Design framework for Selective Laser Melting applied to an aluminium bracket

[1] original bracket

How do you combine design guiderules for SLM and topology optimisation? How do you prioritise the design optimisation tasks while minimising the design iterations?
​
Industrial designers do not yet have a clear framework to review, redesign, and optimise existing designs in order to take full advantage of the benefits that SLM can offer.

In this post, we present the methodology used to redesign an aluminium bracket for SLM in order to save weight while maintaining performance. The key objective is to take into account the manufacturing constraints, user’s requirements and take advantage of technology’s design freedom.

---

[1] von Mises stress graphical representation

[1] Bracket after topology optimisation

​The proposed methodology has five main steps.
1. Capture and analysis of requirements: capture of functional specifications: structural (loads) requirements, DFM rules, material characteristics,…
2. Conceptualisation: creation of initial CAD concepts, FEA analysis (loads map), topology optimisation,
3. Manufacturability assessment: review of concepts for manufacturability for SLM according to process characteristics (layer thickness,… ), SLMed material characteristics (anisotropy, surface roughness… ) best efforts to minimise of post-processing (supports,…).
4. Evaluation of proposed designs against initial functional and user requirements.
5. Decision: multi-criteria decision making to pick most suitable option.

​1. Capture and analysis of functional and user’s requirements

The functional redesign [link] of a bracket conventionally machined out of an aluminum alloy 6082-T6 block using CNC milling aims to

• reduce weight by 20%;
• maintain equal or superior performance;
• prevent permanent deformation.

The redesigned bracket must:

• provide the same recess features as in the original design;
• operate using the existing clamping components;
• be compatible with the interfaces of the existing mounting rail in the bottom of the bracket;
• handle three orthogonal, non-concurrent shock loads equal to 1200N.

Given its high dimensional accuracy (compared to other AM techniques), its reliability and the large amount of feedstock commercially available, a laser-based powder bed fusion machine is selected to build the redesigned component using aluminium alloy (AlSi10Mg) powder.

[1] concept designs

​2. Concepts

Given SLM design flexibility, original concepts can adhere closely to the initial design specifications. Next, FEA is used to predict their stress distribution and deformations under different conditions. Factors such as boundary conditions, materials and of loads distribution can be varied for comprehensive analysis. Once FEA identifies the maximum stress and displacements, topology optimisation is carried out to produce a strong organic structure that efficiently uses material only where it is needed.

In our scenario, the first step is to carry out FEA and topology optimisation of the original bracket as a starting point.
From this, a wide range of 3D models are produced using a standard CAD software in order to explore and document a catalogue of concepts. In turn, FEA is carried out to analyse whether these concepts can bear the mechanical forces without failure.

​3.  Manufacturability

Considering the capability and constraints of SLM, major or minor modifications of the designs may be required. The general following guidelines can be applied:

• avoid closed cavities;
• promote self-supporting structures;
• choose proper clearances;
• consider surface finish;
• consider post-processing.

​In addition, the processing capabilities of the machine and materials characteristics need taken into account.

At the next step, models are extrapolate from concepts 3D CAD and their FEA analysis (ie von Mises stress values, location, etc) to take into account the structural constraints and the capabilities of the SLM machine.
 
This leaves us with a set of potentially viable designs that fit design, functional, manufacturability and structural requirements.

​4. Performance evaluation

To verify the performance of the models (strength, fatigue, stresses, …), a variety of tools can be used.
If some requirements are not met after verification, designers can make minor changes in the design and remodel it. However, if most of the requirements have not been met, designers should start from step 2 and build a new concept.

In our scenario, FEA indicates that maximum von Mises stresses do not exceed the strength of the material. Besides, machine and manufacturing constraints are met (ie no cavities or unsupported structures, etc).

​5. Decision making

Where various concepts fit all the requirements, multi-criteria decision analysis (MCDA) can assist in choosing the most suitable option. MCDA is concerned with defining problems, and forming decision involving to different criteria.
 
In this case, the analytical hierarchy process (AHP) is used to decompose the decision into a hierarchy of easier sub-questions. 

The objective here is to find the best optimised design with respect to the following criteria listed in order of importance by the user:

• light-weight,
• strength,
• minimum displacements,
• manufacturing cost
• and surface quality.

​Using an SLM machine with aluminium alloy AlSi10M powder, the final chosen option is Design 2. It results in decreasing the original weight of the bracket from 70g to 40g, a 43% reduction while maintaining performance.

References
K. Salonitis, S. Al Zarban, Redesign Optimization for Manufacturing Using Additive Layer Techniques, CIRP 25th Design Conference Innovative Product Creation,  Procedia CIRP 36 ( 2015 ) 193 – 198, doi: 10.1016/j.procir.2015.01.058