open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi all,
I was hoping to get some insight about coupling models (T3 +tomawac). I'm using V8p1.

I have a T3 model coupled to tomawac forced by same winds.
There are 2 simulations: (i) coupled T3+tomawac (ii) only T3.

In (i) I get 2 files: T3 results file with UVW & tomawac results files with UV.
In (ii) I get just the T3 results files with UVW.

The issue is that with the coupled T3-tomawac model, the UVW results are identical to the T3 only model. Exactly like the models aren't coupled, so I thought the tomawac results might be zero. However, when I view the tomawac results UV files, the velocities are reasonable and comparable to the T3 magnitudes, so there should be some deviation in the coupled T3 currents.

So it seems tomawac is computing wave driven currents but it doesn't seem to relay it to telemac3d. Am I missing something or if anyone has come across this?

I have:
TOMAWAC STEERING FILE= 'myfile.cas'
WAVE DRIVEN CURRENTS =TRUE <--- should this be 1 instead of TRUE??
COUPLING WITH ='TOMAWAC"
COUPLING PERIOD FOR TOMAWAC=1
The issue seems similar to this 8 yr old issue (fyi).

Hello

For the wave driven currents, yes true and 1 are equivalent.

As for your problem, (no difference between coupled or not), I don't see explanation, but the first thing is you should always use the last version. Some things change between versions and sometimes the problem has already been fixed.

regards

I have V8p1. Do you mean to get V8p2 and see if the issue remains?

You can even try with the newest one : V8P3R1 and see if the issue remains.

Hi again,

I have installed v8p3r0 (link on left of page here) and the results are the same (t3+wave = t3 results).

I also tried to implement t3+3d_tomawac in previous version v8p1.
This worked for a bit before a crash "GMRES(BIEF) in Propogation & Diffusion with wave equation resulted in NaN". I expected the crash since it wasn't tuned (just activated).

However the saved files for this (before crash), show that the (t3+3dwave was different to t3 run), i.e., the wave-driven current and T3_only was roughly aligning to the t3+3dwave results.

The installation seems fine because the t3+3dwave run was exchanging info.

I feel like in the t3+2d_wave option, I'm missing some activation keyword beyond the 4 expected from the manual. Something that says "pass info AND use that info". Or maybe some keyword has been defaulted to True (in version updates) that is preventing the use of the exchanged data.

TOMAWAC STEERING FILE= 'myfile.cas'
WAVE DRIVEN CURRENTS =TRUE
COUPLING WITH ='TOMAWAC"
COUPLING PERIOD FOR TOMAWAC=1

Hello

I don't know what is specific to your case, I just made the test on the example given with telemac, the case 3Dcoupling

I just did the simulation with
t3d_littoral.cas
and then again deleting the lines :

WAVE DRIVEN CURRENTS = YES
COUPLING WITH = 'TOMAWAC'
COUPLING PERIOD FOR TOMAWAC = 10
TOMAWAC STEERING FILE = 'tom_3Dlittoral.cas'

and you see the impact of waves on the velocity in t3D (i join the image of the difference)


I don't see any reason why in your case there would not be any impact.

Thanks for the testing.
I will try this example and see if it works.

The case without coupling gave zero speeds (I assume you got the same!?)
The case with coupling gave a figure similar to your image (I have attached Blue Kenue image).






I have attached my cas files. Please let me know if there is anything is off that could be causing this.

Hi again,
Could this be an issue:
My wind ascii file in t3 has timestamps 0 to 6 days.
The tomawac wind file is in serfind but there is a date stamp on it, i.e., 2021-12-10 1:00:00 etc.

Both files are hourly for 6 days. And I reset the INITIAL TIME SET TO ZERO=TRUE in tomawac and T3.

Could it be they don't couple because tomwac doesn't know where to slot its computation (even though the times are reset)?

Hello

I am sorry but i didn't find any point in your cas file that would explain the problem.

As for time stamp in the wind, the only possible consequence could be your Hm0 is 0 because your wind would be 0. But i guess you have checked that Hm0 is not 0 if you are waiting for differences between coupled and not coupled...

Talking about timestamp you have a keyword TIME SHIFT IN WINDS FILE when the time of your wind file has an offset.