I've gotten the 8-pot runway extension limit into the Ler simulation code.
Along the way I tracked down some bugs that may have been causing problems when I abandoned the project last spring (I think when I was trying to run RIL models).
After much chasing of strange errors and intermediate output, I discovered that most (all?) of the headaches arose from aaply() dropping array dimesions; I fixed these by using the .drop = FALSE argument.