Evaluating the effects of selective laser sintering processing parameters on the melt pool dimensions of tungsten carbide through numerical thermal modeling

Olivier, Carlo Martin (2019-04)

Thesis (MEng)--Stellenbosch University, 2019.

Thesis

ENGLISH ABSTRACT: The selective laser sintering (SLS) industry is a relatively novel industry within the broad spectrum of available additive manufacturing (AM) technologies. This layer wise processing technology has gained popularity over recent years, due to its ability to produce low volume, highly complex components made from difficult to process metal materials. A major driver for this industry is the development of materials for new or improved applications. However, each time manufacturers want to add a new material to their machine, a specific set of processing parameters needs to be developed for that material in order to ensure that high density, high strength components are produced. The primary aim of this developing technology is to produce high quality, low costs components, yet it still struggles to overcome inefficiencies relating to the product development phase. Currently, large quantities of raw material are wasted during the machine-material calibration stage in the SLS industry. The aim of this study was to investigate the possibility of improving or replacing the current inefficient research and development (R&D) methods of the SLS industry with numerical modeling. In this study, a numerical thermal model was developed, which manufacturers can use in their production facility, in order to predict changes in the melt pool geometry for new or unconventional materials by changing certain processing parameters. This will allow them to make decisions regarding other machine input parameters such as the scan track hatch spacing and layer thickness before performing any physical calibration tests. Tungsten carbide (WC) was selected as the main processing material for this study, as it is still a relatively novel addition to this technology field. WC has an array of unique material properties, which makes it ideal for application in industries such as the mining, tooling and oil and gas industries. A numerical thermal model was used to develop a feasible set of processing parameters specifically for processing tungsten carbide. The results from the numerical model were validated by comparing the simulated melt pool geometry to that of experimentally produced samples. The simulation was able to predict the change in the melt pool geometry with the change in processing parameters successfully. However, the result comparisons revealed that there were still deviations present between the simulated and the actual measurements. The highest deviations between the simulated and actual melt pool geometry measurements were found to be 24% and 50% for the scan track width and penetration depth, respectively. Additionally, a cost comparison revealed that the simulated calibration method was 46.5% more affordable than that of the conventional calibration techniques used in the industry. Although, the findings of this study are not enough to suggest that numerical modeling should replace traditional calibration methods completely, it rather motivates using numerical thermal modeling in addition to current techniques in order to assist with decision making and reduce the scope of the calibration process. This scope reduction may lead to cost savings and increased sustainability, which will ultimately improve the efficiency of the SLS process chain.

AFRIKAANSE OPSOMMING: Die selektiewe laserstraalsintering (SLS) industrie is ‘n relatiewe nuwe industrie in die breë spektrum van beskikbare byvoegende vervaardingtegnologië (AM). Hierdie tegnologie, wat staatmaak op die laag-vir-laag opbou van materiaal, het oor die laaste paar jare in gewildheid gegroei wat grotendeels toegeskryf kan word aan die tegnologie se vermoeë om komponente met ‘n hoë vlak van kompleksiteit te vervaardig vanuit materiale wat normaalweg moeilik is om te verwerk. ‘n Groot dryfmiddel agter hierdie industrie is die ontwikkeling van materiale vir nuwe of verbeterde toepassings. Tog, elke keer as vervaardigers nuwe materiaal tot hulle masjiene wil toevoeg, moet hulle eers ‘n stel spesifieke prosesparameters ontwikkel wat sal versker dat hulle komponente met ‘n hoë digtheid en sterkte kan produseer. Die primêre doel van hierdie ontwikkelende tegnologie is om hoë kwaliteit, lae koste komponente te vervaardig, maar dit sukkel nog om ondoeltreffenhede in die produkontwikkelingfase te oorkom. Huidiglik word daar groot hoeveelhede rou materiaal tydens die masjienkalibrasiefase in die SLS industrie gemors. Die doel van hierdie studie was om die moontlikheid te ondersoek om die huidige ondoeltreffende navorsing en ontwikkelings (R&D) metodes van die SLS industrie met dié van numeriese metodes te vervang. In hierdie studie is ‘n numeriese termiese model ontwikkel wat vervaardigers in hulle vervaardingingsfasiliteite kan geburik om die veranderinge in die smeltpoelgeometrie vir nuwe of onkonvensionele materiale met 'n verandering in sekere prosessparameters te kan bepaal. Hierdie sal hulle in staat stel om besluite te maak met betrekking tot die masjien intree parameters soos die skandeerspoorspasieëring en laag dikte, voordat enige fisiese kalibrasietoets uigevoer word. Wolframkarbied (WC) is gekies as die hoof vervaardigingsmatriaal vir hierdie studie aangesien dit nog ‘n relatiewe nuwe toevoeging tot die tegnologieveld is. WC het ‘n hele klomp unieke materiaaleienskappe wat dit ideaal maak vir industrietoepassings soos mynbou, implemente, olie en gas. 'n Numeriese termiese model is gebruik om ‘n uitvoerbare stel prosesparameters, wat spesifiek op wolframkarbied van toepassing is, te ontwikkel. Die resultate van die numerise model is gevalideer deur die gesimuleerde smeltpoelgeometrie met dié van eksperimentele monsters te vergelyk. Die simulasie was in staat om die smeltpoelgeometrie met die verandering in prosesparameters suksesvol te voorspel. Tog, het die resultate van die vergelykings getoon dat daar nogsteeds afwykings tussen die gesimuleerde en werklike afmeetings teenwoording is. Die grootste afwykings tussen die gesimuleerde en werklike smeltpoelgeometrieafmeetings is gevind om tussen 24% en 50% vir die skandeerspoorwydte en penetrasiediepte, onderskeidelik te wees. Daarbenewens het ‘n kostevergelyking getoon dat die gesimuleerde kalibrasiemetode 46.5% goedkoper as die tradisionele kalibrasietegnieke van die industrie is. Alhoewel die bevindinge nie genoeg is om aan te beveel dat numeriese modelleering heeltemal die tradisionele kalibrasiemetodes moet vervang nie, dien dit eerder as motiveering dat numeriese modelleering tot toevoeging van huidige kalibrasiemetodes kan dien met die doel om die omvang van die kalibrasieproses te verklein. Hierdie verkleining in omvang mag moontlik tot kostebesparings en ‘n verbetering in volhoubaarheid lei, om tot einde die doeltreffendheid van die SLS prosesketting te verbeter.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/106093
This item appears in the following collections: