Problem in fire propagation with grid stretching

[Ritambhara Dubey]

Hello,

I was testing a case similar to C064 from the validation of the CSIRO grassland fire. The difference between the validation case file (C064) and my inputs are:

- Horizontal grid size

- Domain height and grid size in the vertical direction (z-direction). 

I tried two cases (Input file attached):

Case 1- uniform grid in the z-direction (2 m)

Case 2- grid stretching in z-direction: polynomial stretching with a minimum grid size of 1 m at the bottom and 5 m at the top of the domain.

For some reason, the fire propagates well for Case 1. But for Case 2, the fire does not propagate well. When the grid size at the bottom is fine, there should be no problem with fire propagation. I am not sure where the problem is coming from.

For reference, I am attaching the following files:

- Input file- case 1 & 2

- Temperature contour for cases 1 & 2 at 25 seconds after ignition.

Thank you for your help.

[Kevin McGrattan]

Why are you doing this?

&MESH IJK = 16,16,1250, XB = 0,20,0,20,0,2500.0, TRNZ_ID='my trnz' /

I would not recommend such small grid spacing in the predominant direction of the wind.

[Ritambhara Dubey]

The two reasons I want to do this are:

- Because I want to capture combustion better (accurate HRR)

- With grid stretching, I can use less grid points in z-direction and carry out LES simulation

Is there any other alternative way to achieve my objectives that you recommend?

--
You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/CAAJimDFD62jKq--aZFtvT9uXjWsJVb%2B%3D3%2BVRw-C4dO-gZWoOAQ%40mail.gmail.com.

[Kevin McGrattan]

Inaccuracy in the flow is greatest at mesh boundaries. Your grid maximizes the grid to grid error. I suggest you use grids like the ones that are used for the cases in the FDS repository. 

I also tend to discourage grid stretching. Your accuracy is only as good as the largest cell dimension.