Abrasive flow finishing

[jb24 at chrysler.com]

>>>I was thinking about the "random bumps reduce drag" effect. Perhaps
there's a
difference between drag on an object in free space (planes, golf balls)
and mass
flow through a tube. It seems weird, but it might be true. Supposedly
when you
have laminar flow over a surface the flow speed right at the surface is
near 0,
increasing (as distance squared) to whatever max value "far" from the
surface.
I *think* that this nice, smooth flow over the surface also provides
maximum
drag, kind of like being perfectly wrapped in sandpaper. The bumps
cause turbulent
flow, which results in little pockets of air where the "sandpaper"
isn't touching
you... and thus less drag.  Where I think the difference between free
space and
a tube lies is that, in a tube, the turbulent flow near a dimpled
surface has to
intrude into the limited flow area of the tube itself, where it
interferes with
and reduces the smooth flow that was there. Clearly more of a problem
in smaller
tubes. Anyway that's my thought. Someone with a better understanding of
aerodynamics please feel free to correct and clarify this!<<<

The explanantion you are looking for is called Reynolds number.
Reynolds number is a dimensionless parameter used to express the
relationship between free-stream velocity and the size of the surface
(length) in the flow.  Skin friction changes with Reynolds number,
very-nonlinearly.  Which is why a very unaerodynamic bumblebee can fly
with little relative drag.  Anyhow, internal flow is much different
than external flow, mostly because of the length of the surface being
interacted with (and thus greater Reynolds number).  All internal flow
will eventually become turbulent, surface finish will just change
when.   Golf balls kick themselves up into a higher Reynolds number,
which also happens to have lower drag.  However, the same dimples on a
volleyball (if it was flying at the same speed as a golf ball) or
landspeed record vehicle would increase drag.  It's complicated, but
really what I am saying is flow losses are different for different
cases.  The difference between smooth and bumpy surfaces is how the
flow turns corners.  The same tiny vortices that allow a smaller wake
(and drag) on a golf ball disrupt flow going through a bend in a pipe
(effectively lowering the cross section and increasing the surface to
volume ratio, and therefore greater flow losses due to friction).
Confused yet?  Turbulence is good or bad, depending.  However,
turbulence is almost always good for combustion.

John Bucknell is jb24 at chrysler.com