Expert: Paul Soderman - 5/19/2011

Question
Paul, How are you doing today? I have actually two questions for you. First, a little background. I'm an avid RC model pilot and have been for many years. I occasionally get a wild hair and design my own aircraft. I do have a technical background but not in aerodynamics. I have read several books on RC aircraft and how to design them. One thing I have never understood is what a Reynolds Number is. Can you explain it to me?
My second question is that just for my own head, so to speak, I would like to take an on-line engineering course in aeronautics. Do you know of any schools offering such a course? Thanks! Dave Bryer

Answer
Dave
Technically, Reynolds number is a dimensionless parameter that is the ratio of the intertia force (drag) to the viscous force in a fluid.  But in practice, Reynolds number is used by engineers to predict how the flow, especially the boundary layer will act on a body.  Consider a cylinder.  At very low Reynolds number (Stokes range) the flow resembles a perfect fluid and no separation is evident.  At a somewhat higher Reynolds number, a laminar boundary layer develops that separates at a certain angle on the cylinder.  At the next Reynolds number range the laminar boundary layer separates but rejoins the cylinder as a turbulent boundary layer.  The separation point moves aft with increasing Reynolds number.  Similar things happen on an airfoil.  So to get a desired aerodynamic behavior we have to achieve the appropriate Reynolds number.  This was a perpetual problem in our wind tunnel research because if we used small models to simulate flow over large aircraft, it was difficult to achieve a high enough Reynolds number to get the desired drag or stall condition, for example, that would occur on the real aircraft.  So engineers used various tricks to simulate a high Reynolds number flow by tripping the boundary layer with sand paper for example.  This is one reason the NASA 40x80 Ft Wind Tunnel was built to test very large models and use those results to correct data from small wind tunnels.  With modern computational tools, large model testing is not quite as necessary, but can still be a great way to benchmark a prediction code.

For RC models this means you can't rely on full-scale aircraft data to predict what will happen to the model.  Though in many cases the differences may be acceptable.  It all depends on what aerodynamic problem you are trying to solve.

I am not familiar with on-line engineering courses, but it would be easy enough to find out by visiting school websites.  Here are a couple to try:  http://www.aa.washington.edu/  and   http://www.erau.edu/
Paul