THE KIRKKONUMMI TEST EMBANKMENT ON DEEP AND MASS -STABILISED SUBSOIL

The Kirkkonummi test embankment was part of The Pavement Structures Research Programme (TPPT), which is part of the Finnish National Road Administration's (Finnra) strategic project "Road Structures Research Programme". The TPPT research programme is being carried out in 1994-2001 with funding provided by Finnra. The Kirkkonummi test embankment was the most remarkable TPPT testing project concerning deep and mass-stabilisation. The test area is located at Kirkkonummi (Main road 51 Helsinki-Karjaa, road section 8), in the Southern Finland. The test embankment is situated in the area of a future multi-level junction of main road 51. The embankment is founded on soft subsoil. Beneath the top layers, which consist of peat and gyttja, there is 7-8 meters of soft clay above sandy soils and moraine. The test embankment consists of a lower (2 meters) and a higher (6 meters) bank. The clay layer has been deep stabilised and the layer of peat and gyttja was mass stabilised under the lower part of the test embankment and replaced by a layer of reinforced crushed rock under the higher part of the test embankment. The lower embankment consists of a 1.7 m thick mass-stabilised peat slab standing on semi-hard columns (c/c 1.5 m, 600 mm). The higher embankment is standing on hard stabilised columns (c/c 1.7 m, 800 mm). Between the columns and sandy embankment is a layer of crushed aggregate reinforced with steel meshes. The functional basic idea and research hypotheses of the structures were to find: 1) Deep stabilisation as ground reinforcement for high embankments instead of piling, and 2) An embankment construction based on deep stabilisation where the load distributing layer is either mass stabilised peat or reinforced layer of crushed rock. The deep stabilisation and mass stabilisation test fields were made in June 1996. The test embankment was under construction, from September to January 1996-1997. Before the construction phase, extensive laboratory tests were carried out with site materials in order to find out the feasibility of the stabilisation. Near the test embankment, test fields for both deep and mass stabilisation were made. In these on one hand the stabilisation capability of different binding agents and on the other hand different stabilisation entrepreneurs' ability to produce homogenous columns was studied. In both rigid and semi-rigid deep stabilised columns the binding agent was raffinate of blast-furnace slag + portland cement 1:1 120 kg/m3 and 200 kg/m3, respectively. In mass stabilised peat and gyttja the binding agent was quick-cement 150 kg/m3. The dimensions of the steel mesh used were 4 mm, # 100 mm x 100 mm. Various instruments were installed in the test embankment: earth pressure cells, vertical and horizontal tubes for radiometric measurements, settlement gauges, pore pressure cells installed in clay, settlement marks, inclinometers and different types of strain gauge applications. The disturbance of the clay during and after deep stabilisation is studied by pore pressure cells and field vane tests. Monitoring continued until summer 1999. The behaviour of the higher embankment was also analysed with 3D-FLAC and PLAXIS computer programs. With 3D-FLAC the embankment was analysed as a whole and with PLAXIS only a single column was modelled. The results underestimated the degree of settlement and the settlement rate and overestimated the loads accumulating to columns even using conservatively selected parameters based on laboratory tests.