Fixed midflameheight to use ceiling instead (test OK).

[drnimbusrain]

No description provided.

[zmoon]

does this imply the top flamelay could be only partial?

[drnimbusrain]

@zmoon Can you further clarify? I just wanted to move it to the upper layer if there is a partial flamelays. Maybe the 2nd ceiling isn't necessary?

[drnimbusrain]

@zmoon FYI. Wei-Ting just tested this new formulation, and the WAF results are looking really good against the Massman example figures in his paper.

[zmoon]

I thought that originally you wanted to ensure that flamelays were all full-flame layers. But with flamelays = ceiling(FLAMEH/MODRES) + 1 the top one could be only partial flame.

Seems like another situation where it would be useful to be able to differentiate between interface and "mass" levels.

[drnimbusrain]

@zmoon Ah, yes I see what you are saying considering my previous thoughts. You are right, and I changed to

flamelays = floor(FLAMEH/MODRES) + 1
MIDFLAMEPOINT = max(ceiling(flamelays/2.0),2)

I change in latest commit, and this formulation seems to give best answer for midflameheight and layer.

[drnimbusrain]

@zmoon Can we now merge this with Main? Thanks!