Laser polishing (LP) is considered as one of the enabling technologies primed to replace time-consuming manual polishing operations. During laser polishing, a thin layer of material is melted as a result of laser irradiation. Since molten metal is characterized by the melt pool relocation capabilities, laser polishing results in a significant decrease of surface roughness. Experiments on flat stainless-steel metal plates with a 1200W fiber laser lead to a surface roughness as low as Sa = 320nm (90% roughness decrease) with a processing time of approximately 40s/cm².
The aim of this study is first to demonstrate the potential for the upscaling of process speed. Upscaling laws based on power density and energy density will be discussed. Experiments are carried out to assess the upscaling from 1200W to a 10kW fiber laser in order to improve processing time. We can expect to reach a processing time of 5s/cm², a diminution by 4 to 6 times comparing to the current process. Under appropriate process parameters, certain classes of metallic materials are suitable for LP and can reduce their average surface roughness by more than 90 %.
Micro laser shock peening (μLSP) with pulse energies well below 1 J proved to be a useful technique to obtain fatiguelife performances similar to those reported for traditional LSP processes on metallic bulk materials [1, 2]. However, it suffers from a lack of productivity as spot sizes are reduced and pulse overlaps are increased in order to obtain compressive residual stresses, deep below the surface of the bulk material. To overcome these limitations of μLSP, we have investigated strategies to scale up the productivity by increasing laser repetition rate while keeping constant the total amount of energy deposited on the peened surface [3-5]. We have built a laser processing cell to meet industrial grade applications. Complex surfaces are mounted on a KUKA robot to control the laser orientation and pulse overlap on the 3D workpiece surface. The pulse energy is provided by an 8 ns, Nd:YAG Laser, operating at 1064 nm, with a variable repetition rate from 10 to 100 Hz and delivering a maximum energy of 450 mJ/pulse on Al 2024-T351 samples with a thickness of 10 mm. We present high speed video analysis as diagnostic tool illustrating limitations in upscaling of repetition rates. As a proof of the μLSP effectiveness we present compressive residual stress profiles with up to -500 MPa peak and a return to zero down to 1.8 mm below the surface. This represents a 5-times improvement of the maximum stress depth, when compared to conventional peening processes widely used in the aeronautic industry.