Date of Award

Winter 2022

Project Type


Program or Major

Mechanical Engineering

Degree Name

Master of Science

First Advisor

Yannis Korkolis

Second Advisor

Jinjin Ha

Third Advisor

May-Win Thein


This thesis details the design and manufacturing of a custom testing machine for thin sheet metal undergoing continuous tension-compression (CTC) loading. The CTC machine concept is based on using a set of intermeshing dies to continuously support the thin sheet specimen against buckling during the test. The upper set of dies are held in place by a pneumatic actuator of 25 kN capacity. The dies can move freely relative to one another, so that both tension and compression can be performed in the same setup. A hydraulic actuator of 50 kN capacity displaces one side of the intermeshed dies. The stroke of the actuator is 63.5 mm, and based on the geometry of the CTC machine and specimen selected, that results into 20% strain in compression and 75% in tension. Strain is measured on the specimen itself using high-elongation strain gages, capable of reaching up to 20-30%, depending on the material. However, based on the hardware currently used, only +/- 5% strain can be read; this is an issue that can be easily fixed with updates to measurement hardware. Chapter 1, which is the introduction, includes the motivation behind this work, as well as other concepts that have been implemented to achieve similar results. Chapter 2 describes the mechanical design of the CTC machine. It includes details of the machine design and functionality, and the strength calculations performed to verify its correct and safe operation. Chapter 3 details the custom data acquisition and control system that was developed. It includes details of the sensors and circuity used for the CTC machine. It also discusses the user interface of the control software, and the pre-programmed functionality of the CTC machine. Chapter 4 describes a full suite of verification tests that were performed on both the CTC machine and CTC specimen geometry, to ensure that the data acquired is accurate and reliable. After the verification testing, Chapter 5 describes a series of cyclic experiments performed on a variety of thin metallic sheets. The research chapters of the thesis are capped by Chapter 6, which discusses a non-linear kinematic hardening model of the Chaboche family, which can be used to replicate the results of the cyclic experiments. Finally, Chapter 7 provides a summary of this work, the main conclusions, as well as proposed future extensions and improvements of the CTC machine. Returning to Chapter 4, the verification tests performed for the CTC machine itself involve tension tests on ASTM E8 specimens using both the present machine and a MTS Landmark 370 servohydraulic loading frame. The agreement is found to be excellent. The CTC specimen geometry is different from the standard ASTM E8 dogbone specimen one, to further prevent buckling during compression. It is confirmed though that the CTC specimen geometry provides identical results to the standard ASTM E8 during tension testing. In summary, the results listed in Chapter 4 show favorable agreement between the two machines, as well as the two specimen geometries. The cyclic experiments discussed in Chapter 5 are performed on a variety of thin metallic sheets: aluminum alloy AA6022-T43 as well as EDDQ, JAC-270D, DP590, DP980 and DP1180 steels. An example of a cyclic experiment is: straining a specimen in tension to +1% engineering strain, reverse loading to -1%, forward loading to +3%, etc. as in -3%, +5%, -5%, then back to 0%. Another experiment is cyclic loading between two equal and opposite strain values, e.g., +/- 2%, for N number of cycles. A noticeable trend with all the materials tested is the amount of tension/compression asymmetry, i.e., where the compressive flow stress is higher than the one in tension for pure compression or tension tests. The Chaboche model in Chapter 6, is calibrated for DP980 steel. An automated parameter determination algorithm, implemented in Matlab, is also described. The code produced is meant to provide the user with an initial best fit to the experiment, so that the user can then improve the fit further as desired, e.g., by manual adjustments. It is expected that utilizing this model in numerical simulations of sheet metal forming processes that include unloading and/or cyclic loading can yield accurate predictions of the springback expected.