Maybe some analysis would be useful if done as follows:
Assume steady state flow and an air mass flow rate thru a tube with
representative pressure and temperature conditions at the inlet or outlet.
The steady state flow assumption should be okay due to the air flow inertia
especially at high RPM.
Calculate the Reynolds number. My old textbook "Principles of Heat Transfer"
by Frank Kreith has a graph of friction factor (or pressure drop) as a
function of Reynolds number and relative surface roughness as many similar
textbooks probably do. This is easier and more direct than trying to
calculate the boundary layer thickness--I'm not sure that a protrusion that
does not stick up out of the boundary layer can be neglected off hand. In
any case, starting with a boundary layer thickness calculation is the long
way around.
Then one could parametrically explore a range of flow rates, tube diameters
and lengths, degree of roughness, and inlet conditions to see what amount of
roughness has how much of an effect.
My Kreith graph covers a range of relative roughness from 0.05 (very rough)
to 0.00001 (very smooth). Relative roughness is defined as the average
height of the protrusions divided by the tube diameter.
Lance
--------------------------
-----Original Message-----
From: land-speed-bounces@autox.team.net
[mailto:land-speed-bounces@autox.team.net] On Behalf Of Kirkwood
Sent: Sunday, June 05, 2011 8:20 PM
To: land-speed@autox.team.net
Subject: [Land-speed] Intake Manifold Question(s)
The founder of Go Power did this work while working on his PhD at Stanford
about 60 years ago. He was an avid go-carter and the emphasis in those days
was "port and polish" . . . with emphasis based on "polish". He debunked
that myth in his thesis.
You are correct in that here is a boundary layer. That layer also has
"static cling" LOL, or more technically "stiction", capillary attraction, or
call what you want, with the passage wall. The thickness of this dead-zone
layer is proportional to velocity. But, regardless of velocity, the flow
attached to the wall is essentially zero. So any protrusion in that area is
irrelevant. Only discontinuities sticking up through the boundary layer
influences flow. IIRC, anything with 60 grit (about 1/16") is fine. As Dave
has already indicated. I think some studies showed even a 120RMS had little
influence on flow.
_______________________________________________
Land-speed@autox.team.net
Donate: http://www.team.net/donate.html
Archive: http://www.team.net/archive
Forums: http://www.team.net/forums
|