I am trying to simulate the soil media hydrodynamics with underdrain on the bottom layer.
Here is the real case. three different soil media and underdrain on the bottom layer which was designed to drain the water from the infiltration. My objective is to size the underdrain pipe.
Here is what I did on the model. The inflow is a constant which was given to the top surface. I set the underdrain with a initial assumed diameter through the curved pipe tool and considered the underdrain pipe surfaces as seepage faces in the boundary conditions. In addition, I set the initial solute concentration on the initial conditions.
After simulation, I don't know what data can indicate the underdrain size. Another question is that the concentration of solute was very high which is not reasonable. Could someone help me with the questions please? I really appreciate your help.
how to check whether the underdrain size works or not for a constant inflow?
how to check whether the underdrain size works or not for a constant inflow?
 Attachments

 yaoanWTS0514400mm.zip
 (1.23 MiB) Downloaded 3 times
Re: how to check whether the underdrain size works or not for a constant inflow?
Although you specified (circular) drain pipes with a diameter of 0.2 m, your FE discretization is so course, that you end up with squared drain pipes (four nodes on the circle is in fact a square). If you indeed want to have a circular pipes, you will have to use Mesh Refinement on Pipe Surfaces. You will certainly need finer spatial discretization. read rules about spatial and temporal discretization (FAQ :
http://www.pcprogress.com/Documents/No ... zation.pdf
Other problems:
1. You specified water content dependence of (solute transport) parameters, without specifying any parameters for that.
2. You selected that Initial Concentration is given in terms of Total Concentrations, while you have no reactions (sorption, volatilization, etc). Why?
3. You run four solutes, while all are considered to be tracers. Why? The results will be the same for all solutes, except for the scale.
4. You will need larger longitudinal dispersion, giving the size of the domain and FE discretization.
5. Also note that your inflow (0.008 m/min) is significantly larger than the saturated hydraulic conductivity of the surface layer. The program will have to generate large positive pressures at the surface to get water in.
J.
http://www.pcprogress.com/Documents/No ... zation.pdf
Other problems:
1. You specified water content dependence of (solute transport) parameters, without specifying any parameters for that.
2. You selected that Initial Concentration is given in terms of Total Concentrations, while you have no reactions (sorption, volatilization, etc). Why?
3. You run four solutes, while all are considered to be tracers. Why? The results will be the same for all solutes, except for the scale.
4. You will need larger longitudinal dispersion, giving the size of the domain and FE discretization.
5. Also note that your inflow (0.008 m/min) is significantly larger than the saturated hydraulic conductivity of the surface layer. The program will have to generate large positive pressures at the surface to get water in.
J.
Re: how to check whether the underdrain size works or not for a constant inflow?
Jirka wrote: ↑Tue May 14, 2019 2:43 pmAlthough you specified (circular) drain pipes with a diameter of 0.2 m, your FE discretization is so course, that you end up with squared drain pipes (four nodes on the circle is in fact a square). If you indeed want to have a circular pipes, you will have to use Mesh Refinement on Pipe Surfaces. You will certainly need finer spatial discretization. read rules about spatial and temporal discretization (FAQ :
http://www.pcprogress.com/Documents/No ... zation.pdf
Other problems:
1. You specified water content dependence of (solute transport) parameters, without specifying any parameters for that.
2. You selected that Initial Concentration is given in terms of Total Concentrations, while you have no reactions (sorption, volatilization, etc). Why?
3. You run four solutes, while all are considered to be tracers. Why? The results will be the same for all solutes, except for the scale.
4. You will need larger longitudinal dispersion, giving the size of the domain and FE discretization.
5. Also note that your inflow (0.008 m/min) is significantly larger than the saturated hydraulic conductivity of the surface layer. The program will have to generate large positive pressures at the surface to get water in.
J.
Thank you very much for the reply.
problem1: yes. I didn't set the parameters which is not proper
problem2: Actually, I more care about the reaction in the liquid which go through the soil media. Probably I need to change the Total Concentration to Liquid Concentration.
problem3: I run four solutes because they are different solutes which is our objective polutant. We want to know what is the outflow concentration for those four solutes. The scale is a lot more than the initial condition I set which is not right.
problem4: Alright. Thanks for the suggestion.
problem5: Yes, that inflow is our design flow. Even it is a lot larger than the saturated conductivity of surface layer, it started from unsaturated to saturated. That is why we set this model to simulate how many and how big the underdrain pipe should be. Is it right to set the underdrain pipe surface as seepage flux?
Re: how to check whether the underdrain size works or not for a constant inflow?
Yes, that is correct. Subsurface drains should have the seepage face BC. J.