Simulation of Inertia Welding process in Abaqus | Fortran Subroutines and Python Scripts

Introduction

The aerospace, automotive, and manufacturing industries widely use inertia welding as a critical joining process to create high-strength, defect-free welds. The inertia welding is a solid-state process that utilizes kinetic energy to join two metal parts together. This tutorial offers a comprehensive guide to simulating inertia welding process using Abaqus, a powerful Finite Element Analysis (FEA) tool. The tutorial focuses exclusively on Abaqus, showcasing its capabilities in modeling complex thermal and mechanical phenomena. By integrating Fortran subroutines, the simulation models frictional heating, temperature distribution, and material behavior under varying thermal conditions with high precision. Engineers utilize Python scripts to automate the remeshing process and generate the model in Abaqus/CAE automatically, reducing manual effort.

Designed for researchers and engineers, this guide covers the setup of an Abaqus simulation for inertia friction welding, as a type of rotary welding. It details how to define axisymmetric models, apply material properties, incorporate custom subroutines, and automate processes with Python. By the end, you’ll have a robust methodology for analyzing inertia welding process efficiently. Simply set the problem parameters in the provided scripts, run them in Abaqus, and let the simulation execute automatically.

In this project, we explain step-by-step how to model inertia welding in Abaqus, based on a example from the Abaqus documentation and a well-known paper. The advantage of this tutorial is that the subroutine is explained line-by-line, and the modeling process is detailed step-by-step. Therefore, by making adjustments to this project, you can simulate your own problem. To simplify this, we modified the Python code and converted it into a parametric form. As a result, you can change the inner and outer radii within the script and model your problem with minimal effort.

The inertia friction welding process​

The inertia friction welding process is a complex task that needs multiple stages as follows:

1. Preparation: The two parts to be welded are aligned and held in place, with one part stationary and the other mounted on a rotating chuck connected to a flywheel.
2. Acceleration: The rotating chuck and flywheel are accelerated to a high speed, storing kinetic energy.
3. Forging: The rotating part is brought into contact with the stationary part, and the stored kinetic energy is converted into heat through friction. This heat softens the metal at the interface.
4. Welding: As the flywheel slows down, the pressure between the parts increases, forcing the softened metal to intermix and form a solid bond.
5. Cooling: The welded joint is allowed to cool, solidifying the bond.

To perform such a process accurately, we need to adjust the flywheel velocity, pressure, and welding time precisely to ensure the desired weld is achieved. Designing these parameters experimentally is a challenging task that requires significant cost and time, often making it impractical. To address this problem, numerical simulations have gained attention in recent years for efficiently analyzing inertia welding process.

Overview of the inertia welding simulation​

Inertia welding process involves pressing a rotating cylindrical component against a stationary one, applicable to hollow cylinders. Friction at the interface generates heat, raising the temperature until the materials become pliable enough to forge a joint. Simulation of inertia welding process in Abaqus involves capturing the interplay of frictional heat generation, thermal conduction, and material deformation. Due to large deformations, remeshing is necessary, adding complexity to the process. The inertia welding simulation workflow, implemented in the provided Python scripts, consists of the following stages:

1. Model Setup
   • Define a 2D axisymmetric model to reduce computational demand while preserving accuracy.
   • Specify material properties such as temperature-dependent thermal conductivity, specific heat, and density.
2. Thermal Analysis
   • Solve the heat transfer problem to compute the temperature distribution over time.
   • Include heat generation due to friction and boundary heat loss through convection.
3. Mechanical Analysis
   • Assess thermal stresses and deformations resulting from temperature gradients.
   • Validate weld seam quality by ensuring the temperature reaches the forging threshold.

Accordingly, key steps in the Abaqus simulation of inertia welding process must be considered, as discussed in the following.

Key Steps in Abaqus Simulation

The key steps in the Abaqus simulation of inertia welding process are:

Preprocessing

• Geometry Definition: The cylindrical specimens are modeled as axisymmetric parts in Abaqus/CAE to simplify the problem while accurately capturing the welding process. The geometry consists of two axisymmetric cylinders representing the welding specimens, with the contact surface defined at their interface.
• Material Properties: Temperature-dependent material properties such as thermal conductivity, Specific heat, and density, could be applied. Material properties are assigned via Python scripts, ensuring precision and efficiency.
• Boundary Conditions and Loads: Rotational motion and axial pressure are applied to the top specimen. The bottom part is fixed. Friction at the interface is modeled using a temperature- and velocity-dependent friction coefficient.
• Meshing: A finer mesh is used at the contact interface to capture steep temperature and stress gradients. Mesh refinement ensures computational efficiency without compromising accuracy.

Customization Using Fortran Subroutines

FRIC and UEL Fortran subroutines enhance simulation precision as discussed below:

• Frictional Behavior: The FRIC subroutine defines frictional behavior as a function of multiple parameters.
• Flywheel Effects: The UEL subroutine models the rotational inertia of the flywheel.

The setup process compiles and links these subroutines, seamlessly integrating them with Abaqus.