Matthias Hirsch

Loose powder detection and surface characterization in selective laser sintering via optical coherence tomography.

Defects produced during selective laser sintering (SLS) are difficult to non-destructively detect after build completion without the use of X-ray-based methods. Overcoming this issue by assessing integrity on a layer-by-layer basis has become an area of significant interest for users of SLS apparatus. Optical coherence tomography (OCT) is used in this study to detect surface texture and sub-surface powder, which is un-melted/insufficiently sintered, is known to be a common cause of poor part integrity and would prevent the use of SLS where applications dictate assurance of defect-free parts. To demonstrate the capability of the instrument and associated data-processing algorithms, samples were built with graduated porosities which were embedded in fully dense regions in order to simulate defective regions. Simulated in situ measurements were then correlated with the process parameters used to generate variable density regions. Using this method, it is possible to detect loose powder and differentiate between densities of ±5% at a sub-surface depth of approximately 300 μm. In order to demonstrate the value of OCT as a surface-profiling technique, surface texture datasets are compared with focus variation microscopy. Comparable results are achieved after a spatial bandwidth- matching procedure.

Focussed arc tungsten inert gas brazing of zinc-coated steels:

The brazeability of automotive zinc-coated steels depends on several factors. These include the morphology of the joint and the welding parameters selected. However, more fundamental material factors such as the composition of the coating, method of coating and coating thickness also have a significant effect. In this study, five commercially available and widely used automotive zinc-coated steels are investigated to assess brazeability. Surface zinc content and the coating type are shown to have a marked effect on the quality of the resulting joint. This is shown by surface analysis of the joint to determine evenness and bridging capability of the filler material and a cross-sectional analysis of the joints. Differences in wettability and contact length of the filler material and zinc-coated steel substrate are observed. It was found that electro-galvanised steel exhibited the best brazeability of the materials investigated here. Wettability of spreading angles as low as 17.3°, most uniform contact length and best bridging capability due to the filler material forming a metallic bond with the substrate were observed. However, pores were present in cross-sections. Galvannealed steel also showed good wetting with no embedded defects. Other steels used (galvanised and magnesium–aluminium zinc steels) presented problems with uniformity, high spreading angles of the filler material and poor bridging characteristics.

Meso-scale defect evaluation of selective laser melting using spatially resolved acoustic spectroscopy

Developments in additive manufacturing technology are serving to expand the potential applications. Critical developments are required in the supporting areas of measurement and in process inspection to achieve this. CM247LC is a nickel superalloy that is of interest for use in aerospace and civil power plants. However, it is difficult to process via selective laser melting (SLM) as it suffers from cracking during rapid cooling and solidification. This limits the viability of CM247LC parts created using SLM. To quantify part integrity, spatially resolved acoustic spectroscopy (SRAS) has been identified as a viable non-destructive evaluation technique. In this study, a combination of optical microscopy and SRAS was used to identify and classify the surface defects present in SLM-produced parts. By analysing the datasets and scan trajectories, it is possible to correlate morphological information with process parameters. Image processing was used to quantify porosity and cracking for bulk density measurement. Analysis of surface acoustic wave data showed that an error in manufacture in the form of an overscan occurred. Comparing areas affected by overscan with a bulk material, a change in defect density from 1.17% in the bulk material to 5.32% in the overscan regions was observed, highlighting the need to reduce overscan areas in manufacture.

In situ low-cost and adaptable braze tool evaluation system with vision analysis

In this study, a low-cost, flexible process monitoring system for focussed arc tungsten inert gas welder tool electrodes is presented. The system is capable of acquiring and processing geometric data that are used to define acceptance criteria. This system is used to characterise tool wear phenomena in repetitive brazing runs ensuring that the tool is operating within its process window. It was found, using this apparatus, that a build-up of zinc oxide, originating from the zinc-coated steel, on the braze tool prohibits arcing and delivery of a good quality joint. The vision system can identify build-up of this layer before failure of ignition. Measurements showed that after 990 s of continuous arcing, the zinc oxide build-up would cause system malfunction. The vision system gave a warning of zinc build-up at 577 s.

Towards in-situ process monitoring in selective laser sintering using optical coherence tomography

Selective laser sintering (SLS) enables fast, flexible and cost-efficient production of parts directly from 3D CAD data. However, compared with more established machine tools, there is a marked lack of process monitoring and feedback control of key process variables to optimize production parameters in-situ. We apply optical coherence tomography (OCT) to evaluate components produced by SLS and suggest a route for its application in in-situ process monitoring within the SLS tool for real-time monitoring of the SLS process for assurance, or even dynamic correction of defects during the build. OCT is shown to be a viable technique for evaluation of both surface and sub-surface features built into a part either by design or from poor sintering or non-homogeneous powder spreading. We demonstrate detection and quantification of surface defects on a ~30 μm scale in a Polyamide (PA2200) part, resolving ‘built-in’ fine features within a 200 to 400μm depth below the surface.