Erfttal Railway Bridge

Client: European Union, Research Fund for Coal and Steel, represented by Rheinisch-Westfälische Technische Hochschule, RWTH, Aachen
Reports: 2009 DETAILS FInal Workshop, Lucca, Italia
Publications: 2008 IMAC 26, Orlando, Florida
2008 Eurodyn, Southampton,UK
2008 IABMAS Seoul, Korea
2009 VDI Baudynamik, Kassel, Germany (in German)

This project was realized in the context of the EU Research Project DETAILS (Design for optimal performance of high speed railway bridges by enhanced monitoring systems), July 1st, 2006, to June 31st, 2009, The project was supported by the Research Fund for Coal and Steel (RFCS). rci dynamics was subcontracted by RWTH Aachen, DETAILS Partner, to organize and accomplish tests on the Erfttal Bridge. This two-track railway bridge is located close the City of Kerpen between Köln and Aachen, Germany.

In April 2007, an experimental modal analysis using Ambient Vibration Testing technology was performed on the Erfttal Bridge. rci dynamics acted as a a project leader and also provided the major part of the equipment used. Project partners were RWTH Aachen, Bauhaus University Weimar, KU Leuven, Belgium, and GeoSIG, Switzerland. All partners co-operated closely providing technical knowledge, test crew members and equipment. The local logistics were organized by RWTH Aachen.

The bridge consists of two 24.6 m long simply supported filler beam girders. The two girders are about 5 m wide each. They are separated from each other through a 200 mm open gap. Their position in the longitudinal direction differs by 4 m. Each of the girders carries a ballasted railway track being used by normal and high-speed (ICE) passenger trains as well as by heavy freight trains (Figs. 1 to 4).

Fig. 1: Bridge Foto with ICE-High-Speed-Train crossing.

Fig. 2: Plan view of the bridge. Dimensions are [m]. The situation underneath the bridge can also be seen: Two 4.3-m-lanes of the Erfttalstrasse, an access road to Motorway A4, a 2 m wide green strip and a 2.5 m pedestrian/cyclist lane to the right.

 Fig. 3: Bridge cross section (above).

Fig. 4: The bridge as seen from below.

The ballast is continuous in the longitudinal as well as in the transverse directions thus covering all gaps between the girders and the bridge and adjacent tracks. The bearings are of the open elastomeric type and are therefore elastic in all directions. It was hence one of the main purposes of the experimental modal analysis to providing reliable data concerning the stiffness of the gaps and the supporting devices. The measurement point layout consisting of 44 measurement points can be seen in the drawings in Figures 2 and 3. With using four references (indicated with black squares), 10 rovers (circles) and 14 channels of the frontend, the measuremenent campaign was divided into four setups. The technical and logistic problems to be solved were quite challenging. Details of how these were solved and how it was possible to acquire time windows of a minimum 410 seconds with purely ambient conditions for all four setups can be found in the publications cited at the beginning.

The colleagues at Bauhaus University in Weimar updated their SLang baseline Finite Element model using the in-house developed OptiSlang software package. As was to be expected, updating of the FE model was a comprehensive task. However, the first seven bridge modes, f = 3.7…21 Hz could be reproduced with MAC > 0.8 (Fig. 5) (2009 VDI Baudynamik-Tagung (in German).

Fig. 5: Comparison of experimentally (AVT) and analytically (updated FEM) determined bridge modes (MAC = Modal Assurance Criterion).