The remaining capacity of the timber foundation of a historical quay wall with relieving platform

Abstract

The increased loading and decreasing strength of historical quay walls calls for remedyand has drawn the attention of many Dutch municipalities, Rotterdam being oneof those. In order to set up proper assessment methods that result in an economicaland ecological policy for renovation or replacement of these structures the Port of Rotterdamhas asked Delft University of Technology to assist them. This thesis focusses onthe elements perhaps most prone to decay and hardest to inspect: the timber foundationsimplemented in quay walls with relieving platforms from the late 19th and early20th century. The main research question of this thesis is What is the (remaining) capacityof the timber foundation of a quay wall with relieving platform? and focusses onthe Maaskade case. The aim is the provision of a structural analysis and appointmentof weakspots on a representative quay section build-up using measured material characteristics.This quay has partially collapsed and has been replaced as of May 2020. 45Timber piles from the site have been investigated in the Rotterdam harbour to derivedensity and (dynamic) Young’s modulus measured by a Timber Grader MTG. A selectionof these piles was then further tested in pieces (21) at the TU Delft structural laboratory(Stevin II) to derive the Young’s modulus and compressive strength. Using the correlationfound between the compressive strength parallel to the fibre ( fc,0) and the dynamicYoung’s modulus parallel to the fibre (E0) the compressive strength values of the entire45 pile group have been derived. This results in fc,0,k = 12.64 MPa and E0,mean = 11.3GPa under 50% moisture content, the saturated state the wood is expected to be in inthe quay. The resulting design strength fc,0,d = 10.7MPa is very much in line with the advisedstrength of 10.8 MPa in NEN 8707 . Missing characteristics necessary for structuralassessment are used from bending class C22 (sawn timber) from EN 338 . A FEMmodel of section 4-1 of theMaaskade under different historical build ups has been set upin Plaxis 2D, and an Ultimate Limit State computation has been carried out on the finalbuild-up before replacement under four load cases. The found normative location thatcan be seen as the weakspot is the joint between the raking pile and the adjacent cappingbeam. A structural assessment based on sectional forces results in a negative outcome ofthe structure even on the self weight governed loadcase as unity checks > 3 are given for thecapping beam. These results do have the shortcomings due to simplifications from 3D to2D and a modelling with linear elements with no volume in the model and no allowancefor further load distribution beyond elasticity. A definite answer on the capacity in termsof additional external loads cannot be given and the need for a model that incorporatesthe exact lay out and positioning of the loads is stressed. Further recommendations forpractice are to find out if the capping beam- pile connections are expected to be able totake tension and to name their state after an inspection.