With the growth of laser welding technology, there is increasing need to improve the widespread empirical method for developing process parameters to reduce development cost.  To achieve this goal, it is necessary to know the process parameters that will produce the required weld dimension (in particular the weld depth) beforehand.  To achieve this goal an efficient modelling technique based on a scientific understanding of the welding process is required. The modelling approach used for this study goes against the current trend of developing increasingly sophisticated models.  Instead, a simple but fast laser drilling model is used to represent the process.  The study was conducted on a series of Nd:YAG pulsed laser spot welds on Ti 6Al 4V, Type 304 stainless steel and carbon steel. 

The experimental data covers a wide range of experimental conditions, with laser beam intensity in the range 0.4 to 50 kW/mm², and with accurate measurement of beam power and beam diameter.   Laser spot welds were made in all three materials, in as-received and surface linished condition, and with or without inert gas shield.  It is demonstrated that material surface condition and shielding have a major effect on bead dimensions, and initial experiments on laser beam absorption also show large differences in measured absorption depending on surface condition. 

The comparison between the experimental data and the model results showed that the model gave reasonable predictions of the different welding regimes and the melt depth given the model's simplicity.

Measured Absorption vs surface condition and intensity (Note: Absorption measurements at highest beam intensity, 51.2 kW/mm2 are not reliable due to through thickness keyhole and material ejection)

Penetration depth model predicts vs experimental measurements for different surface conditions

(Open points – incident beam intensity, solid points - corrected for absorption)