Frequency domain measurements of melt pool recoil force using modal analysis

Tristan Cullom,Douglas Bristow,Cody Lough,Jason Blough,Troy Hartwig,Kevin Johnson,Nicholas Altese,Andrew Barnard,Robert Landers,Edward Kinzel,Ben Brown

doi:10.1038/s41598-021-90423-z

Abstract

Recoil pressure is a critical factor affecting the melt pool dynamics during Laser Powder Bed Fusion (LPBF) processes. Recoil pressure depresses the melt pool. When the recoil pressure is low, thermal conduction and capillary forces may be inadequate to provide proper fusion between layers. However, excessive recoil pressure can produce a keyhole inside the melt pool, which is associated with gas porosity. Direct recoil pressure measurements are challenging because it is localized over an area proportionate to the laser spot size producing a force in the mN range. This paper reports a vibration-based approach to quantify the recoil force exerted on a part in a commercial LPBF machine. The measured recoil force is consistent with estimates from high speed synchrotron imaging of entrained particles, and the results show that the recoil force scales with applied laser power and is inversely related to the laser scan speed. These results facilitate further studies of melt pool dynamics and have the potential to aid process development for new materials.

Highlights

The total energy can be inferred from the energy in the first harmonic assuming that the force response scales with the laser energy as measured by a photodiode response
This gives the fractional energy in the first harmonic a dependence on the duty cycle of the laser’s pulse train
The fractional energy was calculated by taking the ratio of the magnitude at the fundamental frequency to the rms magnitude of the input signal i.e. MFund/MInput

Summary

Introduction

-40 1 2 3 4 5 6 7 8 9 10 11 12 Frequency [kHz] For most of the samples, the higher harmonics lie outside of the bandwidth of the accelerometer used in this study. The total energy can be inferred from the energy in the first harmonic assuming that the force response scales with the laser energy as measured by a photodiode response.

Results

Conclusion