open TELEMAC-MASCARET The mathematically superior suite of solvers

Dear Community,

Ive been trying to model an erosional breaching process of an earthdam embankment that has been done in a wave flume. In order to see if my model works correctly ive compared with some data that has been obtained in the lab ( in this case i was comparing the flow hydropgraphs). Unfortunately my results differ a lot to those obtained experimentally. Ive been computing the flow right after the embankment( between 0 and 0.2 m see black line in Q_area_downstream.png) and i obtain a negative flow( see Q_result.png).This happened mostly all the times i tried, eaven if at some point I was able to get a positive result the flow was mirror inverted and still with 50-60% error.Ive double, tripple checked the dimensions of my mesh the input parameters of telemac and sisephe. I dont know where my Error could be. I know that this is a very superficial description of my problem and the error could be anywhere. I wish i could be more precise with the information. I hope somebody can give me maybe an advise which other parameters i could consider or have a quick look at my data and detect an error... Ive been turning in rounds for a while.... Thank you so much.

Kind regards
Christian

Hello,

looking in the flow direction you should define the line from left to right in which you integrate the discharge.

A remark:
I think your outlet boundary condition is too close to your zone of interest. You provide for the whole simulation a constant depth / water surface level of 1 cm or does this condition resemble the laboratory conditions? Otherwise I would extend the mesh downstream and in this case you could see this part as a collecting pond for water and sediment mass and you don't have to prescribe any liquid boundary anymore.

Once I did a very similar dam-break study for which you can find all the input files somewhere in the forum. I published a paper in the Telemac User conference 2014. Maybe it has some useful informations for you.


Best regards,
Clemens

Hello Clemens thanks for your quick reply i will try it out right now.

I settled my initial conditions that way because i had some unwanted high oscillations at my downstream boundary condition or eaven my bottom got eroded as well. Initially i wanted to set my BC as Open B with prescibed H with a free tracer (444). The inital overtopping water depth i prescribed it in my condin file as 0.4m as depicted in flume.png.So you say i should enlarge my flume and this way i wouldnt need any liquid BC anymore?. But what should i use than as my BC-file? should i prescribe everything as solid walls? Ive been working with this software for 3 weeks now so these questions might be trivial for you ^^ Ive been looking for your documents in your previous posts but i couldnt find them so far... Is there a download page of all the papers of the telemac user conferences?

Thanks again Clemens

Have a nice Day

Christian

Hello,
yes, correct. You can create a pond downstream of the dam and prescribe all the walls as impermeable (2 2 2).
However, if you use the Finite Volume scheme HLLC for the outlet you can prescribe also a "free outlet" (4 4 4) if you are sure that there you have supercritical flow. I don't know exactly why but with the other schemes this condition doesn't work that good.
Enclosed you will find a dam-break case where I defined a constant water level at the inlet.
At the Telemac website there is a link to the 2014 proceedings and if you are logged in you can download the proceesings.

By the way, the questions are not so trivial and 3 weeks in the Telemac world is not that much

Hope this helps,
Clemens

Perfect! Thanks a lot i will play around with it a bit and hopefully i get out of this mess finally ^^ I will let you know how it went and maybe come up with some new questions haha.

Have a good one

Christian

Hey Clemens,
so i tried both variations. First the one including a longer section to my flume and secondly the one you did in your work with Konrad Moser including a basin downstream. The first one gave me a hydrograph similar to the ones i obtained before without enlarging the flume. Only that this time the flow got higher and at some point reflection phenomena probably blurried my results. Unfortunately i wasnt able to model your variation since my bottom and watersurface jumped to infinity when watching the animation so something wrong. Do you know what would be the reason for the higher flow and the problem with your variation?
I will include the files just in case....

Thanks again

Viele Grüße

Christian

Hey Clemens,
so i tried both variations. First the one including a longer section to my flume and secondly the one you did in your work with Konrad Moser including a basin downstream. The first one gave me a hydrograph similar to the ones i obtained before without enlarging the flume. Only that this time the flow got higher and at some point reflection phenomena probably blurried my results. Unfortunately i wasnt able to model your variation since my bottom and watersurface jumped to infinity when watching the animation so something must be wrong. Do you know what would be the reason for the higher flow and the problem i had with your variation?

Thanks a lot again

Viele Grüße

Christian