Baseball Roughness Effect on Laminar Separation: Post #23

Until recently, we used a Hack Attack 3-wheeled pitching machine to launch baseballs. If you’ve ever used one of these, you know they are brutal on the baseballs. Also, the balls we used were Wilson 1010 or 1030 high-seamed balls. I’ve been careful to point out that those results may be specific to that particular ball. Recently, we’ve noticed that new MLB balls often exhibit laminar boundary layers that separate on the front of the ball, forming a very large wake, as described in Post 22. This only happens if there is no seam on the front of the ball on the side (top or bottom) being examined.

[If you are new to our measurements, you may take a minute to read here about vorticity (the colors in our plots), boundary layer separation and wakes]

The present results, as well as the Post 22 results, are for non-spinning balls launched at 90 mph. The photo below shows the Wilson ball used in the measurements discussed in this post, which is also typical for our earlier measurements (Posts 1-20), where the ball was spinning.

The three wheels on the machine accelerate the ball through friction and touch nearly the entire perimeter of the ball. After a few “throws,” the ball becomes uniformly rough. Note the chunks of the ball drifting away after the throw in the video below.

The boundary layer separation on a ball that has been used in the wheeled machine is consistently on the rear of the ball due to a turbulent boundary layer, as shown below. This turbulent boundary layer separation leads to a smaller wake and less drag.

A second example is shown below. Again, the boundary layer is turbulent prior to separation and remains attached past the middle of the ball.

For a smooth ball, such as the MLB balls that we have recently been using, when no seam is present upstream, the boundary layer separates while it is laminar on the front of the ball as shown below (image from Post 22). The separated shear layer moves away from the ball in a remarkably straight line and results in a very wide wake. The wake generates 95% of the drag on a baseball (the rest coming from skin friction) according to a CFD study at USU. A wider wake means more drag, so, counterintuitively, a smooth ball and laminar flow in the boundary layer can increase drag!

We repeated this measurement with an unused, high-seam, NCAA ball and the results were the same as for the new MLB ball. The condition of the leather is what is important here, not the seam height.

I conclude something that some may think obvious (but needed to be proven, in my opinion). That the boundary layer on a baseball at 90 mph may be laminar or turbulent depending on the smoothness of the leather and the presence or absence of a seam upstream. Clearly, a scuff may cause the flow to be turbulent over more of the ball and clearly any ball that has seen several plays will always have turbulent boundary layers. Since turbulent boundary layer separation results in a smaller wake, this will reduce the drag (skin friction drag on a sphere is not significant in these conditions).

This may answer why a spinning scuffed ball behaves differently in a 2-seam and 4-seam orientation. In the 4-seam orientation, there is always a seam upstream and so boundary layer separation is always turbulent. This is not true in the 2-seam orientation.

This also means that in any game where the ball is not replaced whenever it is damaged, the ball will behave differently after it has hit the ground. It will have less drag. After it’s hit the ground a few times, it will behave predictably, but not necessarily the same as when it was new.

I was surprised to learn today that in the NCAA World Series, scuffed balls are not necessarily replaced as they are in the MLB. If anyone can tell me what the practice is in the MiLB, I’d appreciate it.

Coming Soon:

This raises the question: Does rubbing mud on the ball, as is done before MLB games, change the roughness of the leather? Can this lead to a change in the boundary layer separation location?