Automated monitoring of cleaning quality in parts cleaning

Cleaning and drying processes are monitored by a new system with automated analyses and AI-based process control and optimized in terms of quality, energy and resource use.

Stable high process quality and savings in energy, resources and operating costs - requirements that water-based cleaning processes have to meet today. Ecoclean has developed the innovative lab-on-a-chip system to implement these requirements safely and efficiently. It keeps the quality of the cleaning and rinsing baths at an optimum level through automated analyses based on various measuring methods. The new Smart Drying technology enables AI-supported, intelligent control and optimization of drying processes with hot air and vacuum.

Increasing demands on component quality lead to more complex cleaning processes. Predefined cleanliness specifications must be met in a stable and reproducible manner. This is increasingly associated with higher documentation requirements for process parameters. At the same time, there is the requirement to clean in a more energy and resource-efficient manner in order to reduce the CO2 footprint and costs. On the other hand, there is a shortage of skilled personnel with the appropriate training and cleaning expertise. Ecoclean is responding to this situation with innovative developments such as the Lab-on-a-Chip (LOC) system and Smart Drying technology for automated monitoring and intelligent control of cleaning quality. These solutions ensure that cleaning and drying processes run optimally and also meet documentation requirements thanks to the constant recording of parameters.

In order to analyze the concentration of cleaning or passivation media as well as bath contamination, a wide variety of sensors are required, which are used in the baths depending on the respective medium or process. In addition, the measurements usually require numerous manual interventions. The time-consuming determination of the bath quality often leads to the cleaning medium being added regardless of consumption. Likewise, bath changes are still sometimes carried out on a time-dependent basis. Both can lead to sub-optimal bath and therefore cleaning quality and a waste of resources. This changes with the new LOC system, which can be integrated into both multi-bath immersion and chamber systems. The combination of different measuring methods in just one system opens up an enormous bandwidth for detergent and bath analysis. And it does so without equipping each bath with different sensors and independently of the cleaning or passivation medium and process used. Samples are taken automatically from the respective stations, which are connected to the evaluation unit.

The methods of titration, pH value, conductivity and, for the first time, surface tension measurement are available for determining the detergent concentration. In modular cleaning systems, the latter enables the surfactant content in the bath to be determined separately and thus enables the targeted dosing of this component. The contamination of the baths can be measured using fluorescence and conductivity. In future, these measurement methods will be supplemented by methods for determining total organic contamination (TOC) and particle measurement. Which measurements are carried out with the sample taken from the bath can be entered via the system control. Between sampling from the baths, automatic rinsing ensures that the results are shown on the display without distortion and that appropriate actions are initiated if necessary, such as the addition of medium or water or a bath change. With the standalone version, which is also available, the samples are taken manually from the baths to be checked and the rest of the procedure is also carried out automatically. With both alternatives, the measuring methods can be combined as required and additional functions or new measuring methods can be easily integrated.

Drying is not only one of the most energy-intensive processes, but is also the bottleneck in parts cleaning due to the time required. Nevertheless, it is not uncommon for problems to arise - especially when different components have to be cleaned and dried. To be on the safe side, drying processes are often designed for too long based on a worst-case part, which results in increased energy requirements with correspondingly high costs and reduced productivity. The aim is therefore to obtain optimally dried parts from the system with minimized energy and time requirements. Ecoclean has achieved this with its AI-supported, intelligent Smart Drying technology for combined hot air/vacuum drying processes. The basis for this is a humidity/temperature sensor specially developed for this application, which is now standard in various types of system from the manufacturer.

In order to control the drying process in a time and energy-optimized manner, the AI model of the Smart Drying technology was first trained using data from thousands of drying tests with a wide variety of components and target parameters. Based on this knowledge, the AI evaluates the quality of the drying process taking place in each case by analyzing the parameters of humidity, temperature and pressure in the process chamber. It then decides which action makes the most sense for optimum drying. In the case of thin-walled parts or plastic workpieces, for example, this could be the addition of hot air. Stable, reproducible processes are essential for the use of Smart Drying technology, as they ensure that the same conditions always prevail in the process chamber. Worn seals or sensors that are not regularly calibrated, for example, lead to changes here. The next development steps at Ecoclean include transferring the findings from machine learning to the use of AI for the control and optimization of cleaning and rinsing processes (OM-5/26).