- Home>
- Products & Solutions:
- Controllers, Recorders & Data Acquisition:
- Service & Support:
- Tips of the Month:
- June, 2009:
- February, 2008>
- August, 2007:
- June, 2007:
- January, 2007:
- November, 2006:
- September, 2006:
- August, 2006:
- May, 2006:
- March, 2006:
- February, 2006:
- January, 2006:
- December, 2005:
- November, 2005:
- October, 2005:
- September, 2005:
- August, 2005:
- July, 2005:
- June, 2005:
- ns-tips20:
- ns-tips21:
- ns-tips22:
- ns-tips23:
- December, 2005
Tips of the Month
December, 2005
Tuning Controller PID Parameters Manually
Though a variety of methods have been discovered for automatically tuning PID parameters, there are thought to be many cases in which adjustments can be made manually based on experience. The following is a procedure for manual tuning of the PID parameters that form the foundation of PID control. The method introduced here is the most well-known, the Ziegler-Nichols Critical Gain method.
Procedure
- In proportional control mode, narrow down the proportional band from a sufficiently large value to determine the proportional band (critical proportional band PBu) and period of oscillation (critical period Tu) at which the response is a continuous oscillation of a uniform amplitude. (Figure 1.)
- From that value, calculate the PID parameter using table 1, and check the control operation. (Figure 2)
- Watching the control results, finely tune the PID parameter to the optimum value. (Figure 3: Example with proportional band set somewhat large)
Table 1: Ziegler-Nichols Critical Gain Method
| Control | Proportional Band PB |
Integral Time TI | Differential Time TD |
|---|---|---|---|
| P | 2PBu | - | - |
| PI | 2.2PBu | 0.83Tu | - |
| PID | 1.7PBu | 0.5Tu | 0.125Tu |
![]() |
![]() |
![]() |
| Figure 1: Continuous oscillation | Figure 2: After setting PID with Critical Gain method | Figure 3: After fine tuning PID |
Notes
The examples given above are relatively common uses for manual PID parameter tuning, but this tuning method may not be optimal for all processes. In particular, when applying to PH control, reaction canister control, and other processes with a high degree of non-linearity, extra care must be taken since the control output can fluctuate greatly when changing the PID parameter, and the controlled objects (temperature, current, pressure, etc.) may change dramatically.






