Load Tests on a 1:1 Model of a Geogrid-Reinforced Bridge Abutment
The paper deals with geogrid reinforced soil as solution for bridge abutments. Preliminary results are presented of a real scale test of a simulated 4,5m high geogrid reinforced vertical soil block loaded directly on top near the edge by the bridge sill beam, high-lightening the low settlements and horizontal displacements measured. In one test, the reinforced embankment was nearly lead to rupture, what occurred with a load in the order of 3 times the usual one for this kind of structure.
The tests presented here on a geogrid-reinforced soil block simulating a real bridge abutment under a sill beam are in no way intended to be a comprehensive detailed scientific analysis. The exercise is much more about testing the behaviour of a system and its performance reserves in a situation that can be related to practice, from the point of view of "we want to know". The use of an already constructed test object after modification was advantageous in terms of time and money, however, it also brought its own restrictions and deficiencies (Section 3.2), including that we would have to live with the known, somewhat insufficient, compaction in the upper layers and the possible looser fill zones in the front face zone resulting from the tests done for other purposes. The tests described herein are still fairly recent; and so the following remarks are a first, rather incomplete overview, but the most important points are readily recognisable and can be translated into practice.
• The tested arrangement should be seen as a "worst case" scenario:
• The sill beam was only 1.0 m wide and placed only 1.0 m away from the edge
• The front face was vertical
• The outer skin (facing) had no bending stiffness, being only a geogrid-wrapped-back wall, without any form of stiffening
• The degree of compaction of the fill in the most sensitive upper zone was only Dpr = 95%, with probably loosened zones in the front face area near the loading beam, some probably as a result of the previous tests.
The following remarks can be made:
• A contact pressure under the sill beam of up to 650 kN/m2 (approx. 3 times the pressure normally experienced) led to no obvious component or system failure. However, because there were signs of serious effects taking place, the situation could be used as a marker for the ultimate limit state.
• A contact pressure of up to 400 kN/m2 (approximately twice the usual value) resulted only in completely acceptable deformations.
• The tested system exhibited technically advantageous, ductile behaviour with no discontinuities and seems to have a substantial reserve capacity.
• The overall performance can be considered very good despite the previously found soil density deficiencies.
• The facing consisting of flexible geogrids had no bending stiffness but showed only small local and global deformations (marginal in the relevant load range).
• The settlement behaviour of the loading beam (indirectly assessed by converting the modulus of subgrade reaction) was as if it had been sitting on an infinite horizontal plane and not near a vertical slope; the only plausible explanation is the apparently highly effectiveness of the incorporated reinforcement and the geogrids used.
The author would have no reservation using the structure as built and tested (and ideally with better soil compaction) in practice.