Experimental and Numerical Investigation of Part Fabrication by Incremental Sheet Forming

Abstract

In recent years, the demand for personalized products have significantly increased in biomedical, automotive and aerospace industries. However, traditional sheet metal forming processes such as stamping, deep drawing and hydroforming are only suitable for high-volume production due to high initial setup costs and long production lead time. In this research, a relatively new fabrication technique named incremental sheet forming (ISF) process is comprehensively investigated to address the emerging need for customized parts. Major advantages of ISF process are lower forming forces, increased material formability and energy efficiency. However, it produces complex and non-linear strain history that makes it difficult to produce parts with industry standard accuracy. There are multiple variants of ISF process such as single-point incremental forming (SPIF), two-point incremental forming (TPIF) and double-sided incremental forming (DSIF).

In this thesis, incremental micro-forming (µSPIF) process is investigated to improve part accuracy and understand deformation mechanics for its applications in biomedical and micro-electronics industries. An experimental setup is developed in-house that mimics a table-top CNC machine and provides synchronous tool motion in x, y and z directions. Truncated cones and pyramids are experimentally produced on 50 µm thin AL 1100 and AL 5052 foils to quantify the process performance. The parametric analysis showed significant influence of step size and tool diameter on both forming forces and material formability. When the cross-sectional profiles of produced parts are compared with its designed geometry, it showed a deviation of 80-120 µm that could be attributed to the combination of machine compliance and sheet springback.

The capabilities of macro-TPIF process are also explored to investigate and improve its process performance in the production of aerospace parts using 1.64 mm thick AL 7075 material. Besides inferior geometric accuracy, the correct estimation of material squeeze factor is identified as a major research gap in TPIF process. Previous studies have reported material squeeze values of anywhere between 10% to 50% in TPIF/DSIF experiments. However, effective material squeeze obtained in the experiments can be considerably different than its programmed value due to tool deflection and machine compliance. Therefore, a mathematic model is proposed and experimentally validated to calculate an effective squeeze factor based on its programmed value, forming forces and pre-recorded machine compliance. For cone 67 degree part, programmed squeeze factor of 40% results in only 2.3% of effective squeeze in experiments. This effect must be appropriately considered while developing a FE model to obtain accurate results. Further analysis of squeeze factor on heart shape parts showed that increase in material squeeze helped improve geometric accuracy but also accumulated an unwanted bulge at the part bottom.

Finally, it is important to develop a cost-efficient and accurate prediction model to understand the underlying process mechanics. A finite element model is developed that can provide prediction results with good accuracy in minimum computation time. Tensile tests are performed on AL 1100 material to obtain constitutive models. Holloman’s strain hardening law with isotropic hardening provided the best prediction results when benchmarked against experimental data. The techniques of mass and velocity scaling are successfully utilized to artificially accelerate the simulation speed without compromising on accuracy. Forming forces for are predicted with less than 10% error. It is also demonstrated that using techniques such as Rayleigh mass damping and hourglass control can lead to numerical artifacts. So, they must be utilized with utmost care.