Heat accumulation in pulsed laser processing (drilling)

Overview

• This code was intended to be used for the calculation of heat accumulation in pulsed laser processing, primarily ultrafast drilling.
• The key idea is an efficient and flexible implementation of superposition of analytical solutions to the heat equation.
• Determination and superposition of these heatkernels is quite efficient - especially in the case of equidistant timestepping - which lends itself to an implementation based on fast convolution (frequency domain convolution).
• As is typical for these calculations, the temperature is evaluated immediately before the succeeding pulse.
• The solution is radially symmetric.
• The simulation domain is the half-infinite body.
• Sources are mirrored about the surface.
• Convection is not considered.
• Fluid dynamics cannot be considered in this formulation.
• Point / Gaussian and Circular "donut" sources are implemented.

Features

• Pulse repetition rate can be constant or variable (chirped).
• Energy distribution over depth can be variable and time-dependent.
• A special case could be burst-processing.
• Irradiated / residual pulse energy can be time-dependent.
• Gaussian / "donut" source: Lateral extent of source can be time-dependent.
• In conclusion: The implementation is flexible enough to provide (in theory) a useful and fast approximation of heat accumulation in ultrafast laser processes - in particular percussion drilling.

Examples

• Drilling with an energy ramp

• Drilling with a variable pulse repetition rate

• Burst processing

• Not sure what to call this

Get Started

• Simply run main.m
• Beyond that you're unfortunately on your own, but in general I'm happy to answer question. Contact me.
• The code is not pretty, but I think there are some neat ideas / implementations used in heatacc and get2D_Distributed_Heatkernels.

Why wasn't this used?

• Unfortunately both the drilling progress as well as the energy distribution (i.e. absorbed energy over the borehole depth resulting in the temperature increase) is mostly unknown.
• Even worse: Existing models that estimate the drilling progress are based on considerations (absorbed energy distributions) which are at odds with experimental observations.
• Determination of the energy distribution in pulsed drilling is non-trivial and until that problem is solved these calculations are, realistically, nothing more than pretty to look at.