Because sound emissions can reveal whether a laser process is running stably or out of sync, Fraunhofer IWS is developing acoustic monitoring solutions that evaluate laser processes in real time. The aim is to achieve cost-effective and robust process control that can systematically analyse sound emissions during processing in order to detect deviations immediately. The Dresden-based institute recently presented its first market-ready monitoring module at Laser World of Photonics 2025 in Munich. At the same time, research teams at the institute are working on developing approaches for further applications such as laser welding and cutting.

Industry-oriented solution

Quality control in laser micromachining is often time-consuming, technically complex and delayed.

Typical are structures with dimensions in the micro and sub-micrometer range, which can only be monitored and evaluated using complex microscopic methods. The new acoustic-based monitoring module from Fraunhofer IWS now offers an industry-oriented solution: It analyses the sound emissions generated during processing, evaluates the quality immediately and visualises anomalies without the workpiece having to leave the machine.

Acoustics as a data source: real-time evaluation during the process

The system uses robust semiconductor microphones to detect sound emissions, a local evaluation unit and a neural network that has been trained on real quality data. Acoustic signals are broken down into frequency ranges and characteristic parameters such as amplitude, duration or pulse shape are calculated from them. These are fed into the AI-supported evaluation, which reliably classifies the process status – for example, as stable, deviating or faulty. Feedback is provided in real time and independently of additional sensor technology.

Foreign objects, misalignment or uneven ablation become visible

According to Fraunhofer, one outstanding feature is the application for creating an ‘acoustic image’. The system generates a two-dimensional image of the structured surface based on the spatial distribution of the sound emissions. The acoustic image can be compared to classic surface images, but without additional measurement technology, purely from process data. Errors such as foreign objects, misalignment or uneven ablation become visually apparent and can be documented. ‘We wanted a solution that could provide information about process quality directly at the machine without anyone having to take a sample or send it to a separate measurement laboratory,’ says Tobias Steege, group manager for system engineering and sensor technology at Fraunhofer IWS. ‘Our acoustic analysis does not replace classic microscope measurement of a surface, but it enables cost-effective and robust process control for each individual component.’

Lean integration, economical operation

The monitoring module is designed as an industrial plug-and-play solution. It includes standardised Ethernet-based interfaces (TCP/IP, REST), requires no external cloud connection and can be linked to common machine controls. Thanks to its modular design and the use of semiconductor components, the overall costs can be significantly reduced compared to established acoustic monitoring systems. According to Fraunhofer, this enables precise, inline-capable and economical process control even for smaller series productions.

Ready for immediate use

‘The system is designed to be ready for immediate use: no special hardware or external data lines are required; it can be easily connected to the existing machine via a network connection,’ adds Steege. ‘Our strength lies in the combination of process-oriented sensor technology and intelligent software design that is geared towards everyday industrial use.’