HI Muhammad Atif,

It has two forces on the surface: the pressure force acting on a surface element, and the viscous force. The first one is named shape drag and the second one viscous drag, respectively.

Fp = -p.n*dA

Fvis = (-viscosity*d(u)/dn@ wall )*dA

Cd = ||F||/(0.5*density*||U||²

which force you will consider is case-dependent on. The pressure distribution over the body will generate a force if you multiply by the area ( negative signal mean that the pressure acts against the unitary surface normal vector.

The net force will be calculated by means integral on the area. And you can use normal components to calculate each force component or calculate all pressure force on the surface and after that, calculate its area average.

Viscous force is normally calculated by means of wall shear variable - precalculated by solver or post-processing, considering or no the pressure contribution. Is a natural result of most commercial codes such as FLUENT and CFX.

The force can or not consider the pressure force. It`ll depend on the code. In doubt, do it on hand.

With the force on hands, use the velocity magnitude ( next to the wall, the element average, the conservative variable, tangential or normal projected, etc... there are many choices of which velocity will be used and density to evaluate the drag coefficient at each area element. I advise you to create new variables before running in order to have all these parameters calculated during the solver solution. Hence, you will have these fields available for post-processing.

Bests!

hello Muhammad Atif hopefully doing well by the grace of Allah. its a long term process to explain you please find the link attached and see the video hopefully you will get your answer.

Thanks

link:

https://youtu.be/atK1cVtHOms

Choose the frontal area as the reference area in fluent as it would be the wetted area. This area we use in calculating drag coefficient in fluent. Set reference values in fluent correctly to get drag coefficient

Hi Ricardo,

Thank you for sharing detailed answer. I have actually done CFD analysis using ansys and currently working on the wind tunnel. I don't have load cells available to measure drag force directly. Thats why I was thinking to use pressure and velocity field and then use some pressure and force relationships to calculate drag force.

Hello Muhammad, As frequently as my name appears on these articles, I am but a humble optical measurements guys. My colleague, James C. Ross should be able to answer your question.

Hi Muhammad,

We generally follow exactly the path which you have started in order to get force estimates from pressure sensors. From the simulations you have an array of possible pressure distributions, and you can assume that even if they are not exactly correct, that they qualitatively illustrate the set of potential flows which you will measure. Then decide on an integration method and number of sensors, and make a distribution where you extract those surface pressures from the numerical data. Then compare with the known force from the simulation. For instance decide on 10 sensors on the trailer rear and compare the integrated force from the simulation with the infinite integration. Then change the senor positions and repeat.

If you have a good comparison between the integrated force and the full surface output, than you can be reasonably sure that the sensors are monitoring all of the important aerodynamic phenomena. However, even in the best case, the comparison between experiment and CFD should be done by comparing experimental and numerical pressures integrated by the same algorithm. Our experience is that the actual integration algorithm is not very important, and you can just hardcode a constant area multiple for each presure sensor.