Dear all,
I am repeating the simulation with results published in
Article Experimental Measurement of Parameters Governing Flow Rates ...
I follow all of their instructions in the COMSOL simulations, with parameters listed in the supplementary information.
I set up a geometry of 1cm (w) x 10 cm (L) strip as the article suggested. And provided an initial condition of -500e3 Pa all over the geometry.
The boundary conditions were set as the author suggested too. The three edges are set as no-flux condition. And a pressure of 0 Pa for the remaining edge which is initially saturated with water.
With a fine physic-controlled mesh and a step of (0,0.1, 10 min) for a time dependent study, the result, however, shows there is no Darcy flow through the geometry, only a very tiny velocity is observed at the edge saturated at the beginning of the study.
I am writing to see if i have setup something wrong and Thanks in advance for your kindly replies.
Lawrence.
Hi all,
Fortunately enough, I am able to find the answer by myself. The reason of why the simulation did not run smoothly seems to be because of the large pressure gradient between the initial dry condition and the saturated boundary condition.
Rath et. al. indeed changed the initial condition settings to circumvent the problem, by determining an water-saturation-averaged capillary pressure from the water saturation-capillary pressure graph, and using the water saturation at which the averaged capillary pressure is found to be the initial condition in the paper substrate.
Thanks very much for your help.
- Badges
- Science topic