Hey @ceckert ,
Some observations comparing to a similar tutorial we have,
The atmosphere boundary condition would be set up differently with zero gradient to visualize results purely from blade rotation. The following settings that are currently being used would not mean a zero-gradient flow through the domain.
Since you are not using Symmetry boundary condition, you would have to assign all outer faces to this custom boundary condition as follows:
For the direction of blades, I suggest you try to reverse the direction of the blades and check if that gives you the results you want. Everything from a solver and boundary condition perspective looks good other than the previous point I made. Let us know how it works out. 
Best,
Satvik