ANALYSIS OF DYNAMIC REPLACEMENT COLUMN

Nowadays, the most efficient geoengineering methods of soil strengthening under the construction of civil structures are accompanied by shakings which induce vibrations in soil. The vibrations propagate in all directions, having destructive influence on neighbouring buildings or other infrastructure elements. In some cases, the vibrations may result in excessive effort of construction elements, which may result in cracking, decrease of bearing capacity and stiffness. In extreme cases, the construction may even be destroyed. Vibrations accompany e. g. the dynamic replacement method, dynamic  compaction, prefabricated piles driving and sheet piling driving. Similar phenomena also occur in soil strengthening with a heavy vibratory soil compactor. It is obvious that the range of influence mostly depends on the vibration’s frequency. However, other factors such as soil type (more precisely its deformation and attenuation characteristics), the presence of ground water, the depth of rock or applied energy also play an important role. These latter factors mostly concern the use of vibratory soil compactors. Also, the impact on the people in the influenced buildings should be taken into account, especially when conducting works close to office and residential buildings. In practice, special monitoring is used to assess the real influence of the vibrations on the neighbourhood. It is based on in-situ measurements of amplitude and frequency of vibrations occurring on some elements of the observed structures. On this basis and with the use of standards [1] and [2], the harmful impact of the conducted works is assessed and the appropriate measures are undertaken to prevent the negative influence. In extreme cases, it may result in changing the strengthening method. Moreover, the monitoring may be expensive. Therefore it is crucial to undertake research (both theoretical and experimental) which will facilitate the design by simulating the phenomenon of vibrations propagation and their influence on its surroundings. Field tests will be necessary for the calibration and verification of the adopted model and its parameters.

Nowadays, the most efficient geoengineering methods of soil strengthening under the construction of civil structures are accompanied by shakings which induce vibrations in soil. The vibrations propagate in all directions, having destructive influence on neighbouring buildings or other infrastructure elements. In some cases, the vibrations may result in excessive effort of construction elements, which may result in cracking, decrease of bearing capacity and stiffness. In extreme cases, the construction may even be destroyed. Vibrations accompany e. g. the dynamic replacement method, dynamic  compaction, prefabricated piles driving and sheet piling driving. Similar phenomena also occur in soil strengthening with a heavy vibratory soil compactor. It is obvious that the range of influence mostly depends on the vibration’s frequency. However, other factors such as soil type (more precisely its deformation and attenuation characteristics), the presence of ground water, the depth of rock or applied energy also play an important role. These latter factors mostly concern the use of vibratory soil compactors. Also, the impact on the people in the influenced buildings should be taken into account, especially when conducting works close to office and residential buildings. In practice, special monitoring is used to assess the real influence of the vibrations on the neighbourhood. It is based on in-situ measurements of amplitude and frequency of vibrations occurring on some elements of the observed structures. On this basis and with the use of standards [1] and [2], the harmful impact of the conducted works is assessed and the appropriate measures are undertaken to prevent the negative influence. In extreme cases, it may result in changing the strengthening method. Moreover, the monitoring may be expensive. Therefore it is crucial to undertake research (both theoretical and experimental) which will facilitate the design by simulating the phenomenon of vibrations propagation and their influence on its surroundings. Field tests will be necessary for the calibration and verification of the adopted model and its parameters. Dynamic replacement (DR) is one of the most popular techniques often applied in order to strengthen weak soil under road embankments. The method owes its popularity to a large number of road construction projects being currently realized in Poland. DR column is formed by dropping a pounder (rammer) of a specific shape and weigh of 15 – 30 tonnes. This is usually conducted on a working platform, which makes the use of an 80-tonne crane possible. In the first stage, the pounder is dropped from the height of up to 25 m to form a crater, which is refilled with coarsegrained material. The following drops of the pounder form the column. In Poland, the diameter of the columns varies between 1.6 – 3.0 m and their length is up to 6m [11]. This simple and rapid method allows strengthening of weak soil up to the depth of 6 m. It increases the strength of both cohesive soils (clays, silts), as well as organic and anthropogenic soils. Column formation process is accompanied by vibrations perceptible in the surroundings. Therefore, in most cases, the structures located close to the construction sites are monitored so that a safe drop height (and thus the proper energy) might be selected.
This paper features the results of investigations carried out during stone column formation on G1 section of highway DTŚ. Over 30 000 m2 of soil under road embankment have been strengthened. Soil investigations performed before the beginning of the strengthening (CPTs and boreholes) show that the upper part of the soil consists mainly of soft silty clays and medium sands, and the lower part – the bearing layer – is made of firm clays. In some places anthropogenic soils were encountered from the soil surface up to the depth of 1 m. Columns were formed with a 15-tonne pounder. At least 20 drops were performed from the maximal height of 15 m to form a column. The crater was refilled 7 times during column formation. In this case, the columns of burnt (red) shale measured 2.0 – 2.5 m in diameter and 3.5 – 4.5 m in
length. The spacing was variable, ranging from 4.5 m x 4.5 m to 6.0 m x 6.0 m depending on location. Partially excavated columns on the construction site of DTŚ are shown in Fig. 1. During the construction of embankments, columns settlements were monitored. The observed values were closed to the predicted ones [9].

DR column construction process generates significant waves, which may result in soil deformations and changes of soil parameters. Strength and deformation parameters of the soil surrounding the column change during column formation process, which was examined in the natural scale [7]. There are also horizontal and vertical displacements occurring in the soil adjacent to the column [4], [7]. There is a large body of literature on the subject of waves’ propagation in soil. Many articles are based on the works of R. D. Woods, who described how waves propagate in an isotropic, elastic half space [16]. Among works of Polish authors, there is a remarkable report published by the scientists from Cracow University of Technology [3]. The main focus of these reports was the influence of waves on buildings, neglecting the nature of
waves propagation in soil. Therefore [13] presents the principles that should be applied when measuring the acceleration in construction sites. Two standards, [1] and [2], have been elaborated to assess the results of these measurements. It should be noted that in case of typical constructions, horizontal acceleration components are the most dangerous and vertical components are less destructive, although more perceptible for people. Moreover, another important factor is the frequency of excitations. In case of heavy impacts of the pounder, the frequency of the propagating waves is about 10 Hz, which may be the resonance frequency for some buildings constructions. The literature on the subject of vibrations accompanying the construction of DR columns is more modest. On global level, there was a group of scientists from Florida [5], [14], [15] working on this topic, and in Poland, the subject has been elaborated by scientists from the Silesian University of Technology [4], [8]. The acceleration of the pounder (rammer) used in the dynamic replacement method was measured during the in situ tests described in [14]. Thanks to the measurements of rammer acceleration at the moment when it was hitting the soil, the coefficient of effectiveness of the crane could have been indicated. M. Guanarante et al. [5] used the results of vibration measurement to calibrate a numerical model of stone column formation process and to verify the theoretical model of soil reinforced with the dynamic replacement method [15]. The measurements of vibrations occurring in the DR method can also be applied to protect the structures located within the vicinity of the strengthened area. A bridge [4] or a building structure [6] may be the examples of such structures. If there is a risk that the vibrations may damage them, the impact energy is reduced or it is necessary to construct barriers [6].

Stone columns in DTŚ building site were constructed within the vicinity of already existing A1 highway. In the central reserve of this highway, there was a pillar of the constructed bridge. The vibrations were measured in both embankment and the pillar under construction, as the works could have potentially influenced both structures. The column located the closest to the embankment was formed 15 m from its foot (Fig. 2a – the point marked with a stake). In this place, the embankment was 6 m high and the slope was equal to 1:1.5. Motor traffic was not stopped during column formation. The accelerations were measured on
the side of the motorway (Fig. 2b). In the meantime, subsequent drops of the pounder were performed from the height of 5, 10, 12 and 15 m. The column which is located the closest to the bridge pillar was formed at 48 m from its foundation (Fig. 3a – marked with a stake). Also in this place the embankment was 6 m high and the slope was 1:1.5. The accelerations were measured on the foundation of the pillar (Fig. 3b). The pounder was dropped from 5, 10, 12 and 15 m during the measurement. Supporting pillars had been constructed a few days before and the formworks were removed on the day when the measurements were
performed. Three monoaxial, piezometric sensors of accuracy 1,0V/g, measurement range 10m/s2and frequency >0,5Hz were used at each point to measure the accelerations. The results were recorded with a 16-channel diagnostic spectrum analyzer. Figure 4 presents the typical graphs of the registered acceleration changes.

As it can be seen in Fig. 4, the registered values of background amplitude were 50 mm/s2 (vertical) and 40 mm/s2 (horizontal). They resulted from the intensive car traffic during the measurements. As the authors had the results of the research in graphic form only, it is impossible to apply filters eliminating image noises of low frequency. The maximal values of vertical accelerations on the embankment were 300 mm/s2 and on the pillar – 125 mm/s2. The maximal values of vertical accelerations on the pillar were 180 mm/s2 and on the embankment – 130 mm/s2. The attenuation of amplitudes increase after pounder drop occurs approximately after 3 cycles.