As we showed in post 23, a ball with rough leather tends to produce a turbulent flow which delays the boundary layer separation. Roughness, in fluid dynamics, is difficult to nail down. We typically use “relative roughness” which is a ratio of the mean height of the roughness to (in this case) the ball diameter. In other words, relative roughness is the ratio of the average size of the bumps in the material to the ball diameter; a value of zero indicates perfect smoothness and a 1 would be a ball with lumps as big as the ball. Accurately identifying the relative roughness for materials, like the leather on a baseball say, is difficult, but with proper tools (which we don’t have) it can be accomplished.
To better understand the measurements being presented, have a look at our post on Particle Image Velocimetry and one on our current setup. Additionally, the National Committee for Fluid Mechanics has a website with several films explaining some fundamentals of fluid behavior. The videos, which you can find here, are old but do a good job.
I believe one of the largest areas of confusion for individuals learning about fluid dynamics or those interested in drag over a sphere is how roughness affects drag. For flow over a flat plate roughness increases drag. This is intuitive as we can experience this tactilely as we feel how slippery a polished rock can be compared to a rock that hasn’t yet tumbled down any length of stream and still has all its corners, bumps and roughness. However, the increase of drag with roughness is always true for bodies where boundary layer separation does not occur such as flow over a flat plate or a streamlined body (the classic teardrop shape).
This is where the confusion begins. Depending on the roughness and velocity this increase in drag does not occur in bluff bodies (i.e. shapes where flow separation occurs). In fact, the drag will decrease at a certain velocity for a certain roughness. Thrown baseballs are always bluff bodies. This is significant because it means that, at typical baseball speeds (60-90mph), a baseball will have decreased drag with roughness. Another clear example of this is the golf ball. The dimples of a golf ball increase the roughness of the ball and decrease drag.
Why does a rougher ball experience less drag? Barton Smith (no relation!) does a great job in POST #20 explaining some of the nuances of the so-called “drag crisis” and why that is so important for baseballs. Long story short, the rougher a sphere, the earlier boundary layer transition to turbulence happens and the earlier that drag crisis occurs. Our data so far have shown that new, unused, un-mudded major league balls have laminar flows in most (but not all) cases unless there is a seam upstream to trip the flow to turbulent, even at speeds of 90mph.
The question we are trying to answer now is how Lena Blackburne Baseball Rubbing Mud affects the roughness of a baseball and whether that effect is large enough to transition the flow from high-drag laminar to low-drag turbulent. This is significant because Lena Blackburne Baseball Rubbing Mud is used in every Major and Minor League game on every ball. There is an interesting history for why this is and you can read it on their website at baseballrubbingmud.com . You can also get some mud of your own.
Now, contrary to my analogy above, a good grip does not necessarily mean increased roughness. A rough piece of ice may be harder to grip than a smooth piece of leather but the ice is still rougher. While rubbing mud may make the balls easier to grip for pitchers, it may, at the same time, fill in the pores on the leather and make it smoother.
For consistency, we took a previously un-mudded ball for which we had found consistent and clear laminar boundary layer separation and rubbed the mud on. Mudding balls is far from consistent. The umps are in charge of overseeing the mudding of the balls and often do it themselves but some of the teams’ clubhouse attendants do the actual mudding. There is no prescribed process that the umps or clubhouse attendant must use and each person does it differently. There is a prescribed target for how dark the ball should become but ultimately the ump decides what is acceptable. Typically, about 12 dozen balls are prepared for each game and each must be mudded beforehand. This allows for a lot of variation between individuals who are applying the mud and between the balls that one person prepares.
We mudded as close to the prescribed target as we could and measured several of the dimensions of the ball before we began PIV testing. Anecdotally, the ball was definitely easier to grip but that does not mean it necessarily was rougher and I noticed that my hands, after applying the mud and washing my hands, felt as dry as they do when using athletic chalk.
The mud increased the weight of the ball from 143.46g to 143.91g and increased the average seam height by 15%. All other dimensions stayed the same. The increase in seam height is interesting and deserves further attention but we’ll save that for later. Below is shown an un-mudded ball thrown at 75 mph in a “no-seam” orientation (where no seams are in front of the ball in the interrogation plane). The velocity vectors are presented over contours of vorticity (blue meaning clockwise rotation of air and red meaning counter-clockwise; the thick blue line shows a sudden change in flow direction). The distinct blue line shows a sudden change and velocity and this linear behavior represents a laminar boundary layer separation.
The same ball is shown below after being mudded then thrown in the same manner.
The results are perhaps surprising. The mud didn’t change the roughness to an extent large enough to affect the boundary separation layer at this speed. Note also that we tested various levels of mudding and all have still shown this laminar boundary layer separation behavior.
These are preliminary results but I expect them to represent what we’ll find as we do more testing.