CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   convergence check:grid and time-step size (http://www.cfd-online.com/Forums/main/106005-convergence-check-grid-time-step-size.html)

Lilly August 15, 2012 06:53

convergence check:grid and time-step size
 
Dear all,

I want to do a convergence check for the spatial resolution (grid) and the temporal resolution (time-step-size) of my simulations. I want to do some simulations using pulsatile flow and compare them with simulations using the average velocity of this pulsatile profile. Furthermore I want to do some simulations with the profile upscaled about 2 times.
Do I have to do the convergence checks for all of this situations or is it enough to do them with the (let's say) most dramatic situation: pulsatile, upscaled (large velocity and velocity-gradients)?
And is there kind of a rule how to proceed with these 2 convergence checks: how many grids, how many time-steps, how to combine them?
I am really thankful for any idea and help!
Lilly

FMDenaro August 15, 2012 07:57

Quote:

Originally Posted by Lilly (Post 377173)
Dear all,

I want to do a convergence check for the spatial resolution (grid) and the temporal resolution (time-step-size) of my simulations. I want to do some simulations using pulsatile flow and compare them with simulations using the average velocity of this pulsatile profile. Furthermore I want to do some simulations with the profile upscaled about 2 times.
Do I have to do the convergence checks for all of this situations or is it enough to do them with the (let's say) most dramatic situation: pulsatile, upscaled (large velocity and velocity-gradients)?
And is there kind of a rule how to proceed with these 2 convergence checks: how many grids, how many time-steps, how to combine them?
I am really thankful for any idea and help!
Lilly


be careful in taking convergence check by fixing h or dt... the monotone convergence study should be performed fo vanishing ratio dt/h, when you fix one parameter you must be sure that the value is very very small otherwise your convergence will be altered when the error reaches that magnitude order


All times are GMT -4. The time now is 19:59.