open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi there,

I have created a fairly simple model of a channel in which I apply a force to the flow to simulate vertical structures using the dragfo.f file. The north and south walls of the channel are slip surfaces, and the east and west walls are liquid boundaries.

On the eastern wall (channel inlet) I specify the rate into the channel, and on the western wall (channel outlet), I specify a height. I specify an initial free surface elevation when I begin the model. This works correctly.

What I do not seem to understand about the way Telemac works with boundary conditions is that I cannot seem to run a simulation correctly if I apply a volume flux at the channel outlet. It would be quite good to be able to specify a velocity at the inlet and volume flux at the outlet, but I have found that this leads to the surface elevation in the domain increasing way too much. Similarly, specifying a height at the inlet and volume flux at the outlet leads to the solution exploding. Attempting to specify a height and velocity at the inlet and volume flux at the outlet doesn't seem to work either. I notice that in the validation cases, it always seems to be the case that a volume flux is specified at the inlet, not the outlet. Is there a reason why this is so and is there a way to get around it?

Thanks,
Chris

We are going into fundamental hydrodynamic modelling here. For a thorough explanation, read any of the many text books on this topic. In short, you are trying to solve a flow, that in principle is one dimensional (forgettig about any structures you put in). This needs to solve 2 equations, one momentum equation and the continuity of mass. This requires one condition at each boundary to close the problem mathematically. One for the surface level and one for the velocity. Telemac considers drage near the bed. In a channel flow, this means that the change in surface level are related to the velocities in such a way that the surface levels can only be calculated in the upstream direction. Therefore a water level is required at the donwstream boundary. Hope this helps to understand, but I do suggest you read one of the text books on hydrodynamic modelling. Look for 'backwater' calculations as term for channel flows.

Michiel