open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,

Can anyone advise on how to set a fixed value of a passive scalar as a function of time at a specific point inside the computational domain? E.g. time t1, TR= 100, time t2, TR=200 etc.

I have used the ABSCISSAE OF SOURCES, ORDINATES OF SOURCES and SOURCES FILE, e.g.

T Q(1) TR(1,1) Q(2) TR(2,1) Q(3) TR(3,1)
s M3/S NG/L M3/S NG/L M3/S NG/L
0 1 100000 1 100000 1 100000
20 1 100000 1 100000 1 100000
21 0 0 0 0 0 0
85800 0 0 0 0 0 0
86400 1 100000 1 100000 1 100000
87000 1 100000 1 100000 1 100000
87060 0 0 0 0 0 0

This works fine but the issue is that I do not have a source (m3/s), just a fixed value of TR appearing at a particular time at a particular location and lasting a specific time. I am finding that the source (m3/s) is affecting the value of the TR introduced and this value is much less than than the fixed value required (100000).

Thanks,

Tom

Hello Tom,

Can you send your steering file + potential user Fortran files please?

What about the size of elements close to the sources?

Do you also use the keywords VELOCITIES OF THE SOURCES ALONG X/Y/Z in addition to the keywords ABSCISSAE OF SOURCES, ORDINATES OF SOURCES and SOURCES FILE. If yes, what about the surface of the elements close to the sources compared to the ratio flowrate of sources and magnitude of velocities of sources?

Chi-Tuan

Hi Chi-Tuan,

Attached are the steering file, sources_tracers file and an image of the mesh. There are 12 vertical levels and I wish to fix the tracer the values of the tracer inside the circle shown in the image for the 3 layers closest to the surface over a certain time period.

I have not used the keywords VELOCITIES OF THE SOURCES ALONG X/Y/Z as I assumed they would be zero as default.

Thanks,

Tom

Hello Tom,

What about starting by simplifying your model with only 1 tracer and 1 source, and progress step by step?

I have a few questions and suggestions to start:
- what about the bottom elevation compare to value 0 m? You set every elevation of sources at 0 m. The sources have to be inside the 3D mesh (in particular not at the bottom or above the free surface),
- as written in the TELEMAC-3D user manual, it is recommended
to use fixed levels at sources elevations in order to avoid unwanted vertical movement of the sources during the simulation. If every source points are located at 0 m, it is easy: use MESH TRANSFORMATION = 3 + define your mesh distribution in the Fortran file directory with a modified USER_MESH_TRANSF subroutine, see e.g. the one from Cooper example,
- after answering/changing the first 3 items (including decreasing the number of source points) and if not solved, do you have the same behaviour with TYPE OF SOURCES = 1?
- can you try to refine the 2D mesh close to the source point? Too coarse meshes are not well suited to source point feature. Try 2 meshes with element size divided by 2 and 4 around the single source point used for these tests.

Indeed if you have not used the keywords VELOCITIES OF THE SOURCES ALONG X/Y/Z, velocities will be set to 0 m/s by default. Otherwise, there may be a constraint between the size of elements and the ratio flowrate and velocities of the sources.

Hope this helps,

Chi-Tuan

Thanks for the advice, Chi-Tuan. I will certainly try out your suggestions.

Tom