Introduction:
Extrusion plastometers are analytical instruments used to accurately determine the melt properties of thermoplastic materials. Temperature control accuracy and thermal response time are critical performance characteristics for these instruments. To achieve the required accuracy, a lengthy and complex calibration of the system is performed. These calibrations are required both in house following manufacture as well as in the field during equipment servicing and recalibration. Often the calibration process required is considerably longer than any other processes associated with manufacture or service of the plastomer. This causes problems with throughput; scheduling, and field downtime.
Customer Challenge:
Lack of repeatability and lengthy thermal stabilization characteristics were identified as contributing to calibration times reported as ranging from many hours to a day or more. It was indicated that the performance goal for the system was a calibration time of no more than a few hours. This deficiency was characterized as a controllability problem. The controllability problem was attributed to inadequate performance of the algorithm provided by the control unit.
Single Iteration Solution:
Qualitative evaluation of control performance in the extrusion plastometer confirmed the controllability problem. The cycle time of several hours per PID setting was found to be a practical obstacle to optimizing control component performance using classical tuning methods. In addition, lack of repeatability between runs was found to further compound tuning complexity. Poor repeatability is generally symptomatic of an improperly characterized or unstable thermal load; however, this condition was observed under static load conditions. In order to better understand the nature of the apparent unplanned variation in control system load characteristics, the design of the load was investigated. This identified significant issues inherent to the design of the thermal load itself including:
- Inconsistent interface between heat load and sensors
- Unacceptable contact resistance variation in the heater to power lead junctions
- Lack of optimization in heater band resistance wire layout
- Suspect clamping of heating band to thermal load
- Conductive thermal load mounts causing localized losses
Based on physical and electrical examination of the heater, Single Iteration created a model to simulate and analyze the dynamic behavior of the thermal load and determine if a control algorithm could be developed based on the existing control component hardware architecture and heater circuit design.
The results of simulation were used to cost effectively reduce component tolerance specifications and eliminate the sources of load instability. This allowed Single Iteration control engineers to use the optimized thermal load to build a custom control algorithm that offered an ideal balance of performance and cost. This newly designed system offered a much smoother transient response and the desired corresponding reduction in calibration time saving the manufacturer time, money, and industry reputation.
