The Dynamic Hollow Cylinder Apparatus (DHCA) is renowned for its ability to subject soil samples to various cyclic stress paths, allowing for the investigation of the dynamic behavior of soils under complex cyclic loading conditions. This note explores the errors arising from the stress non-uniformity along the height of DHCA samples and examines their impact on the measured soil dynamic properties. After discussing the two globally used DHCA types, this study presents the stiffness and damping of sand derived from a set of Dynamic Hollow Cylinder experiments covering a wide range of dynamic stress paths and shear strain amplitudes. It is highlighted that the deviation of the results from established degradation models is primarily attributed to the errors associated with the stress non-uniformity, leading to up to a tenfold underestimation within the medium strain range. A simple correction to the shear stress amplitude calculation is proposed to minimize the impact of stress non-uniformity and improve the accuracy of the test results.

The strength and deformation characteristics of compacted soils are typically evaluated using triaxial compression tests on specimens that are compacted relatively uniformly in a laboratory. Soil compacted in the field using vibration rollers is nonuniform in the vertical depth direction; this is because the gradient of the dry density and saturation degree in the depth direction of each compaction layer is large, owing to the distribution of the load transmitted from the contact surface. As a quality control method for earth-filling works, nondestructive inspection indices-such as the soil stiffness index-are applied in some cases, and the average value of the nonuniform compaction layers is measured. Further, unsaturated specimens, which were retrieved from the test fill yielded during field compaction tests, and specimens obtained via compaction using the same soil as that in the laboratory were prepared, and then triaxial compression tests were performed. Local displacement transducers were installed to obtain local deformation characteristics, based on the vertical depth of the specimen, which were then compared with the average deformation characteristics of the entire specimen. The results show that the local low-stiffness significantly affects the overall average; this makes it non-negligible because the specimen compacted in the field is nonuniform. However, because field-compacted soil has a local low-stiffness near the surface layer compared with laboratory-compacted soil, evaluations based on the nondestructive inspection of the surface layer may result in underestimations.