CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   Problem with Post calculations (https://www.cfd-online.com/Forums/cfx/20567-problem-post-calculations.html)

 Josip August 18, 2004 17:16

Problem with Post calculations

I have a problem with calculating and/or understanding some essential parameters from the results file in CFX 5.7 Post. The geometry is very simple: a curved pipe of constant diameter with viscous fluid flow.

I have driven the RMS residuals below 1E-6, so the results must be very accurate. At the end of the solver report the "Pressure Moment on Walls" and the "Viscous Moment on Walls" values are displayed. The total torque on the walls around the Z axis are made of the viscous and pressure components.

I wanted to calculate these two components with the calculator using integrals and compare the results with those displayed by the solver report. When calculated the viscous moment by integrating the (Wall shear*area*radius) then I have got the same result as indicated by the solver report.

However, when integrated the pressure forces, I have got something very different from the value in the solver report. Here is the way I have calculated:

User functions:

fi=atan(Normal Y/Normal X)

Ptan=Pressure*Normal Y*cos(fi)-Normal X*sin(fi)

r=(X^2+Y^2)^(1/2)

In the calculator:

AreaInt

Wall

Ptan*r

Or:

Sum

Wall

Ptan*Area*r

While the solver report indicates a pressure moment of about 71 Nm on walls, when integrating the pressure over the tangential surface projection weighted with the radius using the above formulas the result is about 6 E-6 Nm. This is a huge difference, and something must be wrong. When calculating the total torque over the wall surface the result is in consonance with the moment components from the solver report, but not in agreement with the 'manually' calculated components.

If anybody has experience with this issue and knows the answer, please enlighten me where is the error and how should I perform the integration to get the correct result (if the described formulas are wrong). Thank you in advance.

Josip.

 All times are GMT -4. The time now is 15:59.