[bia]

its just my first workbench and it seems like a nightmare. . my problem is : a flow over a circular cylinder (diameter equal to 1 cm ) . the fluid is air .Reynolds number is 100 .unsteady .
i've tried to plot the drag and lift coefficient with fluent MONITORS ->FORCES ->DRAG ect . the ghraph givven is sinusoidal but the values are very low around zero. .and i must find 1.4 for the drag coefficient. . please help me

[tarjeiba]

You need to specify the reference values for the calculation of the drag coefficient. Remember,
,
where you need to provide the wanted flow variables. Fluent finds the drag force by integrating the pressure over the cylinder wall, necessitating a fine grid resolution. It then divides this by the applied reference values, giving you the calculated drag coefficient.

[bia]

THANKS TARJEIBA . but i've found somewhere that fluent gives you directly the drag and lift coefficient
. (that's why when i open the window of monitors
->drag -> edit ->surface = cylinder ) the choice of the surface is obligatory
so do you think that ther's a problem in the mesch (boundary layer ) or in the time step ?

[tarjeiba]

bia, to be able to calculate drag and lift coefficients you need to set the values you are refering to -- this is just part of the nature of such coeffecients. When calculating for a cylinder, it is typical to give the resulting drag force in relateion to the dynamic pressure of the free stream flow values. However, Fluent has no way of knowing what these values are, hence your need to specify them.

Just go to go to Solution -> Reference Values to set the values you want Fluent calculate the coefficients in relation to.

As the drag and lift coefficients are not in relation to the near-wall flow field, there really is no way for Fluent to know what values you want it set to. However, you do need to specify what surface you want Fluent to integrate the pressure forces over, as you've mentioned.

[bia]

thanks i've tried this exp cylinder with diameter D= 15mm so refrence area = Pi* D . depth =1 and length = Pi* D/2 so i found a drag coefficient about 0.988and i want to have 1.33.
the computational box is 750x400mm2.
Re=150.
FOR THE RESIDUAL I SET 1 e-08
max iterations per time step =40
these are tha best values i've made to be so closer but i still far away .
i tried to change the depth i fixed depth = 1.33/0.988 , length =1 .and i found the results exactly like i hope but why i musn't do this the depth must be related to a geometric caractrestic and not a mathematical error or coefficient right ? so how to find the way ? please help me

[tarjeiba]

Originally Posted by bia

thanks i've tried this exp cylinder with diameter D= 15mm so refrence area = Pi* D . depth =1 and length = Pi* D/2 so i found a drag coefficient about 0.988and i want to have 1.33.
the computational box is 750x400mm2.
Re=150.
FOR THE RESIDUAL I SET 1 e-08
max iterations per time step =40
these are tha best values i've made to be so closer but i still far away .
i tried to change the depth i fixed depth = 1.33/0.988 , length =1 .and i found the results exactly like i hope but why i musn't do this the depth must be related to a geometric caractrestic and not a mathematical error or coefficient right ? so how to find the way ? please help me

bia, it seems I've given you insufficient information, apologies. I meant to say that the drag force should be expressed as
,
where A is the area of the cylinder exposed to the stream. (The stream "sees" the cylinder as a rectangle of lenght L and height D.) Use this to set the reference lengths in Fluent.

(I still don't quite see why you need to run a simulation you seemingly already know the desired answer of.)

[bia]

thnaks a lot . in fact i must run the same calculation in a first step to be sure after this of any researched values ...i'll try really thanks a lot .

[tarjeiba]

Ah. Of course, that makes sense ... and sounds like good practice. Good look on the remainder of your research.

[bia]

you are really great i've tried as mentionned and the results were perfect thank you a lot

[flotus1]

You cant just change the reference values in order to make the results of a simulation match your expectations.
The results are inaccurate for some reason.
http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

[tarjeiba]

flotus1, of course I agree that you can't fine-tune your reference values to get the desired output. What you can do, however, as seems to have been bia's misstep, is to make sure that your provided reference values are of the same basis as the ones used in the calculations.

For the drag coefficient calculations, the typical reference area for a cylinder is the projected area (i.e., the diameter times the length of the cylinder), not the surface area of the cylinder.