Macro Processing: Process Monitoring and Control (LiM 2015)

Use of inline coherent imaging for laser welding processes: Process control and beyond
Markus Kogel-Hollacher, Thibault Bautze

The introduction of inline coherent imaging technologies as a sensor for the laser materials processing is accompanied by the integration into several applications. One of these is the measurement of the depth of the vapor capillary for laser welding applications, now allowing to keep record of the welding depth with an accuracy of micrometers and a sub millisecond temporal resolution. The broader achievement is the closed-loop control of the welding depth that was not available in industrial environments till now due to the lack of an adequate sensor. Further use includes the acquisition of 3D images around the laser process itself, allowing for coaxial integration of pre- and post-process sensors. These applications are demonstrated by using the In-Process Depth Meter (IDM).

Keywords: laser welding; process monitoring; process control; low-coherence interferometry; sensor technology; penetration depth


Seam tracking for fillet welds with scanner optics
Friedhelm Dorsch, Holger Braun, Dieter Pfitzner

Fillet welding requires a high positioning accuracy of the laser beam in respect to the joint. Workpiece manufacturing tolerances and variations of the clamping require the individual positioning for every workpiece, i.e. an online seam tracking is needed. The proven concept of simultaneous observation of the joint and the laser spot by a coaxial camera sensor and real-time image processing are adapted to the PFO 3D scanner optics. Light-section measurement of the joint ensures robust detection and insensitivity to workpiece surface attributes and to variations of the workpiece cut shape. The system has shown excellent performance at different workpieces and different weld configurations (e.g. incident angle) and is presently under industrial qualification

Keywords: Laser fillet welds; seam tracking; online sensor system; real-time image processing; light-section measurement;


Using optical measuring techniques to investigate the hot cracking susceptibility of laser welded joints
Nasim Bakir, Andrey Gumenyuk, Michael Rethmeier

The safety of components or constructions is of great importance in the manufacturing and processing of metallic materials. Solidification cracking as well as the weldability of materials remain still for many years a highly contentious issues, particularly with regard to the causes of the hot crack formation. Many of studies have been conducted to determine the critical condition of occurrence of the solidification cracking. In this study different digital image correlation measuring techniques in conjunction with laser diodes as the illuminating source have been employed to measure the arising strain field during the laser welding process at the surface of the workpiece directed to the laser beam in the close vicinity of the weld pool. The Controlled Tensile Weldability test (CTW) was used to apply an external tensile load during the laser beam welding in order to generate the solidification cracks. The results showed that by means of those techniques it is possible to measure the strain field without any disturbances from the intense welding light or the smoke. Additionally, the strain and the strain rate as a critical factor determining solidification crack formation can be measured and analyzed.

Keywords: solidification cracking; laser beam welding; optical measuring technique; hot cracking test.


How fast is fast enough in the monitoring and controlling of laser welding?
Felix Tenner, Florian Klämpfl, Michael Schmidt

In the present study we show how fast the fluid dynamics change when changing the laser power for different feed rates during laser metal welding. By the use of two high-speed cameras and a data acquisition system we conclude how fast we have to image the process to measure the fluid dynamics with a very high certainty. Our experiments show that not all process features which can be measured during laser welding do represent the process behavior similarly well. Despite the good visibility of the vapor plume the monitoring of its movement is less suitable as an input signal for a closed-loop control, due to its high noise. Additionally, the plume does adjust with a delay on a change of process parameters. This physical limit restricts the maximal possible monitoring rate. Therefore, a reliable real-time control of laser welding over a wide range of process parameters might not be possible by the monitoring of the vapor plume. The features measured inside the keyhole show a good correlation with changes of process parameters. Due to its low noise, the area of the keyhole opening is well suited as an input signal for a closed-loop control of the process.

Keywords: Laser welding, monitoring, closed-loop control


Autocorrelation analysis of plasma plume oscillations in deep penetration laser welding
Libor Mrna, Martin Šarbort

The light emissions of plasma plume in deep penetration laser welding are typically characterized by irregular short-time pulses. Their timing is closely related to the dynamics of the keyhole formed within the workpiece and the surrounding weld pool. The nature of pulses limits the use of Fourier analysis because in most cases the frequency spectrum corresponds only to a colored noise. In this paper we present a study of the plasma plume oscillations using an autocorrelation function. We show that the autocorrelation function is an efficient tool to detect period of oscillations which are typically in the order of milliseconds. Finally, we compare the characteristics of autocorrelation function and the geometry of resulting welds carried out on a 2 kW Yb:YAG laser welding machine for the steel workpiece and various welding parameters settings.

Keywords: laser welding; plasma plume oscillations; autocorrelation analysis


On the detection of defects and of incorrect actuator settings in laser machining
Ralph Hohenstein

The detection of “unknown objects” is commonplace in several technical fields including sonar, radar and ultrasonic inspection. The techniques used in these fields have one common approach, that can figuratively be abbreviated as the “send & listen” approach: the emission of known wave fields and the observation of how these fields are refracted, attenuated or retarded by the unknown object. The usability of the send & listen approach for detecting defects and their causes and for silencing inherent random signal fluctuations has been researched during remote laser welding with a 4kW disk laser. We used the processing laser as a wave emitter (“send”) and observed wave reception (“listen”) within a multi-sensor arrangement. Sensors included band-filtered photodiodes and sound pressure detectors, both of which varied in position or direction. Our results show that the application of send & listen to laser machining may be beneficial for detecting changes in process results and for discriminating which machine setting is responsible for the individual change. In our spectral analyses we found unique spectral footprints for common defect occurrences such as lack of fusion and full/partial penetration and for common causes such as changes in axial focus, in lateral focus, in workpiece thickness, in the gap between joined parts, in laser power and in lateral beam velocity. This paper details our send & listen approach, shows some of the found spectral footprints and outlines their usability for process monitoring and control.

Keywords: laser, machining, welding, defects, detection, causes, process monitoring, control, send & listen, multi-sensor, multi-tone,
footprints, neural network, microphone, photodiode