OPT FOR SECURITY AND PERFORMANCE
Improve the security and performance of your plant with alarm management, performance supervision, asset supervision, web-based data visualisation, optimisation, advanced control, and tuning with the new TaiJi tool.
Rethink your working methods and obtain immediate results thanks to our services, trainings, and consulting expertise. Manage changes intelligently with our proven solutions.
certified service providerSeptember 2006
In this issue
 |
Classes |
|
Trick of the month: How long does it takes to stabilize a control loop? |
|
To contact us |
|
|
Coming classes and seminars
PlantTriage - Performance supervision and Process Control Optimization
September 26-28, Houston, TX
PlantTriage, What's new in V6?
September 29, Houston, TX
Detailed Loop Analysis with ExperTune
October 16-18, Hartland, WI and
October 16-18, Antwerp, Belgium
ExperTune Advanced Tools
October 18-20, Hartland, WI and
October 18-20, Antwerp, Belgium
Seminar on modern control methods
October 31-November 2, Vancouver, BC
See our calendar for 2006
|
|
This month's tip : How long does it takes to stabilize a control loop?
Questions |
Answers |
| What is your worst enemy in process control? | Dead time. |
| What can you do to reduce the settling time in a loop? | Reduce dead time. |
| Is it possible to quickly estimate the settling time? | ~ 10 * Dead time. |
When using PID control, the settling time after a set point change or a disturbance is approximately 10 times the dead time. If sluggish tuning parameters are used, the settling time will be longer.
Examples:
Belt conveyor
A weighing scale is installed on a belt conveyor; to adjust the amount of material on the belt, a sliding door is moved by a PID controller signal. The distance between the measurement and this door is 15 meters. The belt is moving at 0.3 m/s. Hence, the dead time is 45 s and the settling time (if the loop is properly tuned) is ~450 s (7.5 minutes). Reducing the distance to 5 meters will reduce the settling time to 2.5 minutes.Water temperature
A temperature sensor installed in a pipe measures the temperature of water. To control water temperature, hot water is injected. The distance between the sensor and the injection point is 21 m. The velocity is 3 m/s; the transportation time is 7 s. Total dead time of 18 s on this loop has been found, using bump tests. Reducing the distance would reduce the dead time - but on the other hand, we need good mixing. The sensor is moved at 6 m from the injection point. The transportation time is now 2 s. The dead time has been reduced from 18 s to 13 s; the settling time has been reduced form 180 s to 130 s.Positioning
A positioning system has an apparent dead time of 30 ms and a measured settling time of 290 ms. Mechanical parts are rigid; a zero backlash gear box is used. After analysis, we note that this dead time is coming from 3 time constants: 3 ms, 15 ms, and 12 ms; the real dead time is 0 but the apparent dead time is 30 ms. After modifying the electronic circuit, 15 ms has been removed; the settling time has been reduced to 160 ms.Conclusions
It is a lot cheaper to combat dead time at the design level.When discussing with production people, the theoretical limit to stabilize a loop is estimated using the rule of thumb: tsettling ~ 10 tdead time. Retuning loops will change this value, but only slightly.
|
each valve or transmitter will add 0.5 to 2 s to the dead time (use 1 s for your estimates if you cannot make a good educated guess). Do not hesitate to contact us.
To contact us
|
