**3. Experimental sand compression with oedometer**

Compression tests were performed using computer controlled (fully automatic) oedometer test device. The same sand soil used for morphology parameters investigation was examined in compression tests with the standard oedometer device (ring height *H* = 3.39 cm; ring diameter *D* = 7.14 cm). Testing program consists of the study of the influence of vertical stress ramp (**Figure 6**) on the compression results, the study of crashing of separate particles, and upgrade of testing procedure, respectively.

**Figure 6.** Vertical load ramp influence for soil deformation.

Obtained results, which are presented in **Figure 6**, clearly show that loading ramp is not having any effect for soil deformation. Due to this reason, it was accepted to use 400 kPa/min loading ramp for rest of all tests with oedometer. From this point, it is necessary to check two things before numerical oedometer modelling:


The solution for the checking of particles crushing is to provide a series of oedometer tests with different maximum vertical stress. After each test, it is necessary to run sieving test and to check granulometry change. If the granulometry composition is not changing, that means, particles with accepted maximum vertical stress loading are not crushed. Small change of mass after sieving test can be revealed, but this happens due to particles loss during sieving procedure. Normally, particle loss can be up to 1%. Obtained sieving tests results after different maximum loading are given in **Table 2**.


**Table 2.** Soil sieve test results after different compression steps [19].

Knowing that tested soil particles cannot have crushing effect (the analysis of obtained results revealed that sand particles are not crushed up to 800 kPa maximum vertical loading) makes more simple DEM oedometer test modelling.

The main finding from the compression test procedure is presented in **Figure 7**. Comparison of the standard and improved procedures consists of two different stages. The first comparison stage involves vertical loading values from 0 to 100 kPa. In this stage, the standard test procedure cannot evaluate immediate settlements [20], since for contact ensurance between porous stone and soil, 10 kPa vertical stress is established. Therefore, **Figure 7** shows results not from 0 kPa.

**Figure 7.** Comparison of soil compression methodologies: blue—standard procedure; red—improved procedure.

The improved testing procedure is better than the standard one, as porous stone is positioned in the calibrated height, selected according to the initial sample height. Performing compres‐ sion test with the improved testing procedure, it is possible to analyse compression curve from 0 kPa.

The standard and improved testing procedures differ in their obtained compression curves values only, and the character of their vertical strain values remains the same. In the standard procedure, a smaller compressed sample deformation (*ε* = 1.6168%) is obtained comparing with the improved testing procedure (*ε* = 3.1763%). The comparison of different compression procedures allow the judicious selection of DEM compression modelling as the preferred methodology since it is an improved testing methodology and allows the analysis of com‐ pression results from 0 kPa.

The soil compression curve mostly obtained in literature is given in **Figure 8** [21].

**Figure 8.** Soil void ratio versus normal stress.

Obtained results, which are presented in **Figure 6**, clearly show that loading ramp is not having any effect for soil deformation. Due to this reason, it was accepted to use 400 kPa/min loading ramp for rest of all tests with oedometer. From this point, it is necessary to check two things

**1.** Does the examined particles having any crushing effect with used maximum vertical

**2.** Where are the errors on computer controlled oedometer testing procedure, because in

The solution for the checking of particles crushing is to provide a series of oedometer tests with different maximum vertical stress. After each test, it is necessary to run sieving test and to check granulometry change. If the granulometry composition is not changing, that means, particles with accepted maximum vertical stress loading are not crushed. Small change of mass after sieving test can be revealed, but this happens due to particles loss during sieving procedure. Normally, particle loss can be up to 1%. Obtained sieving tests results after different

**mm 0 kPa 200 kPa 400 kPa 600 kPa 800 kPa** 2.0–1.18 0.000 0.000 0.000 0.000 0.000 1.18–0.6 2.145 2.003 1.999 1.864 1.816 0.6–0.425 39.593 39.234 40.478 39.298 39.053 0.425–0.3 91.319 90.727 90.984 90.466 90.640 0.3–0.212 65.780 66.430 64.859 65.803 65.380 0.212–0.150 6.188 6.217 6.206 6.185 6.190 0.150–0.125 0.297 0.306 0.318 0.347 0.359 0.125–0.075 0.121 0.153 0.204 0.324 0.358 0.075–0.063 0.026 0.023 0.026 0.030 0.031 <0.063 0.000 0.045 0.059 0.122 0.095 SUM: 205.469 205.138 205.133 204.439 203.922

Knowing that tested soil particles cannot have crushing effect (the analysis of obtained results revealed that sand particles are not crushed up to 800 kPa maximum vertical loading) makes

The main finding from the compression test procedure is presented in **Figure 7**. Comparison of the standard and improved procedures consists of two different stages. The first comparison stage involves vertical loading values from 0 to 100 kPa. In this stage, the standard test

before numerical oedometer modelling:

250 Modeling and Simulation in Engineering Sciences

maximum loading are given in **Table 2**.

**Sieve size, Particles content, g**

stress 400 kPa on provided oedometer tests.

**Figure 6** given results start not from 0 kPa.

**Table 2.** Soil sieve test results after different compression steps [19].

more simple DEM oedometer test modelling.

Analysing soil compression curve given in **Figure 8**, it comes obvious that soil density or void ratio for low stress values does not change. Void ratio change is obtained only when soil is loaded with high stress values. Such interpretation of the compression curve is not reliable due to two reasons:


If the soil sample has loading, unloading and reloading steps, **Figure 8** given compression curve is suitable only for the reloading step. Other authors present soil compression curves with vertical loading from 5 to 10 kPa [22–26]. In this case, it is not necessary to show what happens with the soil compression curve when vertical stress is from 0 to 5 or 10 kPa.
