|
Test Methods
Nestled in the small university town of College
Station , Texas , the Oran W. Nicks Low Speed
Wind Tunnel has an undeniable reputation as the
place to go for cycling related aerodynamic
research. With a test section that
measures seven feet wide by ten feet tall,
adequate flow qualities, and dedicated cycling
fixtures, this facility is convenient for
industry outsiders while still providing quality
data. Tunnel director, Jorge Martinez, has
claimed that the wind tunnel balance will
provide force data that is accurate to +/- 0.05
lbs. While not as accurate as is
desirable, it is still good enough to identify
trends and shed some light on a variety of
cycling specific aerodynamic questions.
All testing was done using a wheel attachment
fixture developed by John Cobb and the tunnel
staff. Two streamlined struts allow a
wheel and fork combination to be attached to the
tunnel force measurement system (also known as a
balance). This setup is as realistic as
one can get while still minimizing the variables
in the experiment. It could have been
possible to mount an entire bicycle, complete
with brakes, etc., but these additional
components add to the noise and repeatability of
the experiment. For simplicity and to
minimize experimental variables, a single
rotating Zipp 404 clincher wheel (with a Kevlar
700x20c Continental Ultra 200 tire inflated to
110 psi) was used as the test platform.
This type of setup allows for a comparative
evaluation while still respecting the most
important of “real world” concerns.
Figure 2 . Experimental setup at Texas A&M
low speed wind tunnel.
A small electric motor spins an inch and a half
steel roller that, under a light spring force,
ultimately rotates the wheel. Wheel
rotation is matched to the wind tunnel speed
using an infinitely adjustable speed controller
and a speed sensing system that is very similar
to modern day cyclecomputers.
Figure 3 . Close-up of wheel strut and
motor.
In order to insure that all forks were tested
identically despite differences in steer tube
length and diameter, a cover was placed over all
steer tubes. A round, ten inch long piece
of 1 ¼” inner diameter PVC pipe was roughened
and installed prior to the forks being tested.
Figure 4 . Steer tube fairing and
measuring the head tube angle.
One of the benefits of testing bike parts in a
wind tunnel is the ability to simulate different
wind conditions independent of other variables.
A word that may be familiar to some is the term
“yaw” which is a carryover from the aerospace
industry. Yaw is the angle the bike
velocity vector makes with the relative wind
vector. For example, if there is no wind
and the rider is traveling 25 mph, the yaw angle
would be zero degrees. However, if the
same rider experienced a pure crosswind of nine
mph the yaw angle would be 20°.
Figure 5 . A top view of the
experimental setup and definition of the
wind-axis terms (drag, side force) as well as
the body-axis term (axial force).
Figure 6 . Side view of the fork test
configuration and head tube angle.
In order to evaluate overall aerodynamic
performance, it is essential to measure the
sample’s properties over a representative range
of yaw angles. It was assumed that the
appropriate range of yaw values for cycling is
0° to 20°. There are certainly occasions
in which the yaw angle exceeds 20 degrees, but
it can be argued that these situations are rare.
Another terminology issue that needs to be
discussed is that of lift, drag, side force, and
axial force. The force of interest for
determining overall cycling performance is the
axial force, or the force that opposes the
direction of travel. Drag is the force
that acts in a parallel direction to the wind
vector, and is only applicable when referring to
the wind-axis coordinate system common in the
aerospace industry. Lift is the force that
acts in a direction perpendicular to drag
(usually “up” when thinking about an airplane
wing).
In cycling, lift is not important, since forces
in that direction do not affect performance.
Of importance, though, is the side force, or the
force that tends to make the bike steer to the
left or right. If anyone has ridden aero
wheels on a windy day, they have experienced the
adverse affects of the side force component; too
much side force and the bike will become
difficult to control.
It is the ratio of side force to drag that helps
determine the overall aerodynamic efficiency of
wheels, forks, or any component on a bicycle.
This ratio is the fundamental principle behind
the often referred to “sail effect” - it is
theoretically possible to have a propulsive
resultant force which allows the bike part to
“sail upwind”. In order for this to occur,
though, the side force to drag ratio must be
greater than one, and the flow must remain
attached for the extreme high yaw angles.
Simply put, an overall higher ratio of side
force to drag means that the axial force (the
force opposing forward motion) will be lower –
and a lower axial force means higher speed for a
given power output. If one wants the gory
math, it can be shown that the axial force is
derived as follows:
Faxial = Fside*sin(yaw°)+Fdrag*cos(yaw°)
Equation 1 . Coordinate transformation
from wind axis to body axis.
The discussion of axial force and drag may seem
like a debate in semantics, but it should be
re-emphasized that the variable of interest when
it comes to determining overall cycling
aerodynamic performance is the axial force.
Drag is only directly applicable when the yaw
angle is equal to zero. For this reason,
fork data will be reported using axial force and
not the more familiar quantity of drag.
Results & Discussion
A little over six years of fork development has
netted the consumer approximately 0.15 lb less
axial force at 30 mph, which corresponds to a
wattage savings of 9 watts or around 25 seconds
for a flat 40 k time trial.
Figure 7 . Average axial force values over
the 0° to 20° yaw angle range (measurement error
+/- 0.05 lbs).
It should be clear that all three aero forks
offer a measurable improvement over the
traditionally shaped fork. The Reynolds
and Oval forks perform nearly identically, with
the True Temper fork performing just within the
uncertainty of the measurement method. The
above results can also be presented slightly
differently in order to better differentiate the
aero fork products.
When the data is summarized for crosswind (yaw
of 10° to 20°) and calm conditions (yaw of
0° to 10°), the Reynolds and Oval products are
still extremely close; however, the difference
at larger yaw angles between these forks and the
True Temper fork becomes slightly more apparent.
Figure 8 . Axial force averages for
calm and crosswind conditions (measurement error
+/- 0.05 lbs).
Yet another way to look at the aero data is
to determine the efficiency of the forks in
reducing the axial force at yaw. The
higher the side force to drag ratio, the more
efficient the fork. Again, it can be seen
that all of the aero forks offer an improvement
over the traditional fork.
Figure 9 . Average side force to drag
ratio for 0° to 20° yaw angles (error +/- 0.07).
The same trends in tunnel acquired side force to
drag ratios can be seen in the geometric area
ratios of the forks. This observation is
consistent with aerodynamic theory since it has
been shown that drag is proportional to frontal
area and side force is proportional to projected
side area. In the chart below, the
ordering of geometrical area ratios is seen to
be an accurate predictor of tunnel performance.
Figure 10 . Geometric area ratios of the forks
correlate well with wind tunnel results (+/-
0.1).
Of particular interest in comparing these two
plots is that the wind tunnel results show a
smaller difference between forks than do the
geometrical results. This is almost
certainly due to the dominating influence of the
Zipp 404 wheel on the overall aero performance
of the combination. However, the fork
still does play a role as can be clearly seen in
comparing the Kestrel data with the aero fork
data.
How can these results be translated into
information that is useful to mortals unable to
ride consistently at 30 mph? Below is a
table summarizing time savings and wattage
savings (relative to the traditional Kestrel EMS
fork) at different speeds. It should be
noted that the greater time savings at slower
speeds are a result of the aero benefits
occurring over a longer duration of time.
Table 2 . Typical wattage and time savings
for several different average speeds on a flat
time trial course (note: minor variations are
due to rounding).
The data above appears to be consistent with
aerodynamic theory. The aero forks have
more streamlined blade shapes than the
traditionally shaped fork. The result is a
larger side force to drag ratio and, therefore,
a lower axial force.
Summary
If one takes their racing seriously, attention
to detail is important. Cyclists need to
train their body, their position, and then they
need to start thinking about what equipment to
use. In the absence of wind tunnel data, a
good approach to determining the aerodynamic
performance of a fork is by comparing area
ratios and/or blade length to width ratios.
An aero fork is a bit further down on the list
of equipment that one should address, but if a
championship is on the line, the fork under your
nose matters. Aero forks work.
<insert sell sheet for full data set here>
|
