How believable are published laboratory data? A deeper look into system-compliance and elastic modulus

• 作者: Aly Abdelaziz Giovanni Grasselli
• 来源期刊: 岩石力学与岩土工程学报（英文版）
• 年，卷（期）: 2021,（3）
• 页码范围: 487-499
• 页数: 13页
• 语种: 英文

摘要

Elastic modulus (E) interpretation is debatable with limited literature detailing the impact of system-compliance. To address this impact, a comprehensive testing schedule using an aluminium 6061 (Al) sample is carried out on several systems under various test setups. Al is chosen as it is extruded and adheres to well defined shape tolerances and elastic properties. A robust method, using the Savitzky-Golay filter, is introduced to identify significant slope changes in the stress-strain curve. Since the load in the test system is well defined, the recorded deformation is corrected to the expected value of Al resulting in a system-compliance factor. The results across the testing systems and test setups showed significant variance, with the recorded E always lower than the anticipated EAl. The number of compo-nents within the system over which the deformation is measured had the most significant impact, lowering the expected E by up to 50％. Additionally, the system-compliance factor is inconsistent across different systems and setups. Thus, it is evidently proved that each setup must be separately evaluated for its system-compliance and that no single value exists across systems and setups. The findings are then projected onto a series of uniaxial compressive strength (UCS) tests carried out on Stanstead granite (SS GR) samples. The corrected Et50 and Eavg values for system-compliance of the samples are within ±1％for each system as opposed to being ±50％ pre-correction. The findings conclude that it is deemed necessary and of utmost importance that the deformation be corrected to accommodate the system-compliance to obtain reliable results.