[stud-many]

Hey there.


I have issues with a LargeEddyModel which i built using OpenFOAM (1612+)


geometry, case:


chordlength c =0.13 m
domain_length = 3*c =0.39 m

Mach_inlet = 0.2
Re_chord = 600.000
compressible LES / rhoPimpleFoam

WALE-SGM



mesh:


nodes : approx. 11million
type & topo :blockstructured / hexa


BC:



U: DFSEM-inlet & wave-transmissive (non-reflective) . fieldInf at INLET
p: wave-transmissive. fieldInf at OUTLET
T: wave-transmissive. fieldInf at INLET


1:1 periodicity in pitch-&span-direction



After a while I observe oscillations that propagate in pitch-direction. I can't figure out where those oscillations develop and where they get their energy from. Using inviscid walls instead of periodic BC's do help with the issue - but this is clearly not an option.







CLICK HERE FOR ANIMATION OF OSCILLATIONS USING PRESSURE-FIELD
https://gifyu.com/image/SMPaT








what I tried until now:


-no synthetic turbulence (only wavetransmissive)

-mesh optimization, especially at Leading- and Trailing-Edge
-different incidences
-bigger inflow-length (2*c = 0.26m)
-higher viscosity (lower reynolds-number)
-inviscid walls (NO OSCILLATIONS HERE)
-bigger pitch (currently working on it)




Does anybody have ANY idea where those oscillations are coming from? I am kind of frustrated as I am out of ideas...


If you need more information (BC's, plots, ...) please let me know