error exists (steady deviation from the set point), the integral term will
continue to increase, adding more control action, driving the error toward zero.
DERIVATIVE ACTION (D):
With derivative action, the controller output is proportional to the rate of change
of the measurement and is primarily used to avoid overshoots. Derivative action