CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)

 Jonas Pedro Caumo April 15, 2007 13:03

Wind tunnel of a turbine blade tests.

I´m doing a simulation of a wind tunel cascade of turbine blades, i have a lot of variables from my project calculation and i know that teh results are right, because this turbine is operating in reality, but i want to improve its efficiency, so i decided to use CFX to simulate the flow in 3D. At first i try to validate the programa by calculating the cp (pressure coefficient), i have Mach=0.5, and Reynolds=4,5x100000. I´m using hex mesh, aboutr 1,5 millions. My inlet region is 1.0(chord) from the blade and at exit i put 2.0(chord). I used a lot of configurations on CFX-Pre, i try to put pressure (total) at inlet and static pressure at exit (0 atm relative)(1). I put pressure (total) at inlet and mass flow at exit(2). I put mass flow at inlet and static pressure (0 atm relative) at the exit(3). I found a lot of diferent results. What do you think it is better (1), (2) or (3) ? I know that (1) have the better convergence. I´m calculating cp (pressure coefficient) by exporting the results and using excel or tecplot to make the graphics in the middle of the blade height. I know that the maximum value of cp is 1.0, but i´m not achieving a good result. Is there anybody to help me. Thanks for the help.

 Ermanno April 16, 2007 02:43

Re: Wind tunnel of a turbine blade tests.

In my simulations I use configuration # 1 with a convergence RMS < 10^-6, when it is possible : generally when I calculate the loss coefficient (I have some experimental values of that one) I find values a little higher than the experimental results, but they are acceptables. In your case what is the amount of the difference (about) between the value of a variable that you find (in configuration 1,2 or 3)and the value that you expected ? Are they much difference?

Ermanno

 Jonas Pedro Caumo April 16, 2007 17:02

Re: Wind tunnel of a turbine blade tests.

Comparing one to each other they are in percentage 8%different, but comparing with the experimental results i found a little higher values than the experimental results. I use RMS = 10^-6, and for this configuration, when i put pressure (total) at inlet and static at exit i found the better convergence, but why i´m doing this i´m telling the program the loss in a inderectly. And about your elements, how many ? Thanks. Jonas.