CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Curved flow (http://www.cfd-online.com/Forums/cfx/127919-curved-flow.html)

kmgraju December 25, 2013 22:08

Curved flow
 
Dear all
My geometry has a curved pipe. the fluid entering +y direction (inlet as pressure). The outlet axis coordinate has (+x,-y) coordinate. So I introduced outlet (velocity) as Normal to the surface.

My question is if I rotate the component in the workbench so that the outlet is along X axis, How about my result in this case and previous case? is it same or different?

Thank you

Govind

ghorrocks December 26, 2013 00:13

If you rotate the thing and do the same simulation you should expect the same result, just rotated. Is that what you are asking?

Or are you asking if you can use a previous rotated simulation as an initial condition? That is harder - in general, no, you cannot.

kmgraju December 26, 2013 05:31

4 Attachment(s)
Quote:

Originally Posted by ghorrocks (Post 467687)
If you rotate the thing and do the same simulation you should expect the same result, just rotated. Is that what you are asking?

Or are you asking if you can use a previous rotated simulation as an initial condition? That is harder - in general, no, you cannot.

Thank you Very much
I have another question. Please find the attached pictures. I am confusing about turbulence model.I run the steady state first and then transient . My inlet is pressure and outlet is velocity . I am expecting low Reynolds turbulence occurring near the curve and the area restriction . Is my solution is converged or not.I tried SST model .I also tried tried many models like RNG K-epsilon, K-omega etc , but the solution doesn't meet the convergence criteria 10^-4
Give your advice
Thank you
Govind

ghorrocks December 26, 2013 05:46

FAQ: http://www.cfd-online.com/Wiki/Ansys...gence_criteria

kmgraju December 26, 2013 10:36

Quote:

Originally Posted by ghorrocks (Post 467717)

Dear all
please find the following statement.U Mom, v-mom, Vmom, P-mass all are F.
This will start from 70th iteration t0 100 iteration . what does it mean.

Thank you

+----------------------+------+---------+---------+------------------+
| U-Mom | 1.00 | 3.8E-04 | 9.5E-03 | 1.0E+00 F |
| V-Mom | 1.00 | 3.8E-04 | 7.5E-03 | 1.0E+00 F |
| W-Mom | 1.00 | 5.2E-04 | 1.7E-02 | 1.0E+00 F |
| P-Mass | 1.00 | 1.0E-04 | 5.4E-03 | 8.9 1.0E+00 F |
+----------------------+------+---------+---------+------------------+
| K-TurbKE | 0.99 | 8.7E-05 | 4.4E-03 | 5.7 8.1E-02 OK|
| O-TurbFreq | 1.05 | 8.3E-06 | 8.6E-04 | 10.5 5.8E-07 OK|
+----------------------+------+---------+---------+------------------+
| Rethetat | 2.17 | 6.1E-04 | 2.8E-01 | 5.9 1.0E-02 OK|
| Intermit | 0.77 | 6.0E-05 | 1.4E-02 | 5.7 5.8E-03 OK|
+----------------------+------+---------+---------+------------------+

================================================== ====================
OUTER LOOP ITERATION = 88 CPU SECONDS = 2.320E+03
----------------------------------------------------------------------
| Equation | Rate | RMS Res | Max Res | Linear Solution |
+----------------------+------+---------+---------+------------------+
| U-Mom | 1.00 | 3.8E-04 | 9.5E-03 | 1.0E+00 F |
| V-Mom | 1.00 | 3.8E-04 | 7.5E-03 | 1.0E+00 F |
| W-Mom | 1.00 | 5.2E-04 | 1.7E-02 | 1.0E+00 F |
| P-Mass | 1.00 | 1.0E-04 | 5.4E-03 | 8.9 1.0E+00 F |
+----------------------+------+---------+---------+------------------+
| K-TurbKE | 0.99 | 8.6E-05 | 4.3E-03 | 5.7 8.0E-02 OK|
| O-TurbFreq | 1.03 | 8.6E-06 | 9.1E-04 | 10.5 5.3E-07 OK|
+----------------------+------+---------+---------+------------------+
| Rethetat | 1.16 | 7.1E-04 | 2.6E-01 | 5.9 1.1E-02 OK|
| Intermit | 1.06 | 6.4E-05 | 8.7E-03 | 5.7 7.1E-03 OK|
+----------------------+------+---------+---------+------------------+

ghorrocks December 27, 2013 05:34

This is explained in the documentation. It means the linear solver is diverging. Check that your simulation is correctly setup, improve your mesh quality, use smaller time steps, double precision numerics or a better initial condition.


All times are GMT -4. The time now is 13:01.