Beyond Trial and Error of Sinter-Based Additive Manufacturing

Mastering Sinter-Based Additive Manufacturing with Predictive Simulation

Sinter-based additive manufacturing processes, such as Metal Binder Jetting (MBJ), Metal

Injection Molding (MIM), and Cold Metal Fusion (CMF), are revolutionizing the way we

produce complex metal components. By decoupling the shaping of the green part from the

metallurgical bonding of the metal, these technologies offer unprecedented scalability and design

freedom.

But there is a catch: the furnace is a black box.

In all sinter-based AM processes, parts are printed or molded in a highly porous "green" state. To achieve their final mechanical properties, they must undergo thermal debinding and hightemperature sintering. During this cycle, the metal powder essentially turns into a highly viscous paste. Driven by surface tension, the particles fuse together, closing internal pores. This results in massive volumetric shrinkage, often up to 20% linearly.

As the material softens and densifies, it loses its structural integrity. Gravity takes over, leading to severe slumping, sagging, and anisotropic distortion. For years, compensating for this distortion meant relying on expensive, time-consuming physical trial-and-error. Today, as we will explore in this post, predictive simulation has fundamentally changed the game.

The Simulation Advantage: By solving the viscoplastic continuum theory of sintering mathematically, engineers can predict exact dimensional shrinkage, pinpoint gravity-induced drooping, and evaluate geometric suitability in minutes, entirely eliminating the need for wasted physical furnace runs.

Validating Reality: A 316L Tee Connector Case Study

To demonstrate the accuracy of modern sintering simulation, we replicated the numerical and experimental work published by Torresani et al. in their paper, "Sintering model for predicting distortion of additively manufactured complex parts."

The subject were 316L stainless steel "Tee Connectors" (Geometry 1-3 ). This specific geometry is notoriously difficult to sinter because it features a heavy vertical tube intersecting an unsupported horizontal tube. In a physical furnace, the weight of the structure causes the horizontal tube to sag, turning its perfectly circular cross-section into a collapsed ellipse.

Figure 1: The thermal furnace profile simulation input. The software tracks the exact time-temperature dependencies, ensuring the Arrhenius-driven shear viscosity and surface tension physics trigger at the correct stages of the heating cycle.

Using a fully coupled thermo-mechanical solver (Rheona), we inputted the exact initial green

density (55.1%) and the precise thermal cycle. Crucially, by carefully calibrating the material's

shear viscosity and surface tension, the digital model behaved exactly like the real-world physical part.

Comparing Simulation to Empirical Data

The results of the simulation were nothing short of astounding when compared side-by-side with the physical CMM (Coordinate Measuring Machine) data from the Torresani publication.

1. Volumetric Densification

In the physical experiment, the MBJ 316L parts achieved a sintered relative density of 99.25%. Our simulation seamlessly tracked the pore closure throughout the thermal cycle, locking in a final predicted density of 99.98%. This allows to conclude that the mathematical densification rate, driven by grain boundary diffusion and Laplace pressure, is highly accurate.

Figure 2: Relative Density Map and shrinkage animated over time during sintering. The simulation confirms full metallurgical densification across the geometry, shifting from a porous 55.1% green state to a solid >99% state. Total simulation time 7minutes.

2. Gravity-Induced Distortion and Droop

The true test of a sintering simulation is not just how much it shrinks, but how it warps. The

physical G1 part in the study started with a green Z-height of 63.71 mm and shrank to 50.29 mm, a total Z-shrinkage of approximately 13.42 mm.

Our simulation achieved a final Z-shrinkage of -13.34 mm. We predicted the macroscopic vertical displacement within a fraction of a millimeter of the real-world furnace run.

Figure 3: Distortion and drooping comparison publication vs. the simulation results of the RheonaTM platform; the simulation beautifully captures the displacement, revealing the heavy pull of gravity on the upper vertical tube as the material softens.

The Power of Prediction: Aiding in Setter Design

Why does predicting this droop matter so much for MBJ, MIM, and CMF? Because knowing how a part will fail allows you to prevent it entirely before ever turning on the furnace.

When engineers can visualize a 13.34 mm collapse in a digital environment within minutes, they are empowered to design highly precise setters (live supports). A setter is a custom-machined ceramic or printed shrink-matched support structure placed under the overhangs of a green part.

Because the simulation outputs the exact displacement vectors of the sagging horizontal tube over time, an engineer can export the distorted mesh, invert the deformation, and design a conformal setter that perfectly cradles the horizontal tube as it shrinks. Instead of running three different physical experiments with randomly guessed support structures, the optimal setter is designed on the first try.

Figure 4: Simulation results from the RheonaTM platform for Geometry 3 show distortion and drooping; the simulation accurately captures displacement, though this variant exhibited minimal "droop".

Conclusion

Sinter-based additive manufacturing is the future of mass-producing complex metal components. However, the economics of these processes rely heavily on first-time-right manufacturing. Running iterative trial-and-error cycles in industrial hydrogen or vacuum furnaces takes days and wastes thousands of dollars in powder, binder, and energy.

As demonstrated by our 316L Tee Connector case study, modern FEA simulations leveraging the continuum theory of sintering are no longer just academic exercises. They are robust, deployable engineering tools. By predicting geometric suitability, validating densification, and aiding in precise setter design beforehand, simulation transforms the furnace from an unpredictable black box into a tightly controlled manufacturing asset.