Expert: Paul Soderman - 11/20/2007

Question
i am 4th year mechanical engineering student at institute of engineering, pulchowk campus, nepal.
for my 4th year project am planning to built a model wind tunnel. am planning for open circuit wind tunnels.
am planning for circular crossection and a cone for the contractor.i have heard that a 'S' shaped contraction is preferable. but its hard to build and specially for circular crossection.
i have a 4 bladed 21 inch diameter exhaust fan producing a velocity 7m/s can it be used for my wind tunnel?
i need about 40m/s at my testsection.
i am planning for entrance diameter of crossection as 24.5mm and exit diameter i.e. test section diameter as 10mm. am confused with the length of contractor. what is the minimum length with moderate steadyness in air velocity with a conical contractor?
the main problem is with honeycomb and contraction selection.
as for contractor i have only planned with circular crossection as i have read that circular crossection of contractor n test section is better than the poligonal esp square and rectangular. can i try with polygonal? how much do i lose choosing a polygonal crossection?
also for honeycomb, i have to make it myself as no local manufacturers are available in country also because of money. i do not have enough money to buy it. am planning to built it myself. but how is a question i can't answer i found various altenatives but could not declare one as best. hope you understand the question and will answer.
thanks in advance.

Answer
Santosh
Those are very good questions.  I recommend the book Low-Speed Wind Tunnel Testing by W. Rae and A. Pope (John Wiley & Sons, Pub.) for much more detailed answers than I can give here.
An open circuit wind tunnel does not need an S-shaped contraction because you are not drawing air from a duct but from free space.  A simple bell shape (like a trumpet) will do nicely.  Your contraction is small (6:1 by area), usually they are on the order of 20:1, but yours may work if the streamlines follow the wall nicely.  The contraction lengh can be as small as 2.5 times the test section diameter.  Look up the NPL wind tunnel on line.
The fan looks good if the fan to test section area is on the order of 7:1 or 8:1.
Many good test sections are rectangular.  Circular is better as there are no losses in the corners, but as long as your model does not enter that region everything will work in a rectangular test section.  
You can make your own honeycomb out of anything that has a minimum length to cell size of 6-8.  People have even used straws, but they can be lossy if diameter is small.  Check out the above book for loss factors of honeycomb.  You could also try screens instead of honeycomb.
Good luck.
Paul