Experimental Studies and Numerical Modeling of Strain Rate and Temperature Dependent Material Behavior in Dynamic Processes

Abstract

The rate of loading plays an increasingly important role in many technological processes and applications, and even in many everyday events. In all these cases, it would be important to know in advance how the material will react to the loading conditions where the strain rate can exceed what is generally considered ‘normal’ or ‘conventional’. In this thesis, the main objective was to examine technological processes and events involving time and temperature dependent material behavior and to develop modeling concepts based on experimental materials research for simulation purposes. As a rule of thumb, the instantaneous strength of virtually all materials increases with increasing strain rate and decreasing temperature, which can have both positive and negative implications, depending on the case. In the crusher pressure sensor case, the consequence of increasing strain rate is that the obtained pressure values start to increasingly deviate from the correct values, when the gun pressure increases. From the materials science point of view, this problem can be easily solved by proper calibration that accounts also for the strain rate effect, but there are still issues related to knowing the actual strain rate, as will be shown in the thesis. Cold heading, in turn, is a process where metal wire is deformed to the desired shape in a die. With increasing production rates, also in this application the strain rate can reach levels where it must be accounted for. In this work, the properties of the cold heading steel were determined at wide ranges of strain rate and temperature using both hydraulic materials testing machines and the Hopkinson Split Bar testing techniques. Based on the experimental results, an ‘ad-hoc’ material model was developed and implemented in finite element software to be used in numerical simulations. To validate the simulation results, a relatively simple case that could be carried out both numerically and experimentally was chosen. The results show that the agreement between the experimental and simulated results is much better with the developed model that takes into account the strain rate and temperature effects than with the Johnson-Cook model based on the same data, or with the ‘ad-hoc’ model based on the quasi-static data only. There were, however, still some minor differences observed between the experimental and simulated results, which could be attributed mostly to the nonhomogeneous properties of the cold heading steel and the deviations of the steel wire shape and size from the ‘ideal’ ones used in the simulations. The other applications studied in this thesis deal with the optimization of paper machine roll covers with finite element modeling, the effects of microstructure on the dynamic strain aging (DSA) of carbon steels, and the high temperature high strain rate testing of a titanium alloy with the tensile Hopkinson Split Bar technique. In the first case, comparison of the experimental and simulated data shows that a hyperelastic model of the cover material is more suitable for the prediction of the contact conditions between the rolls than an elastic model. The main result of the DSA studies is that under certain strain rate and temperature conditions, the DSA effect can be rather strong and should be taken into account especially when modeling the behavior of this type of materials at elevated temperatures. Finally, the round-robin type high temperature testing with tensile Hopkinson bar devices underlines the importance of the specimen fixing method on the quality of the test results, especially at higher temperatures, as well as the effect of heating time on the usability of different methods, in particular adhesive bonding of the specimen to the stress bars.