2D to 3D rotateFields discontinuities

[jackr25]

I created a 3D UNDEX case based off the case setup from the mappedbuilding3D tutorial case, but I have noticed that there are some sharp discontinuities between the initial conditions and the first timestep after using a rotate fields.

Here you can see a very sharp pressure spike as though the initial charge has been reset and detonated again.

Which does seem to be the case based on the lambda field.

Attached is the full case for 2D and 3D. My rotatefields is mapping lambda, and is using the -uniform option.
RotFieldsFull.zip

I have tried a few different EOS for the reactants, sticking with JWL for the products and Linear Tillotson for the water. I am unsure if this is a bug or user error.