Accuracy analysis of Young’s modulus and stress-strain curve in the elastic stage of materials using Hopkinson bar experimental method

The stress-strain data obtained from split Hopkinson pressure bar (SHPB) tests include both strain rate effects and structural effects, where the structural effects result in non-uniform stress in the elastic phase of the stress-strain curve. The elastic phase is a critical focus of study for materials like concrete with low sound velocity or certain metals under high strain rate loading conditions. In this paper, we focus on one-dimensional rod systems and employ one-dimensional elastic incremental wave theory to derive analytical expressions for stress-strain curves and Young’s modulus under one-dimensional stress wave conditions with linear incident waves. We investigate the effects and mechanisms of stress difference and velocity difference at both ends of the specimen on the accuracy of stress-strain curves and Young’s modulus. Furthermore, we provide a method for determining stress-strain curves and tangent Young’s modulus during the elastic phase for arbitrary incident waveforms. We analyze the influence of the incident wave slope and shape characteristics on the stress uniformity in specimens and stress-strain curves. We establish the inherent relationship between stress uniformity and experimental stress-strain curves, and clarify the relative accuracy and applicability conditions of tangent modulus and secant modulus. The results indicate that stress uniformity is a key factor affecting the accuracy of stress-strain curves and Young’s modulus. However, the accuracy of Young’s modulus is not solely dependent on the change in stress difference at both ends of the specimen; it is also related to the factors such as the incident wave slope, shape characteristics, and the elastic segment range of the specimen. An increase in the linear wave slope leads to a greater difference between the tangent modulus and the secant modulus from the actual values. For larger slopes, the accuracy of the secant modulus is higher than that of the tangent modulus. When the incident wave shape is considered as a reference, curves with low initial slopes, such as sine waves, have higher accuracy for the tangent modulus compared to the secant modulus, whereas curves with high initial slopes show the opposite trend. For concrete specimens, we verify the influence of incident wave slope on Young’s modulus and evaluate the maximum incident wave slopes for concrete specimens to reach accurate values, which are 0.128 MPa/μs for the tangent modulus and 0.319 MPa/μs for the secant modulus.