Résumé
The classic experience of filling a container with a granular medium poured froma source point reveals the presence of the intergranular forces network by themanifestation of the Janssen effect (Janssen, 1895). Indeed, the measurement ofthe stress applied at the base of the container according to the bed particleheight, shows that the static mechanical state of the ensiled granular medium isnot comparable to a hydrostatic one which is characteristic of a fluid. The lateraldeflection of gravity forces to the walls via the network of intergranular contacts,allows to explain the lowering of vertical stress (σzz) in relation to hydrostaticpressure. Moreover, the sliding threshold condition must be established at thewalls to explain the "saturant" shape of the σzz profile (Ovarlez and Clément,2005). If mechanical properties are involved to explain this apparentphenomenon understanding as a static equilibrium, the analyze of the free surfaceduring pouring correlated to the local mechanical state allow to indicate thatkinetic energy had an influence on the Janssen effect (Mandato et al. 2012; Duri etal. 2018). In order to revisit the Janssen effect, experiments and Discrete ElementMethod (DEM) simulations are compared to investigate the structure of an ensiledgranular medium poured in a cell by a single point at different initial drop heightsand flow rate (parameters which allow to control kinetic energy).Experimental device. The granular medium consists of a population of glass beadsof 1 to 2 mm diameter with a small span value (50/50 in volume). The experimentalset-up, developed in Mandato et al. (2012), consists of an open glass cell container(51 x 100 x 160 mm) made of transparent glass walls. A flat and rectangular steelprobe is especially designed to fit into the glass container. A small probe allowslocal measurements of the vertical stress in the powder bed. A large probe is usedto simulate the bottom of the cell and allows a global measurement of the verticalstress, as in Janssenʼs experiment (Janssen, 1895). Whatever its width, the probeis screwed on a rod that is linked to a load cell of a texture analyser (TA.XT2, TableMicro System), which is used as a force sensor. The filling is provided by a funneland the flow rate is modulated by the output diameter. After pouring, the slopeangle and the arrow of the free surface are measured by image analysis.Numerical simulation. We develop our own homemade code in c++using the DEM.Then we decide to develop also a code using the coarse-graining method(Weinhart and Luding, 2016) to highlight on the different fields. This last methodwhich doesnʼt give access to the compactness near the wall, is replaced by a newmethod using two grids in order to increase the accurate. With these three codes we can now explore all the contacts between the particles by the calculus of (i)normal and tangential forces, (ii) moment, (iii) mobility, of each particle.Results. The variation of the initial drop height and the flow rate impacts the slopeof the free surface: crater form in high kinetic energy conditions, heap form inlowest conditions and intermediary slopes between these conditions (from plate to“camel-like” form). The measured and calculated values of the arrow and of theangle of repose are in very good agreement in each case. Experiments andnumerical simulation highlight the same layout of the vertical stress field (Fig. 1). Inthe upper part of the particle bed, there is a layer in which the stress is generallyassimilated to a “quasi-hydrostatic” state. It can be seen that under this first layer,Janssenʼs hypothesis that stresses are lateral uniformity is not valid: there is astrong local heterogeneity of the stress within the granular medium. It can be seenthat the redirection coefficient defined by Janssen (Janssen, 1985) ratio betweenlateral stress and vertical stress, is not constant. The study of thickness and meanstress in each different zone of the bed versus kinetic energy during the fillingstage had been realized. The static mechanical state of the particle bed is also dueto this dynamical parameter and this point is usually not taking into account.