[IADInocencio]

Hi all, I just wanted to ask if there would be a way to model a perforated pipe running through gravel, essentially a French drain, in Reef3D? Where water would flow on the surface of pavement on the right of the image, to a gravel cover on the left with the pipe?

[kamath]

Hi Ismael,
Given my autocad and civil engineering design days are far behind me, I didn't follow the hatchings in the sketch. Could you dumb it down a little bit more and draw an arrow for the possible pathway of water so that I can comment on your problem?
Thanks

[IADInocencio]

Hi Arun,

Thanks, I have uploaded a more detailed schematic for your reference. In essence adjacent to a road there will be a gravel overlay with perforated pipes underneath, which in turn drain to a larger canal. I know Reef3D can do porous media, but I'm not sure about it's ability to transmit flow from the media into an open pipe as shown.

[kamath]

Thanks for that clarification, Ismael!

The pipe is buried inside the porous medium and the interior of the pipe will therefore be "non-porous"/ regular domain. Such an "inserted" domain might not work.
A work around would be to define the porous medium such that it does not really overlay/ overlap with the regular medium.
See if that works for you.

(Though currently there seems to be a bug pointed out by
After using the B270 function, the force has a null value.
whereby concurrent occurrence of S and B 270 leads to a crash on the latest release. We are working on identifying and fixing that.)

[IADInocencio]

Hi Arun, thanks for the quick reply. Would it be possible to "cheat" in a way and model a highly porous cylindrical material within the gravel as an alternative to the pipe?

[kamath]

Yes, that would be a possible cheat. But I advice caution both with respect to the physics and the numerical stability.

We have used "nested" porous layers in the simulation of breakwaters (they are generally composed of three layers with different porosities) and that works fine. You could have a look and then test the extent to which you could stretch this
Eg.
10.1080/21664250.2020.1734412

or

DOI: 10.2112/jcoastres-d-19-00147.1

[IADInocencio]

Thanks Arun!
I will look into this and see how workable it would be for this type of analysis.