Rail Structure
Interaction
Case Study Modelling and Benefits
Jeremy Barnes Associate Director Hewson Consulting Engineers
Nathan Griffiths Design Engineer Hewson Consulting Engineers
Introduction
Development of Rail Fixings
The Need for Rail Structure Interaction
RSI Modelling Issues
Case Study 1 Jakarta MRT
Case Study 2 North Wales Viaduct
Development of Rail Fixings
Historically jointed rail used
Maximum 37m rail lengths
Fishplate joints with gaps left between rails
Rail movement accommodated at joints
No thermal rail stresses built up
Poor ride quality
Problematic for track circuits
Maintenance liability
Development of Rail Fixings
Continuously Welded Rail Developed
Joints in rails eliminated
Long lengths of rail can be used
Good ride quality
More compatible with track circuits
Low maintenance but more expensive to install
On UK Network most jointed track replaced by
CWR
Thermal stresses built up in the rail
Development of Rail Fixings
The Need for RSI
CWR designed to resist thermal stresses
Structure movement increases rail stresses
Rail re-distributes traction & braking
The Need for RSI
Lengthy structures traditionally isolated with rail breather joints at ends
Maintenance liability & rail authorities keen to eliminate
The Need for RSI
Keep spans short and simply supported
May not be the most economic structural solution
Creates a different maintenance liability
The Need for RSI
Undertake rail structure interaction analysis
Models effect of structure on rail and vice versa
Guidance given in UIC774/3 & EN 1991-2
RSI Modelling Issues
Ballasted Track Bi-linear Springs
70
Resistance of the track, k [kN/m]s
60
50
40
30 Unloaded
Loaded
20
10
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Displacement [mm]
Rail to Structure Connection is non-linear
Represents the build up of resistance and slip of connection
Different frictional values for loaded and unloaded track
RSI Modelling Issues
Bi-linear behaviour of the track
70
Resistance of the track, k [kN/m]s
60
Unloaded [Ballast]
50
Loaded [Ballast]
40
30 Unloaded [Direct Fixing]
20
Loaded [Direct Fixing]
10
0
0 1 2 3
Displacement [mm]
Different stiffness values for ballasted and direct fixing
Different frictional values for loaded and unloaded track
RSI Modelling Issues
Long lengths to be modelled to minimize end effects
Abutments, station structures and embankments are stiff elements
Traction and braking forces are attracted to these elements
RSI Modelling Issues
Limitations on rail curvature in UIC774 &
BS EN 1991-2
Rail curvature limited to 1500m radius
Derogation required in UK to use on curved
track with a radius less than 1500m
Additional lateral restraint required or
reduced allowable stresses
Case Study 1 Jakarta MRT
Mass Rapid Transit (MRT) Line in Jakarta, Indonesia
4.7 km of viaduct decks typically double tracks
4 elevated stations
Case Study 1 Jakarta MRT
Precast segmental internally prestressed
Simply supported spans, typically 36.65m
Case Study 1 Jakarta MRT
Direct Fixing track
Fully CWR no breather joints
UIC 54 rail
Case Study 1 Jakarta MRT
3D Model of rail and viaducts developed
Stiffness of the rail, superstructure, substructure and stations are modelled
Case Study 1 Jakarta MRT
Rail stresses
Action Value Permissible stress (UIC)
Temperature +/- 41 N/mm2
Rail traffic +/- 26 N/mm2 72 N/mm2 (Compression)
TOTAL +/- 67 N/mm2 92 N/mm2 (Tension)
Case Study 1 Jakarta MRT
Longitudinal bearing forces from railway secondary loads
180
160
RSI decrease in bearing 140
forces 120
Force [kN]
100
Without RSI
More economical bearing
80 With RSI
design
60
40
20
0
Typical Bearing
Case Study 2 North Wales Viaduct
Single track bi-directional line in North Wales
Significant Rail curvature with track lateral restraints
Low speed line
Case Study 2 North Wales Viaduct
Multi-span concrete beam and slab bridge over river
Fully integral piers to eliminate bearings
Thermal movement of piers affects rail stresses
Case Study 2 North Wales Viaduct
Fully integral abutments
Sleeved piles at abutments to allow movement
Pway supported on embankment behind
Case Study 2 North Wales Viaduct
3D MIDAS model of rail and structure developed
Rail and structure connected with MIDAS multi-linear springs to model track
bi-linear stiffness relationship
Multi-linear springs at 1m centres
Case Study 2 North Wales Viaduct
Stiffness of abutment modelled
Fill resistance behind abutment modelled
Embankment modelled as stiff restraints
Case Study 2 North Wales Viaduct
Lower and upperbound models developed to reflect variability in ground
conditions
Lowerbound model included design effects of scour
Case Study 2 North Wales Viaduct
Rail stresses
Case Study 2 North Wales Viaduct
RSI distributes horizontal loads to all piers and to the embankments
Conclusions and Benefits
RSI analysis opens up the use of multi-span continuous
construction
RSI analysis can allow integral construction to be used
RSI offers economies as longitudinal forces are distributed over
longer lengths reducing the forces on piers and bearings
Thank you