Tacloban Charity Project: London Update

For the past nine months Philippe and I have been offering technical engineering support for the design of the dormitory structure in Tagpuro in the aftermath of super-Typhoon Yolanda. For more information on the project visit our previous blog post here and the Workshop blog here.

Prior to Typhoon Yolanda, the local design code stated a wind speed of 250kph should be modelled. Yolanda created wind speeds of 320kph over prolonged periods of time, meaning the codes became outdated. When you have such high wind speeds it’s really important to understand how it is likely to behave around a structure. The protruding rafters of the dorm block create chaotic wind behaviour which is best modelled using a wind tunnel or Computational Flow Dynamics (CFD).  We worked in conjunction with Chris Ochyra, an expert in Advanced Engineering from Ramboll’s Southampton office, to model the wind around our four buildings. It’s important to account for the topography and land features in the local geographical area which means the CFD boundary conditions are critical. The ANSYS CFD model outputted values for the pressure coefficient (Cp) which we converted to member and area loads and applied to the global structural SAP model. We modelled the structure with and without louvre panels which allowed us to understand whether it would be best to open or close these removable panels during high-wind scenarios.

We found that the lead engineering practice (based in Manila) underestimated the loadings upon several key areas and we hence undertook a thorough review of structural members. This highlights the importance of using CFD for analysis because the values they had taken from literature were far too low and members could have failed during high winds.

The Philippines and Tagpuro is also susceptible to earthquake and is therefore important to ensure the building behaves well under seismic loadings. A seismic review was carried out in coordination with Davide Pedicone. Davide has worked extensively on the seismic rebuilding after the 2009 L’Aquila earthquake in Northern Italy and in California. His experience has been invaluable for this project.

In the dormitory building, the three large concrete cores provide the building’s stiffness and strength. In between the cores the building is based on a series of timber frames, rafters and purlins. The loads generated from their seismic mass will be transferred through to the cores and down to the foundations.

The results from the seismic model highlighted some of our concerns about lateral drifts (in excess of 150mm!) and the adequacy of the design of the bracing. The bracing is provided by large timber cross beams in the transverse direction and by the purlins in the roof. In light of our seismic study we have recommended additional diagonal steel bracing rods connecting the roof to the core which will limit drift to an acceptable level (8mm). In seismic design, members are designed beyond their elastic limit during large earthquakes. The result is some plastic deformation and damage such as cracking. The timber cross bracing was designed fully elastically to avoid a loss of strength or stiffness and so maintain the integrity of the building in an earthquake. In earthquakes it is also important to design for redundancy, or in other words the loss of key members.