[mattijs]

a) all patches are 'polyPatch' (base type). So any fvPatchField type that only needs a polyPatch will run happily (unless an explicit check has been coded)

b)any fvPatchField that needs extra information will (or should) check the patch type.

(for me symmetry on wall patch gives:
FOAM FATAL ERROR patch 0 not symmetry type. Patch type = wall

cyclic ,, gives:
FOAM FATAL ERROR : Attempt to cast type wall to type lduInterface

empty ,, gives:
FOAM FATAL IO ERROR : patch 0 not empty type. Patch type = wall
)


The behaviour of blockMesh comes from it trying to interface with FoamX and preserve the 'physicalType' FoamX sets. The coupling is not optimal as you've noticed.

[maka]

I get the same messages in case b) but in case a) I only get a message in case of symmetryPlane.

Actually, I ran a computation where I changed the b.c. in blockMeshDict to empty and forgot to change it in the field file and I got no warning. Later I wanted to check what b.c. I applied so, I checked the filed file which was (fixed value) inconsistent with the b.c specified in blockMeshDict. That is why I thought that if it is easy to implement such checks, in this way one can't get a wrong impression about what b.c. he actually applied, whether he checks the field file or blockMeshDict. Offcourse, one can always check the result.

Thanks for the clarification regarding connection to FoamX. Thanks!

Best regards,
Maka

[qiu]

hi, all
how to create a new boundary condition changes with both time and coordinates,the boundary is still ,only the inlet velocity changes.I tried to modify from the oscillatingFixedValue,but as
I'm a freshman to C++,I failed all this week.Could someone be so kind to help me ?Thanks.

qiu

[johny]

Hi OF'ers,

I'm simulating a nozzle and need to try with fixedvalue bc for my inlet. Pressure for the inlet is given zerogradient. And the bc's are switched for the outlet.
If I give an fixed value bc for the inlet. let's say 10...somehow when it moves on to the next time step it automatically changes to 1. I'm working with rhoSimpleFoam.
Do anyone have an idea why it happens?
Looking forward to your reply,

Thanks

[Luise]

Hello,

i don't understand the behaviour of freestream/freestreamPressure BCs.
Here is what i can figure out by reading the sources:

- freestream is inherited from inletOutletFvPatchField<Type>. There is no updateCoeffs(), so it doesnt calculate anything. However the inletOutlet-Class calculates phip, and so evaluates the blendingfactor for zeroGradient and fixedValue.

- freestreamPressure is inherited from zeroGradientFvPatchScalarField. There is a updateCoeffs() function declared and defined, which calculates phip by the flux normal to the patch.

As far as i understand, by using freestreamPressure for pressure and freestream for velocity, the pressure is set to zeroGradient everywhere and the velocity is kept as the inletOutlet-BC prescribes.

But then, what are the reasons for implementing those boundaries? And why is phip evaluated in the BC for pressure?

i hope somebody can help me on that.

Luise