A secret ingredient for faster, easier & more accurate control solution for EIMMs
- Automation
- Plastics
- 14.3.2024
- Reading Time: {{readingTime}} min
- Share Article
Contents
This year it will be 40 years since we at KEBA started controlling the complex injection molding process. This fact however does not make KEBA rest on their laurels, though. The same applies to the entire industry – the IMMs (injection molding machines) are continually becoming more sophisticated. Especially in sectors which require the highest process precision and parts accuracy, e.g. health care, the end-customers, though, had been signalizing room for improvement.
“The challenge was put on the table: The improvement of process quality requires much better closed loop control quality”, outlines Michael Petruzelka, Vice President Plastics at KEBA. Slow reaction time, even when using fast closed loop control technology on industrial PCs still sporadically leads to unwanted under- or overshoots – resulting additionally also in higher mold wear.
The recent KEBA developments and testing efforts revealed what our process experts have had on their minds quite for a while: there is a new, fully undiscovered potential for even better control of injection molding processes – predominantly applicable for full-electric machines, partly for hybrid ones. So, if you want to get more from your new but also existing machine fleet, check this out!
“Short ride around the predictive feed-forward loop”
To bring everyone to the same starting line, let’s start with a simple analogy illustrating and helping to understand the predictive feed-forward loop.
Imagine you are driving a sports car on a scenic highway. You want to go as fast as possible while also keeping control of the car and avoiding the over-use of the vehicle, which could have a negative effect on your bank account. What factors do you need to consider and how to keep up with the execution of this task while still enjoying the view through the window? You are right, you can turn on autopilot!
Autopilots use a variety of sensors to gather data about the vehicle and its surroundings, continuously processing and monitoring the data by the onboard computer using complex algorithms, provide real-time feedback and take corrective actions – all just in a few µs. It can, for example, monitor and keep the same speed – even when the outdoor conditions change. This is how the predictive feed-forward control logic works.
What do sport cars & injection molding machines have in common?
If you have assumed that both can use predictive feed-forward concept to efficiently control their processes, you are already on the right track.
Surely, we could find more similarities, like highest automation and easiest handling etc., but that is not the focus of this article.
Predictive feed-forward control is a control strategy that uses a mathematical model of the system being controlled to predict the future behavior of the system and generate a control signal that anticipates changes in the system.
In the context of an injection molding machine, the objective is to constantly monitor and optimize the process by controlling the injection pressure, speed, and position of the molds, while improving the quality of the final product, decreasing energy consumption and increasing productivity.
Pressure closed-loop on PLC (Programmable Logic Controller) – still a golden standard?
The FPGA (Field-Programmable Gate Array technologies) used in the CP05x KEBA controllers form the basis for the implementation of highly efficient control concepts directly in the controller hardware. This means less demand on the software, and the necessary modules are implemented in the hardware with top efficiency.
Especially when considering electric injection molding machines, it has been the state-of-art closed-loop control practice since a few years to connect the inject pressure sensor directly as well as the 3rd party drives with EtherCat (e.g. with 500 µs) to the PLC in order to achieve the fastest and most accurate control using predictive feed-forward technologies. The results are generally satisfying, it usually takes 3 cycle tics (each ca. 500 µs long) from reading the sensor data, through calculation, to sending the output to the drive, amounting in ca. 1,5 ms. However, the “times are changing”.
KeDrive D3 - The new benchmark in control technology for high-end requirements
The latest comprehensive tests on real machines conducted by KePlast experts revealed striking synergies between KEBA´s PLCs & drives and shifted the control technology at KEBA to a new dimension. The inject pressure sensor is in this case connected and controlled directly by the drive. The pressure close loop functionality runs on servo drive. This grants the cycle time of ca. 125 µs, the maximal reaction time of ca. 500 µs and the shortest ever possible dead-time.
Having a closer look at the test results comparing the above described two scenarios of fast inject / hold ramp-up, with a 3rd party drive (red) there was an overshot of over 30 bar and a rising time of 900 ms to reach the needed pressure. With pressure closed-loop on KeDrive D3 (green) we achieved overshoot of only 1 bar and a rising time 450 ms (slow due to the machine type), typical rising time of 30ms. The mentioned shorter cycle times combined with a reduced oscillation led to extended lifetime of the very pricy molds and significantly improves parts quality.
The implications for a daily business of plastics producers
“Today, lot of end-customers classify machines due to repeat accuracy,” explains Günther Weilguny, Product Manager Plastics at KEBA. “The highest requirements come especially from the manufacturers in health care or automotive industry, which produce complex parts. To succeed in these sectors, the manufacturers need exceptional parts quality” adds Weilguny.
The KePlast Process Expert Gerald Reindl dives even deeper to the root of the issue. “If the reaction time is not fast enough, the filling is not complete, the parts quality is inconsistent and the whole process is considered unstable.” Here are the most common implications of such an unstable process:
Part defects: The high-pressure pulse during overshoot can cause overpacking, sink marks, or warpage in the molded part as well as affect the strength or functionality.
Mold damage: The high pressure can damage the mold surface or cause wear and tear on the mold components.
Machine damage: The repeated overshot can cause damage to the injection molding machine components, such as nozzles, barrels, or screws.
Drive utilization: a more stable closed loop control needs less energy resulting in cooler drives and motors, enabling us to push the machine to higher pressures without increasing the size of the drives.
“Monitoring the pressure signal during the molding process with the pressure closed loop on the drive helps to detect and to correct any overshot instantly and avoid costly issues,” summarizes Reindl.
“If the reaction time is not fast enough, the filling is not complete, the parts quality is inconsistent and the whole process is considered unstable”
Advantages of seamless integration between PLC and drives
Additional perks for both OEMs & end-customers are hidden in the integration of the controller and servo drive. The two mostly benefit from quick commissioning - meaning less down-time for the machines in use due to the simple drives exchange process and the fact that no additional tools are required.
Moreover, an extra help for the machine producer is the easy diagnostics thanks to integrated tools at PLC. Moreover, tuning of controller parameters becomes uncomplicated.
Conclusion
At the end of the day, the new solution provides room for process, material and maintenance cost-saving as well as profit generation in the form of new product and market possibilities.
