UIUC Low-Speed Airfoil Tests
Wind Tunnel Model Construction Notes (Version 6)
Michael Selig, Rob Deters, and Paul Gush
Department of Aerospace Engineering
University of Illinois at Urbana-Champaign, Urbana, Illinois 61801
Last update 10/9/06 modelNotes-versionChanges.html
- Span = 33 5/8 inch
- Chord = 12 inch
- Distance between tubing centers = 5 inch
- Tolerance = +0, -1/64 inch for span and +/- 1/64 inch chord length
- Tolerance = +/- 1/32 inch for all other dimensions
- 1/4 inch I.D. brass tubing for right side of model
- Two 5/16 inch collars with set screws for right side of model
- 17/32 inch I.D. brass tubing for left side of model
- CNC machined airfoil master template (if requested). Two part sliding template (example):
- The wing rod line should be located in the same place on both ends
of the model. Determine the location of this line by centering the
larger 17/32 inch I.D. tubing between the upper and lower airfoil
surfaces on the aft end of the airfoil. The wing rod line and chord
line, however, should be parallel. If this is not possible, please
- Special care should be taken not to deform the brass tubing.
- The brass tubing should be inserted into the model with the uncut
ends facing outward. Important: (1) Put the brass tubing in last and
glue it in permanently, (2) don't get glue or paint on the inside of
the brass tubing (cover the hole with tape if you have more work to do
on the model), and (3) as a final check see if you can fit the next
smallest size brass tubing into the 1/4 inch and 17/32 inch
I.D. tubing. If you drop the tubing supplied, chances are that the
ends will get bent and the test above will fail. So buy or ask for new
tubing if those supplied are dropped. In case you're wondering, we do
fit the 17/32 inch and 1/4 inch O.D. size rods into the model in the
lab before installing the model in the wind tunnel. Also, please
completely install the collars (glue them in) with the set screw
extending completely through the 1/4 inch I.D. tubing. If you follow
the steps above, it will eliminate 99% of the problems that we have
with wind tunnel models, and we will end up taking more data.
- The 9/32 inch I.D short tubes (sometimes supplied, see collar
details in drawing above) should be cut and then slipped over and
glued onto the 1/4 inch I.D brass tubing at the location on the
collars. The collars should then be slipped over these short tubes,
and the set screws extended through holes in both brass tubes so that
they can lock onto a 1/4 inch wing rod. On this rod, we have machined
a flat for the set screw.
- The set screws must be accessible from the bottom of the airfoil. Large cavities
around the collar/set screw should be avoided.
- It is also preferred to have the collars mounted just inside the right end of the model
to minimize aerodynamic disturbances. (See figure).
- The surface finish must be smooth.
- Try to avoid bumps that form along glue seams, since they can cause early and
unwanted transition to turbulent flow.
- The trailing edges should be as sharp as designed. If this is not
possible with the given construction method, XFOIL can be used to
modify the coordinates in the trailing edge region for added
thickness. Any questions on this subject should be sent to Prof Michael Selig.
- No wing twist should be present.
- The first 70% of the upper surface is the most important part of the airfoil, and
therefore it should receive the most attention.
- After that, spend any remaining time on the leading edge region.
- The center of the line on airfoil plots represents the exact coordinates.
- Color: We prefer black models. If the model is going to be
covered with Monokote or some other similar covering, we especially
prefer black. The black covering allows use to readily do surface oil
flow visualization should we want to do that.
- If you have any questions, please call Prof Michael Selig or the
- We prefer the models to be as light as possible, but please
remember that the forces on the models can be substantial at
higher speeds. We have had flutter problems with models that are
particularly heavy, so please keep this in mind when selecting
materials for manufacture. Be careful not to sacrifice strength
and stiffness in order to achieve light weight though.
- Please use a spar. At the higher Reynolds numbers, models without
spars become quite bowed so the potential is there for
- In the past we would sometimes allow the rod spacing to be
decreased to 4 1/2 inches. Due to limitations on our set-up, we
now prefer it if you could keep the spacing to 5 inches. In some
instances this will mean that the rods cannot be aligned with the
chord line anymore. If that is the case, please contact us before
installing the tubes.
- If it is at all possible, please use cores made from spyder
foam. Some preliminary studies indicate that the results are much
more accurate than cores cut from common white, blue, or gray
foams. During the process of cutting the foam, its surface is
melted first on one side and then on the other as the cutting wire
passes around the core. We think this process produces stresses
that get locked into the core and cause it to develop a chordwise
warp. With spyder foam's higher density and different structural
properties, it seems to be much less susceptible to this type of
Wind Tunnel Model Accuracy
The accuracy of the wind tunnel models cannot be over-emphasized. All
airfoils are round in the front and sharp in the back; what separates
the good from the bad is what happens in between. The difference in
shape between a "good" airfoil and a "better" airfoil is not
large. However, since the difference in drag can be quite significant,
it makes it worthwhile to be as accurate as possible. Accuracy is
important from a different standpoint as well. Extremely accurate
lift and drag measurements will be taken in a wind tunnel with
excellent flow quality, and each airfoil will require approximately 16
hours of tunnel time to span a range of Reynolds numbers and angles of
attack. If the models are of comparable quality, then the measurements
are meaningful; otherwise, they are not. To aid in building the most
accurate model possible, you will be sent if requested CNC machined
female templates of the airfoil contour.
Suggestions for Being Accurate
The most important step involves making the rib templates (or foam
core templates). This step should not be rushed. After the model is
assembled and ready to be sanded, check the profile with the female
templates. Put the female template on the model and shine a light from
the other side. If light shines through between the model and the
template, the model should be sanded until the template and the model
agree. A separate template should be used to check the model leading
edge in the same way. If you are aware of inaccuracies with your
model, rework the model until it is accurate.
Shipping of Models
- It is highly recommended that the shipping box be made out of
wood. We have never received damage models shipped in wooden
containers. We have received many damaged models shipped in
cardboard containers. Most of these were returned to the builder
for repairs. Tunnel time is pretty special, so we generally don't
have time to make fixes while taking data during a tunnel slot.
We usually have only two tunnel slots in a given year.
- Cardboard shipping boxes can be found at bicycle shops and moving
companies, eg U-Haul, etc.
- If you do use a cardboard box, a "light" cardboard box is easiest
for shippers to carry and stack. Also, lighter boxes are stacked
on top of heavier boxes (e.g. wooden boxes) in U.S. mail or UPS
trucks, and stand less chance of getting crushed (vs "heavy"
- Write "FRAGILE" on the box.
- The model should be cushioned inside of the box with foam rubber,
newspaper, Styrofoam "peanuts," or similar material to absorb
- The models should not be securely strapped to the box. Models
strapped to the box will be cracked if the box cracks from being
dropped or thrown around.
- For best packaging, place airfoil models in foam cores, with
packaging material around the foam cores. Do not allow tape to
stick to the model, especially if the paint is still drying(!).
- Plenty of tough packaging tape should be placed over the seams of
the box and at the corners.
Models should be mailed to the following address:
Prof Michael Selig
Dept. of Aerospace Engineering
University of Illinois at Urbana-Champaign
306 Talbot Laboratory, 104 S. Wright St.
Urbana, IL 61801-2935
Finally, if you build a model, we're going to send you a free copy of
the next airfoil book. Let us know if your address changes.
[UIUC Applied Aerodynamics Group]