CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   NUMERICAL INSTABILITY FOR FLOW OV. SWIRLING VANES (http://www.cfd-online.com/Forums/fluent/35682-numerical-instability-flow-ov-swirling-vanes.html)

D February 8, 2005 15:44

NUMERICAL INSTABILITY FOR FLOW OV. SWIRLING VANES
 
Hi!

Has anyone observed the numerical instability over the swirling vanes. What we have is the flow over a set of vanes. We have pressure inlets and one pressure outlet. What we are observing is the mass flow rate through the pressure outlet is oscillating in a sinusoidal fashion.

Has anyone come across this kind of situation and has any comments/suggestions. We have tried with lowering the under relaxation factors. Even to a small number like 0.001, the mass flow rate is still sinusoidal.

helper February 8, 2005 17:36

Re: NUMERICAL INSTABILITY FOR FLOW OV. SWIRLING VA
 
Some strongly swirling flows can have their vortex cores precessing - the vortex axis undulates in a snaky and unsteady fashion. Some advanced RANS models can often capture this phenomenon showing a limit cycle-like behavior of the residuals or oscillations in the solutions fields even in steady simulations. So, what you're witnessing may not be "numerical instability" but probably a physical one.

nobody February 9, 2005 16:20

Re: NUMERICAL INSTABILITY FOR FLOW OV. SWIRLING VA
 
where is your pressure oulet located? is it at appropiate location?

D February 9, 2005 16:35

Re: NUMERICAL INSTABILITY FOR FLOW OV. SWIRLING VA
 
pressure oulet is about an inch downstream from the end of the end of the vane.


All times are GMT -4. The time now is 06:18.