I'm having trouble with measuring the torque created by a fluid drag force over a surface, so I create a simple model to try to understand what is going on. In the below picture, here are the results I get for my 3 goals (using the global coordinate system) :

Top Face -0.042Nm

Bottom Face 0.013Nm

Both Faces -.0.29NM

I expected both values to be negative, but the bottom face is positive. Why is this? Also, this causes the total torque to be less (because one face is subtracted instead of added).

For the more complicated version, I am trying to measure power = rpm * torque off a shaft, and my torque value is based on two faces which are giving me opposite (+ and -) values, which cancel out to give me much smaller torque than expected. I am wondering if there is a "face normal" in the reported result somehow?

Any help would be greatly appreciated.