Field measurements at a composite bridge with composite dowels as shear connectors - Economical Bridge Solutions based on Innovative Composite Dowels and Integrated Abutments - page 66

As already mentioned, forces acting on the rail support point are not allowed to exceed
limit values predefined by DB Netz AG. Within the scope of the bridge monitoring these
vertical loads were measured separately for compression and tension forces, based on the
corresponding load bearing mechanism.
In the case of an external tensile loading F acting on the rail support point, the tension
force is transferred from the ribbed base plate R through eccentric insulating bushings with
collar E into two pre-stressed stainless steel threaded bolts (measurement bolt) B , anchored
in the superstructure. The determination of these tensile forces was realized by a direct
measurement of normal strain in the measurement bolts for which special strain gauges are
glued into the bolts (“Fig. 7”, D and “Fig. 8”).
The threaded bolts are not pre-stressed against the ribbed base plate R but against the
eccentric insulating bushings E . Hence a compression loading of the rail support point did
not lead to a measurable decrease of preload force in the measurement bolts, as the pres-
sure hull (caused by pre-stressing of the bolt) built up in the eccentric insulating bushings E
were not decreased by a lowering of the ribbed base plate R (“Fig. 7”, “Fig. 11”). Therefore
compression forces were determined by a measurement of the lowering of the ribbed base
plate R combined with the known stiffness of the elastic intermediate layer. This approach is
equivalent to the determination of vertical spring stiffness for design purposes as described in
chapter 3.1. The actual deformation of the ribbed base plate R was gauged by four inductive
displacement transducers relatively to the slabbed track (“Fig. 7”, I and “Fig. 8”).

Figure 8: Measurement devices for monitoring of forces acting on rail support points

(1 st , 2 nd and 4 th rail support point equipped)

In the first instance all measurement systems were calibrated at the Institute of Steel
Construction of RWTH Aachen University in a phased process. Initially a linear calibration
factor ktension,1 was determined. This factor was used for the conversion of measured elastic
strains into equivalent bolt forces. Based on the concept of measurement bolt calibrations
already successfully applied in the former research project RFCS HISTWIN [13] [14], the
measurement bolts were loaded displacement-controlled in a two-column testing machine up
to a maximum load of F =50 kN. The strains measured during testing were plotted against the
corresponding jack forces (“Fig. 9”) to derive a linear calibration factor ktension,1 for each