Just how (in)efficient is my laser system? Identifying opportunities for theoretical and auxiliary energy optimization

In manufacturing, there is increasing recognition of the need to increase energy efficiency, both to reduce process cost and improve carbon footprint. In order to achieve this, it is necessary to understand how manufacturing systems use energy directly and indirectly. These types of analyses have been carried out at the process level for traditional machining processes, as well as at the factory level to understand macroenergy flows and bottlenecks. Other researchers have accomplished considerable energy optimization work for laser processing. However, the emphasis of this work has been on the optimization of the laser–material interaction. This focus has overlooked the whole system viewpoint and the significance of supporting equipment. Laser welding, using a 300 W fiber laser, was chosen as the subject for this study; first, due to its ubiquity in many high-value manufacturing industries and second due to its potential as a gateway into other manufacturing processes, such as directed energy deposition and additive manufacturing. In this paper, the initial work was to produce a framework for categorizing the process states and subsystems found in a standard or generic laser machine tool. An electrical energy meter was used to measure the energy consumption for individual subsystems when creating autogenous weld tracks in 316L stainless steel. Analysis of these data showed that the laser is only 18% of the total power consumption, the most significant being the water-cooling subsystem (37%). Reported here is a complete analysis of laser welding energy efficiency at a system level. This primary analysis of current equipment typical energy consumption can be used to identify future strategies for energy efficiency improvements beyond the direct laser interaction. By focusing on the most energy-inefficient parts of the system, the greatest potential for improvements to the carbon footprint of laser processing can be quantified.