Influence of the Test Method in the Assessment of Concrete Sensitivity to Explosive Spalling

In the design phase of structures for which explosive spalling is a critical issue, preliminary screening of concrete mixes plays a key role in pursuing the optimal compromise between costs and efficacy. Generally, the best way to limit or even prevent concrete spalling in structures exposed to fire is to properly define the mix design, especially in terms of fibre type and content. Within this context, in the experimental assessment of spalling sensitivity, it should be kept in mind that there are many influencing aspects, such as specimen geometry, heating curve and possible loading level. All these factors should be accurately defined in order to obtain results representative of the actual service conditions. In the present study, three different test setups have been adopted for the evaluation of spalling sensitivity of two concrete mixes (fcm = 45 and 50 MPa) containing different amounts of polypropylene fibre. The testing procedures include: (I) 1m side cubes directly exposed to a localized jet flame (hot spot test), (II) 0.8x0.8x0.1 m slabs heated at the intrados under a constant membrane compression (biaxial loading test), and (III) a loaded tunnel lining segment exposed to RWS fire curve (full-scale test). These methods entail an increasing experimental burden and different levels of detail in the monitored parameters. In the paper, the different typologies of tests are discussed by comparing experimental and numerical results in order to highlight the main influencing factors.