Lead-containing alloys play an important role in the casting industry due to their good product properties. However, harmful emissions pose a problem in production. The TH Köln and Martin Luck Metallgießerei GmbH (MLS) are therefore developing an AI-based centrifugal casting process that reduces the lead content in the component and improves energy efficiency.
"Cast components such as pipes, discs or plain bearings are important components in mechanical and vehicle engineering or in the offshore industry. Copper-tin-lead alloys are often used to produce workpieces with high density, good electrical conductivity and advantageous friction and sliding behavior," says Jens Reuß from MLS, describing the initial situation.
In the production process, these advantages are offset by a disadvantage: Since the machines are often controlled manually, employees in the casting industry are exposed to dust and gas that are harmful to health. In addition, the process is very energy-intensive.
Saving energy with a prototype
During centrifugal casting, liquid metal at a temperature of around 1,000 degrees is poured into a cylinder called a mold, which rotates on its own axis. Due to the centrifugal forces, the melt collects on the mold wall and solidifies there, creating a rotationally symmetrical workpiece. A central goal of the project is to develop a new type of production process.
A prototype for a new casting machine is being developed by MLS and is intended to enable the rapid production of up to 15 cast parts per day with a height of around 25 centimeters and a diameter of 25 centimeters in accordance with DIN standards. Since heating the mold consumes a lot of energy, the focus is on an energy-saving heating system that heats the mold evenly using an air-gas mixture. In addition, the cooling system is to be adapted and automation components integrated.
Lower lead content with the same quality
With the help of alloy modeling, the Materials Laboratory at TH Köln is investigating how a new alloy could be composed whose relevant properties do not deteriorate, even though the lead content is only 6.5 to 8.5 percent instead of the previous 13 to 17 percent.
"The properties of the finished cast parts depend not only on the lead content, but also crucially on the casting process. We are therefore also examining how various process parameters affect the product properties," explains Prof. Dr. Danka Katrakova-Krüger from TH Köln.
These include the temperature of the melt, the dimensions of the workpiece, the temperature and amount of cooling water, and the duration of the cooling process.
The scientists at the Institute for Materials Application are approaching the perfect microstructure using a simulation. To do this, an existing model for the solidification of cast components is optimized and adapted to the framework conditions in the project. The researchers are also working with MLS customers to develop quality criteria for the cast parts.
"The computer then calculates how far the lead content can drop so that the products are on par with those made from a conventional alloy," says Prof. Dr. Stefan Benke from the TH Köln.
Process control through AI
The results from these sub-projects flow into an AI model from the Institute for Automation & Industrial IT. This uses the process, laboratory and simulation data to determine the optimal process parameters for sufficient workpiece quality.
"The fully trained artificial intelligence will be able to determine the optimal settings for the system depending on the shape of the cast part, such as the rotation speed or the preheating temperature of the mold. It also determines at what point the casting process ends and cooling begins," says Prof. Dr. Christian Wolf from the TH Köln.
Thanks to the newly implemented automation components, the prototype should be able to implement the parameters determined by the AI without human intervention.
"Since systems have so far been controlled manually, employees spend a lot of time waiting next to the machine. During this time, they are exposed to emissions that are harmful to health, which can be avoided in the future. In this way, we are achieving a central project goal," says Reuß.
The expected energy and material savings are to be achieved through the innovative heating system and the optimized production processes. The AI-based process settings are also intended to ensure that there are fewer defective castings that have to be melted down again at high energy expenditure.
The project
The project “Simulation and Cl-supported process development to reduce lead content and lead emissions during centrifugal casting of copper-tin-lead alloys while maintaining material properties” will be carried out from October 2023 to September 2026 by three institutes of the TH Köln and Martin Luck Metallgießerei GmbH (MLS). Participants include Prof. Dr. Danka Katrakova-Krüger from the Materials Laboratory, Prof. Dr. Christian Wolf from the Institute for Automation & lndustrial IT/:metabolon Institute and Prof. Dr. Stefan Benke from the Institute for Materials Application. The Federal Ministry for Economic Affairs and Climate Protection is funding the project through the Central Innovation Program for SMEs (ZIM).
