Hi all

Iam simulating convective heat transfer in a closed cylinder. I want to calculate the total heat flux from the wall at different temperature input. As iam increasing grid size, the value of Q is also increasing.. How do i know my results are correct and grid independent?

Thanks

regards Rashmi

Hi,

Heat flux calculation is not only depends on the no of elements. Check you Y+ value, near wall treatment in turbulence model and reference values on each case file. If there is variation, it affect the heat flux calculation. Otherwise try with higher order schemes especially for energy with good convergence.

with warm regards, JSM

Hi

Iam simulating stationary fluid is a closed cylinder. lest wall is at 310K and right wall at 285K. the curved surface is adaibatic. iam uisng higher order scheme and my solution is converging at 1e-6

regards

Hi,

You'll see grid independence if your values 'hone in' on a particular value. Just keep refining your grid, start with coarse and refine, not the other way round like you appear to be doing.

As JSM said, keep an eye on wall y+ values you want them between 30 and 300 using standard wall functions. If grid refinement is making the y+ too small then I suggest a super-fine grid with enhanced wall treatment but y+ needs to be pretty small, less than 5 I think - check the user guide though to be sure.

Once you have done your verification you need some form of validation to get an idea if your results are in the correct range. I suggest a literature search for heated cylinders!

Good luck,

Carlos.