[fimbull]

Hello,

im trying to reactivate some OF Code, wich is made in 2.1.x.

I have a rotating inner Region and a not moving outer region in my case.

At the interfaces i have used cyclicAMI BCs but after a few timesteps the simulation is aborted.

I think i have figured out the problem. I has something to do with my BCs at the sliding interface plane.

So my question is: Do i really need fixedValues for the cyclicAMI Boundary condition?

[Antimony]

Hi,

For AMI, to the best of my knowledge, the BCs for the cyclicAMI should simple say cyclicAMI.

Example, in the "p" file if there is a boundary patch called "hello" that is an AMI patch, then:

hello
{ type cyclicAMI; }

Hope this helps.

Cheers,
Antimony

[fimbull]

Well to make my problem more clear, i will post exactly my current cyclicami patches and my thoughts about it (Sorry english isn't my first language).

Im using 3 variables in my case, these are p_rgh, U and alpha.

for my patches in the 0 Folder im using these conditions:

for U

...
    insideSlider
    {
        type            cyclicAMI;
        value           uniform (0 0 0);
    }
    outsideSlider
    {
        type            cyclicAMI;
        value           uniform (0 0 0);
    }
...

for alpha1

...
    insideSlider
    {
        type            cyclicAMI;
        value           uniform 1;
    }
    outsideSlider
    {
        type            cyclicAMI;
        value           uniform 1;
    }
...

for p_rgh

...
    insideSlider
    {
        type            cyclicAMI;
        value           uniform 105000;
    }
    outsideSlider
    {
        type            cyclicAMI;
        value           uniform 105000;
    }
...

I dont think it make sense to have a plane inside my mesh in which are all variables fixed. This will somehow split my calculations into one in the rotating part and in one in the static part of my mesh. Am i right?
These BC's are given from my predecessor who have worked on these case in the past.

[Antimony]

Hi,

As long as you have stated "cyclicAMI" as the type, the "value" field is not needed. So the value fields of "(0 0 0)" are redundant.

In your constant/polyMesh/boundary file, make sure that the cyclicAMI patches have been defined properly. Since yours is a stator-rotor problem, I suggest you look at the mixerVessel2DAMI tutorial under pimpleDyMFoam.

Hope this helps.

Cheers,
Antimony