The improvement of the loosely deposited sands and silts with the Rapid Impact Compaction technique on Brežice test sites

• Improvement of sands/silts with Rapid Compaction was investigated at two sites.
• Test sites were investigated pre/after compaction by in-situ and laboratory methods.
• Effectiveness of the Rapid Compaction method was determined.
• Correlation between DPSH and LWDPCT measurements was established.
• Coal particles influence on the success of compaction was highlighted.

A possibility of in-situ ground improvement with the Rapid Impact Compaction (RIC) technique was investigated at two test sites where the construction of the high embankments is planned. The location is characterized by up to a 5 m thick top layer of loosely deposited silty sands with a medium to very high concentration of coal particles, which strongly influences the in-situ water content. Materials at both test sites were investigated before and after compaction with in-situ measurements (DPSH, LWDCPT, CPTU and DMT) and the laboratory tests on samples taken from the excavation pits. Each test site was divided into four fields, which had the different compaction raster and transferred energies applied to them. It was concluded that for the optimal compaction more than 50% of the field has to be covered with the compaction points. The presence of the coal particles did not affect the results of compaction at the time of the test since the water content was close to the optimum. However, the water content varies seasonally and this could potentially have a strong influence on the levels and uniformity of compaction. The DMT and CPTU tests captured the effect of compaction but since only a few measurements were executed, the results were inconclusive, which was nevertheless surprising. On the other hand, a good correlation was established between the DPSH and LWDCPT tests, which both clearly indicated the difference between different fields and could be used for compaction control. It was also shown that the DPSH test is insensitive for very loose materials (blow count values N1(60)30SB lower than 3) and in that case the LWDCPT tests should be used instead.