Originally Posted by Crossplot
IThe path past a cylinder is a symetrcal cursive "e".
The primary acceleration is normal to this path.
I am not sure what you mean here, maybe you could draw it.
There is no disagreement that centripetal acceleration takes place and is proportional to the gradient of the pressure in the direction normal to the path. However, please re-read this quote:
air flow near the aerofoil follows the geometrical
shape of the upper surface generating a pressure
gradient and acceleration, both perpendicular to
the streamlines and directed to the centre of the
This states that the acceleration is normal to the streamline, which is mostly not true. Consider the a fluid particle very close to the stagnation point. Its velocity is almost zero, so there is very little centripetal acceleration, but there is a large tangential acceleration. As you point out, flow over a cylinder illustrates this nicely:
Also as you point out, the acceleration of the flow at the surface is only normal to the streamline at the top of the cylinder (theta=pi/2).