Structural analysis of the Arundo Donax Pavillion at HISTeRIA 2012

Kanela pavillionIn the summer of 2012, four organizations gathered to make possible  yet another structure made out of cane (ESFA-InvestigacioCanyera, Sustenta, Canyaviva and the yet incipient Alitura). The project was developed within the research effort of all four groups in order to obtain a mechanical characterization of the Arundo Donax as a building material.

Due to the lack of time and resources, an initial project for load testing of the structure, as part of an extension of the research currently under development between Universitat Politècnica de Catalunya, ESFA, Sustenta and Canya Viva, was not possible.
Unfortunately the scheduled visit of structural engineer Oriol Palou (Sustenta) for such purpose was canceled for health reasons. Also other data besides the purely technical and constructive were not taken. These are parameters such as humidity within the columns, geometry of the final structure by means of photogrammetry, etc.
Kanela pavillion-trans
However, a Finite Element model was possible in order to analyze the structural behavior and compare the numerical versus the field displacements. This would allow us to approximate the Young’s modulus value for use in the design of further structures.
Irrespective of whatever program one uses, the main parameters that need to be defined in a FEM model are as follows:
  • Density = Specific mass = mass per unit volume: 40 Kg/m3
  • Young’s modulus: 250.000 N/mm2
Element sections
  • Section area: 3.000 mm2
  • Moment of inertia:  2.500.000 mm4

There are many others, mostly depending on the software one chooses, but what matters is that, in the used abstraction for this particular and simple case (called Timoshenko Beam), the only really important information used is that above.

Density (specific mass) and Young’s modulus were estimated iteratively, modifying their values in the program until the calculated displacements were somewhere near reasonable. A further more accurate measure of the specific mass of a bundle of cane and of the displacements in the structure will be in order and soon published.

Regarding the element sections, a CAD software was used to calculate the values from a general section of a bundle. From the past experiments made (see an article on the topic in ACE journal: it is still unclear whether the inertia of the bundle or its Young’s modulus is reduced about the middle of the arch.


Future research efforts should be oriented towards simplifying these assumptions but most importantly towards a methodology for obtaining these values in the field with some acceptable accuracy and speed.