CFD Online Discussion Forums

CFD Online Discussion Forums (
-   STAR-CCM+ (
-   -   non-physical solution with periodic boundary in ccm+ (

arihant August 12, 2009 15:53

non-physical solution with periodic boundary in ccm+

I am testing starccm+ for simulating fully developed laminar flow in a rectangular channel and circular pipe. I am using the periodic boundary condition and I am getting non-physical results.
I created a Periodic interface by selecting the inlet and outlet boundaries.
I selected Fully developed Interface as the property under Interfaces node and selected Pressure Jump under Physical Conditions node and specified a Pressure Jump value under Physical values. I am using Segregated solver with 0.001s as timestep.
I am simulating a Reynolds number of 100. Channel dimensions are 3cmx3cmx10cm. I specified Pressure jump from \Delta P = \rho f L u^{2}/2D_{h}, where f = 64/Re (probably not true for rectangular channel but just as a test) and Hydraulic dia. D_{h} = 4*Area/Perimeter = 3cm. Density of water is 1000kg/m^3. So, \Delta P = 10.67 Pa. for an Avg. velocity (u) of 0.1m/s

After about 400 iterations there is reversal of flow in the entire domain which is non physical. I changed the Fully developed flow option to Mass Flow rate and specified a Mass flow rate of 0.2 kg/m-s. This time there is flow from center of channel to both inlet and outlet which again is non physical.
Am I missing something in specifying Periodic boundary condition ? Can any one point out where I am doing wrong.


SKK August 17, 2009 05:16

We had a similar problem with periodic boundary conditions showing flow reversal and when contacted adapco was told there is a bug in that version. It was in version 3.0... I would assume its been fixed in version 4. Best is for you to contact the support directly about this.

arihant August 17, 2009 15:30

Thanks SKK. The version I am using now is 4.02.007. I will contact the tech support and see what they say.

All times are GMT -4. The time now is 02:15.