open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,

I want to model an open channel case of constant slope bathymetry (2m - 2.4m) , on the intake side of channel the water is pumped in at constant rate of 1.9 m/s, and on the other side the channel empties into a pit where the water falls freely.

For the first boundary I imposed a discharge boundary condition by giving value 4 5 5 in boundary condition file, however I am not sure as what boundary condition I have to impose on the other end where the water is falling freely. I have run the case by giving free H, free V and free U due to which the boundary condition is 4 4 4, and simulation runs properly by giving the exact outfall discharge as the intake.

I want to know whether the method of boundary condition for free fall outfall condition is the correct method or not. If not then what possibly I may have to make change in the boundary condition.



Thanks,
Gourish

Hello,

Yes this is correct, if the exit has a Froude number greater than 1, in which case all the conditions at the exit are given by what happens inside your channel. I would advise in this case to use the method of characteristics for the advection of velocities, so type of advection : 1;5

With best regards,

Jean-Michel Hervouet

Hi, dear everyone who use telemac2d.
I'm on simple mobile bed experiment by using telemac2d.
my experiment is on 4m width, 40m long.
and discharge is 9.1m^3/s for inlet.
precribed elevation is 0.95m for outlet.
but i'm not sure i can use 9.1 value for prescribed flowrate.
i think it is just for m^2/s.
so I wonder i should change the value of discharge as m^2/s

and the second question is also about boundary condition.
when i did like above, error message is appeared.
" inlet boundary should have velocity profile or check water depth."
i tried variable flowrate with time, velocity, and elevation.
but failed.

Could you give me some advice?
Thank you for reading this question.
Hope you have a nice result for your work too.

um, i also attached my
stering file : cas4
geo : ex3
boundary : conlim

I hope some kind person help me.
it doesn't work...ㅠㅠ

Hello,

The discharge at the entrance is meant for the whole boundary, so in m3/s.

For the second problem, the principle is that the program computes velocities that will give the discharge, considering that the depth is taken at the previous time step, so in your case in the initial conditions. If there is no water in the domain in the initial conditions, it cannot work. So:

* either you add initial conditions, like constant depth or constant elevation, see user manual (you can also program them).

* or you really start from dry conditions but then the flow will be supercritical at the entrance, and then the depth must be also prescribed and you must give its value.

With best regards,

Jean-Michel Hervouet

very thank you, Jean-Michel Hervouet.
You are very kind, so I'm impressed.

I have another question.
I can compute this simulation by using time step 0.01.
but I'm not sure it computes well.
becuase, when I give time step 0.01, fudaa can't accept this value. it can accept just until 0.1 for time step.

Does telemac2d just allow time step until 0.1?
or does it have some error?

Hello,

There is no limit in the time step, and if Fudaa does not accept it (I did not know that...) you can just edit yourself the steering file. However 0.01 is small, unless you work with a scale model, so you can try higher values, look e.g. at the number of iterations of solvers in the listing, normal numbers are let's say, 5 to 15, and the important number to look at is the CFL number U*DT/DX, values around 1 are OK. Knowing the velocity U and the average mesh size DX, this gives you an idea of an acceptable time step.

Regards,

JMH

Hi Noah,

You may have look at a sample case which I have made of a 3m channel, with 8 inflow and 1 outfall. The outfall is with free fall condition as I have described earlier. I have used a time step of 0.05 s, and also tried with a lesser value of time step,the simulation had run properly. Apparently I don't about the exactness of the simulation results (Water depths and velocities) since I have not yet compared it with any analytical results (I may still have to correct the mesh and the simulation parameters in steering file)

The mesh size was very small, of order 0.2 m. I have used the criteria given by Jean-Michel Hervouet, however based on mesh size DX to get the acceptable time step.

With Best Regards,
Gourish

Hi, dear telemac user.
Now I'm going to simulate real river case.
its shape is like meandering channel.
so there are two curves.
but when I compute it,
Water flow is appeared near contour(boundary line)
I just put flowrate for inlet and elevation for outlet.
so I wonder about boundary condition for both sides which has not information.
If flowrate is not appeared around side boundary line,
I think it is ok. but it is appeared.
so I think that water can't flow through boundary but reflected.
It may occur error for real case.

Is there any idea to compute correctly?

Thank you to read my question.
Hope you have a nice day.