open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I'm working on an existing model (with Mascaret 3.2.1) with 2 junctions.
In the original model the option "automatic head loss at junction" is not used. There are only 3 reaches for each junction and I get results alright.

I will need to compare results obtained with the unsteady and transcritical kernels. So I modified the original model in order to work with the transcritical kernel (eg. weir adapted for transcritical kernel). I entered the junctions' normal coordinates and selected the "automatic head loss" option, which is compulsory for the transcritical computation.

I get errors messages such as :

Problem 1 with transcritical kernel : (Error 107)
"Probleme de definition de la geometrie au confluent No 2"
"Geometry definition problem at juction Nb 2"
=> and in this junction, reaches 2 and 3 in the Mascaret listing box are indeed computed a second time (with normal coordinates and angle totally unkown to me...)

Problem 2 with steady kernel : (Error 88)
"Le nombre de branche aval : 2 de la confluence No 1 est superieur à 1"
"The number of downstream branch : 2 of junction Nb 1 is superior to 1"
Yet I only have 3 reaches going to each junction.

I tested several cases (such as erase and re-enter the junction Nb 2) and always gets the same errors. Does anyone know where theses problems could com from and how to solve them ?

Thanks in advance for any tips !

Hello,

First of all, the options "automatic loss" is not available with the transcritical kernel.
Secundly, i think that you have an error in the definition of the junction : the link between reach and junction.

Best regards

N. Goutal

Thank you very much for your answer !

I probably mixed up 2 things indeed. When the "automatic head loss at junction" is selected with the unsteady kernel, I get access to the local 1D-2D modelisation of the junction, similar to the single one available with the transcritical kernel. I therefore used the same "name" in both cases, my mistake.

So far, I just attached the reach to the junction and the connection was taken into account with a red spot. Would there be a special link to create between reach and junction, to get a proper definition of this junction (especially with the local 1D-2D modelisation) ?

Thanks