Three metal 3D-printing machines produced overlapping contour melt-pool widths without producing identical surface roughness.
That result from the In718 comparison in the study separates a useful trend from a setting that could be transferred unchanged between machines.
What the contour scan changes
Laser powder bed fusion selectively melts layers of metal powder.
Bulk scans fill solid areas within the layer, while contour scans trace its edges with the goal of improving surface finish.
A melt pool is the liquid metal produced while a laser heats metal powder.
The reported measurements distinguish average roughness, Sa, from maximum valley depth, Sv.
A trend within a machine is not an exact prediction across machines
For stable In718 contours, increasing melt-pool width was associated with lower Sa and Sv.
Across the studied range on machine A, Sa fell from 10 to 5 micrometers and Sv from 60 to 30 micrometers.
Yet neither contour width nor linear energy density alone predicted exact roughness across the three In718 machines.
All three produced widths between 60 and 120 micrometers; machines A and B had similar Sa, machine C had lower Sa, and the three machines did not agree in Sv.
Knowing the direction of a trend therefore differs from knowing the value another machine will produce.
Machine C also showed an increase in average roughness above a contour width of 125 micrometers.
Changing the alloy changes the comparison
For AlSi10Mg, increasing contour width did not show a change in Sa or Sv.
Stable contour categories had lower values than balling or indistinct categories.
The useful distinction in that comparison was the melt-pool category, rather than a wider-is-smoother rule.
A smoother surface leaves an internal question
The authors explain that approaching keyhole melting may lower surface roughness while risking subsurface porosity that could harm mechanical performance.
A better surface measurement therefore does not establish a better subsurface condition.
Identical laser speed and power also do not guarantee a fixed melt-pool size, because local geometry changes heat-transfer conditions.
Matching input settings and matching the resulting melt pool are separate questions.
Across the alloys and platforms studied here, a specific melt-pool width could not predict a specific surface roughness value.
Sources:
- https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=957361
- https://www.nist.gov/news-events/news/2017/05/tucking-nists-3d-printer-testbed
Reader links:
Top comments (0)