Machine tools: These technologies make future-proof

A bit of cautious optimism could be heard despite the difficult economic situation for the German machine tool builders when Franz-Xaver Bernhard, chairman of the VDW (German Machine Tool Builders' Association), commented on the current situation of the

 German machine tool industry in January. The weak international demand, declining order intake, another expected production decline for 2025, as well as developments in the markets in China and the USA are just some of the challenges that German tool builders have to respond to.

Therefore, according to Bernhard, the technical leadership role of the industry is an important advantage. “Relatively stable, companies spend around three percent of their sales per year on research and development,” says Bernhard about the production location Germany.

Checking tools in automated processes in the machine room

Companies are relying on this innovation potential to respond to the market situation. It is clear: efficient timing and little waste are fundamentally essential aspects for sustainable manufacturing and high productivity. Tools are the core elements of the system. If they no longer work reliably, damage occurs.

The tool breakage sensor TD 110 from Heidenhain checks tools directly in the machine room, ensuring process reliability. The tool breakage sensor TD 110 from Heidenhain helps prevent damage in subsequent operations by detecting a broken tool on its way from the tool magazine to the machine room or back while passing by. This prevents the tool from being used again. The sensor detects a tool breakage for tools with diameters starting from 0.4 millimeters with a length change of 2 millimeters contactlessly via an inductive sensor - directly in the machine room with a rotating spindle at working speed.

This significantly increases process reliability. The signals are simply transmitted to the control via the probing system interface. This then triggers a message, an NC stop, or a user-specific stored reaction. For this purpose, the tool breakage sensor can be retrofitted in the working area of almost any machine tool in close proximity to the table. A pragmatic solution that ensures less downtime and secure processes.

Connect and optimize production

Process efficiency and thus market success fundamentally depend heavily on automation and digitalization. A partner that can support machine tool builders in implementing an intelligent factory is the Forterro Group. The company was founded in 2012 and operates as a provider of software solutions for the industrial mid-market.

Forterro has now expanded its portfolio to include the use of AI through the acquisition of Prodaso, a start-up specializing in the interface between IoT and AI applications for manufacturing processes. Customers can now optimize their productivity thanks to complete transparency and real-time analysis. The system continuously learns through artificial intelligence by analyzing historical and newly acquired data, with the goal of intelligently linking the data, recognizing patterns, drawing conclusions, and making predictions to initiate adjustment measures.

The benefits of this integration include improved resource efficiency and delivery reliability through intelligent control and planning with self-optimizing AI algorithms, or higher quality and error reduction through early detection of anomalies and optimization potentials through pattern recognition in production processes. The company also highlights another advantage: "With the AI-supported IIoT platform, all production facilities can be networked quickly and easily - from modern electronic machines to older equipment and even manually operated workstations."

Cybersecurity solutions for machine tools

With increasing digitization, the importance of functional safety (FuSa) and cybersecurity is also growing for machine builders. With the Machinery Regulation and the Cyber Resilience Act (CRA), the European Union has formulated laws aimed at regulating rapidly evolving future technologies. In particular, manufacturers of digital products or products that use digital technology must comply with strict security standards. Examples include embedded control of functionality, connectivity, artificial intelligence, and generally the provision of software. The regulations are to apply comprehensively from 2027. Companies must therefore be prepared accordingly by then.

Against this backdrop, Cadfem GmbH, a provider of simulation technology and digital engineering, and Clockworkx GmbH, a consulting company focused on cybersecurity and functional safety, have agreed on a strategic partnership. Together, they aim to support companies in integrating functional safety and cybersecurity early in the development and production process in view of the new requirements. The offering from Cadfem and ClockworkX includes consulting, training, and technical support, as well as assistance with the implementation and certification of security standards.

Josef Overberg, managing director of Cadfem Germany GmbH, emphasizes: "Especially our customers in mechanical engineering must prepare in time for the new legal requirements." Dr. Christian Geiss, managing director of Clockworkx GmbH, adds: "For medium-sized businesses, this means they must adapt their processes and products to meet the new requirements. We want to provide the necessary expertise and the right tools to overcome these challenges."

Process monitoring with AI for older systems

Quality, efficiency, and reliability are certainly the most important requirements that a machine tool must ensure. To ensure these, the use of AI is increasingly coming into focus, especially for error detection or predictive maintenance.

It is known that artificial intelligence in process monitoring can reduce waste, improve component quality, and relieve personnel. The recently completed research project Autopress by the Institute for Integrated Production Hannover gGmbH (IPH) and Jobotec GmbH now shows that expensive investments in new machines are not always necessary for this.

The researchers have developed a system of sensors and AI that allows old machines to be retrofitted. The system was developed using an older spindle press as an example. However, according to IPH, it can also be applied to other machines and systems. The spindle press was equipped with laser distance sensors, tension sensors, and temperature sensors. Various AI models evaluate the measurement results and compare them with the ideal parameters. If deviations are detected, the person operating the system receives feedback:

• "Warning, the tool is incorrectly installed!"
• "Warning, the semi-finished product is not centered!"
• "Warning, you have inserted the wrong material!"

The process monitoring detects parameter deviations with a success rate of 95 to 98 percent. "What was developed in the Autopress research project using a spindle press as an example can also be applied to numerous other machines and systems," say the scientists at IPH.

Generative AI for condition monitoring

The use of modern technologies in condition monitoring has changed the maintenance and servicing of machines and systems. For example, operators now receive early indications of potential problems through the analysis of machine parameters such as vibration levels, temperature fluctuations, and power consumption. This allows maintenance measures to be better planned and carried out in a timely manner.

A research project at Luleå University of Technology (LTU) in northern Sweden is now investigating how the possibilities of generative artificial intelligence can be integrated into the world of machine reliability. The basis is pre-trained models that serve as base models. This involves processing huge amounts of data.

One goal is, for example, the diagnosis or prognosis of bearing faults. The project is supported by the strategic innovation program of the Swedish process industry, PilA (Process industrial IT and Automation). It is being conducted in collaboration with SKF, a Swedish manufacturer of rolling bearings, seals, and components for lubricating parts, as well as mechatronics, based in Gothenburg.

The scientists know that it is difficult to automate condition monitoring. “In the laboratory, the relationship between vibration signals and certain bearing faults has been sufficiently investigated and established,” explains Karl Löwenmark, a doctoral student at Luleå University of Technology (LTU), who already dealt with the topic in his dissertation in 2020. “In a production environment, however, it is much more complicated and requires an application engineer with many years of experience,” he identifies a challenge. “Reading problems from the machine signals requires a lot of expertise. But the sheer volume of data alone is overwhelming.”

Is AI-assisted condition monitoring ready for practical use? Löwenmark is cautiously optimistic. "I would say we have a stable framework, but every condition monitoring system is different. This means users need to train the models with their own data and tailor them to their specific operations and processes. AI systems can use reinforcement learning to gradually improve the relevance and precision of their results based on user feedback."

Automatic coolant management for machine tools

But even with highly automated systems and processes, the basics must be ensured. The coolant, for example. It cools the tools and flushes chips from the workpiece and the workspace. Without it, the machine stops. The consequences are often expensive: production downtime and possibly a defective component.

A small machine can ensure reliability here: the Fluidworker 50 from MAW Werkzeugmaschinen GmbH. It automatically supplies the machine with fresh coolant and periodically measures the concentration and temperature. The device only requires a fresh water connection, is manufacturer-independent for most coolants, and can be easily retrofitted to existing machine tanks.

“Initial experiences with the Fluidworker show increased productivity, reduced machine downtime, and potential savings in coolant lubricants. A manufacturing company documented a 50 percent saving in coolant concentrate after just a few weeks of use. According to customer statements, this saving alone can amortize the investment,” says MAW.

System solution for coolant fluid management

Coolant lubricants are a key factor for quality and cost-effectiveness in production in the metalworking industry. They become contaminated during use by metal abrasion, dirt, and decomposition products due to thermal influences.

Gimat Liquid Monitoring, a provider of complete solutions for monitoring coolant lubricants, and the system house SD Nord, a provider of B2B solutions, are now jointly offering a comprehensive system solution for the fluid management of coolant lubricants. The goal is to monitor coolant lubricants (KSS).

Digitized measurement systems, for example for oil concentration and pH value, coupled with a comprehensive fluid management system for coolant lubricants, will support customers in meeting the requirements of the Technical Rule for Hazardous Substances (TRGS) 611. From recording measurement values at the machine tool to machine-related documentation of KSS quality and various evaluations, a comprehensive solution is offered. This is modularly structured and available in different performance packages. The Fluidas fluid management system is used for this purpose - an app- and web-based system that enables documentation, visualization, and evaluation as well as cost-efficient and safer planning.

“Customers always ask us where they should put their measurements and whether we offer options to document the data paperlessly,” says Dr. Volker Koschay, managing director of Gimat Liquid Monitoring. “Until now, we couldn't satisfy the customer needs for a seamless solution. That has changed with Fluidas,” Koschay continues. “It was the same for us, just the other way around,” adds Lars Löhner, managing director of SD Nord. “Our customers asked us for measurement technology that is compatible with our Fluidas software. Now we have what the customer needs,” Löhner continues. “No more tedious typing in and out of measurements.”

Edited by Julia Dusold