Multiple long-term constant rate of strain (CRS) tests were performed on Zegveld peat from the Netherlands. The aim of performing these tests is to check the validity of the Isotach Framework. Different isotachs can be visualized by changing the applied displacement rate during CRS testing. The trajectories of these isotachs are determined to conclude on both the level of parallelism as well as the mutual distance between different isotachs. The results show a high degree of parallelism. The mutual distance between the different isotachs remains largely unchanged. Transient behaviour is observed around a change in strain rate.

The obtained results show a decent decent of parallelism within the studied strain rat ereegime. The mutual distance between the different isotachs remains largely unchanged within this strain rate regime. However, the results show that the trajectory of the isotachs corresponding to the lowest strain rate seem to diverge from those at higher strain rates. In practice, diverging isotachs result in stress-dependency of creep. Furthermore, it is concluded that the mutual distance between different isotachs increases with a decrease in strain rate. Additional CRS tests underline this observation. This increase in distance with lower strain rates in practice results in non-constant creep behaviour on logarithmic time scale, giving rise to tertiary creep. The observations made are important since field strain rates are shown to be a few orders of magnitude lower than those applied in conventional CRS tests.

The performed CRS tests showed transient behaviour around a change in strain rate. The time needed for the soil to fully adjust to the new strain rate increases with decreasing strain rate. This could give rise to invalid parameter determination since the soil’s state has not yet moved to the isotach corresponding to the new strain rate. The obtained results of the step-changed CRS tests are simulated using the NEN-Bjerrum Isotach model, the abc-isotach model, the Soft Soil Creep model and the MIT Elasto-Viscoplastic model of Yuan and Whittle. Overall, a satisfactory fit is found between the models and the actual CRS test data. The MIT Y&W EVP model and the Soft Soil Creep model are capable of accurately simulating the observed transient behaviour around a change in strain rate.