CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   Light Post (https://www.cfd-online.com/Forums/main/9705-light-post.html)

fj August 16, 2005 10:18

Light Post
 
Hello... I have not much experience with CFD and I do not know if is really necessary to use it in my problem. I need to estimate the stresses in a light post, the velocity of the wind is about 150 kph. I know also the conditions of humidity and dry bulb of the wind. In fact, I really do not know how to estimate the boundary conditions in the structural problem (without modelling the F-S interaction). Do I need to use ideal gas eq to estimate the mass and then to integrate the velocity in a lapse of time in order to calculate the force? (F= m dv/dt) I know that a good model of FSI could solve my problem, but I have no much time to explore it, but I want to estimate the stresses in a first hand in a simple structural model, without fluid. Maybe later if I have time, I will spend a little in a FSI model. I have experience with ansys. Thank you in advance! frank

Jim_Park August 16, 2005 11:19

Re: Light Post
 
I'd start by looking the cross-section of the post, which (I guess) is circular. Flow over a circular cylinder has been studied a lot. For a classic book with lots of data and photos, see Schlichting, Boundary Layer Theory. Probably any edition will do. The objective is to ESTIMATE the force from the drag coefficients tabulated for this flow.

First I'd determine if the conditions are right for vortex shedding from the post. That would result in period forcing of the post, which would I think result in a periodic boundary condition on your structural analysis. Without forcing, you still have to deal with any variation in your wind velocity.

Is it possible that you don't need detailed wind forces, just an overall force vs. time that can be imposed on the post model in your sress analysis? I'd guess yes.

Please note the liberal use of the word 'guess' above. Perhaps this will give you a place to start thinking about the problem?

fj August 17, 2005 10:41

Re: Light Post
 
Thank you Jim your advice was really useful.


All times are GMT -4. The time now is 11:55.