Investigation of the parallel gradation method based on the response of track-bed materials under cyclic loadings
2020
Abstract Ballast/fines mixture is often found in railway substructure and the overall behaviour of tracks is strongly dependent on the mechanical behaviour of this mixture. Since directly testing such mixture with large grains is limited with common laboratory equipment, the consideration of a model material at smaller size is advisable. Parallel gradation method is widely used for this purpose. This study assesses the validity of this method in the case of ballast/fines mixture. Large-scale cyclic triaxial tests were carried out on the ballast/fines mixture at six volumetric coarse grain contents. The results obtained were analysed, together with those obtained previously from small-scale triaxial tests on a microballast/fines mixture whose microballast grain size distribution was determined by applying the parallel gradation method. The cyclic parameters (permanent strain and resilient modulus) were obtained for the two types of mixture. Results show that for all mixtures two distinct soil fabrics can be identified according to the variations of permanent strain and resilient modulus with the volumetric content of coarse grains fv: a fine-fine contact structure for fv ≤ 20% and a grain-grain contact structure for fv ≥ 35%. In the case of fine-fine contact structure, the permanent strains and resilient modulus values of the ballast specimens are consistent with those of the microballast specimens, evidencing the validity of the parallel gradation method. In the case of grain-grain contact structure, the permanent strains and resilient modulus values are found to coincide globally at the two scales, also justifying the validity of the parallel gradation method, the slight differences between the two scales being attributed to the irregular grain sliding and the distribution of the fines soils.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
37
References
4
Citations
NaN
KQI