Take a look at this and let it sink in for a while. (Sorry about the pixellation, I just used the simplest BWheel drawing I could find and enlarged it.) Every Bhaskara wheel design, including Abeling, has these characteristics.
First, do a calculation of the actual turning moments given by the weights and their positions wrt the vertical centerline of the wheel. The horizontal distance (green) from centerline to the center of gravity of the water or mercury in each chamber, times the weight, gives you the turning moment of each chamber. Add them all up for the left side and the right side, and you will find that the turning moments are equal, hence there is no reason for the wheel to turn.
Next, look at the weights relative to the horizontal centerline (blue). There is always more weight on the _bottom_ half than on the top half. That is, the center of gravity of the wheel is always _below_ the center of the wheel. Again, there is no reason for the wheel to turn.
"The easiest person to fool is yourself" -- Richard Feynman