Research Blog 2
Research Overview
For our research, we are testing the effects of varying trailer panel attachment angles on a scale model tractor trailer in a wind tunnel. Using a force meter, the force of wind drag is measured and the mean force of each angle is compared to the other populations to find the ideal trailer panel angle of attachment.
Trailer Panel Cut Angle
A self-derived equation is given. The trailer panels are essentially trapezoids that are glued together at their non-parallel edges. The angles of the trapezoid panels that meet at the corner of the truck trailer are calculated through the equation to the left. The x variable is simply the angle inward. For example, when you want to find the cut out angle of the thirty degree panel, set x=30, and your output will be 63.4349 degrees. To cut out the trapezoid, the smaller angle is 63.43 degrees and the larger angle is 116.57 degrees.
Experimental Setup
The first step of acquiring the materials necessary for pre-trials was ordering the scale model truck. Once the model arrived, the trailer panels were constructed by calculating the dimensions and angle necessary to create trailer panels at the needed angles. The formula used for calculating this angle is shown above. A hole was then drilled into the wind tunnel using a drill driver and a smart pulley was placed in the wind tunnel just behind the hole as shown in picture one below. An ample length of string was measured, attached to the force sensor, and threaded around the smart pulley to be attached to the front of the truck. As a result of the wind tunnel's velocity being too slow to visibly move the truck, wood blocks were placed under the tunnel to place it at a 1.5 degree angle, allowing the truck to roll so the force of drag can be more easily measured. The final steps of setting up the experiment included securing the force sensor to the table, creating a randomized order of trials, attaching the anemometer to the rear end of the wind tunnel, and using a power strip to allow both fans to turn on simultaneously.
Force Sensor
The force sensor is attached to the table using duct tape, and the black cord runs from the force sensor to the lab quest to record the data. The string attached to the hook also runs up through the hole in the tunnel and around the pulley.
Panels 1, 2, and 3
This picture shows the three panel assemblies. Panel 1 is the zero degree panel on the left. Panel 2 is the 15 degree panel in the middle. Panel 3 is the 30 degree panel on the right.
Anemometer
Here, the anemometer was placed at the end of the wind tunnel where the air exits. The anemometer measures the air velocity and ensures that the fans are running consistently. Most trials gave an average wind velocity of 5.7 m/s.
Problems Experienced
One problem which occurred during pre-trial experimentation was possible inconsistency with the force meter's measuring. This was due to the method used to secure the force meter, which consisted of duct taping it to the table as shown above in picture one. While this produced consistent data for the most part, over time it is possible that the force meter became progressively looser, which could result in inaccurately lower force values. Although this method has not been replaced yet, there is a plan to do so for data-trials. A hole will be drilled into a wooden block to secure it to the force meter using a screw. This block will then be clamped to the table, resulting in no force meter movement.
There are three samples taken during trials: the zero, fifteen, and thirty degree panel assemblies as shown in picture two above. The amount of trials taken for each sample was not the same for each day. For example, on the fourth day ten samples were taken of the zero degree panel, twelve of the thirty degree panel, but only two trials for the twenty degree panel. As a result of variation in the force sensor measurements each day, some days would record a higher overall force across all three samples. This 'noise' can me minimized by doing the same amount of trials across each sample for each day, but it becomes a problem when the amount of each sample trials vary. The last day had a smaller average force reading. This caused sample two, the fifteen degree panel, to have a higher force mean as less trials were taken on the day when the force sensor was giving low values. To solve this problem, a new, randomized order of trials have been made for data trials with equal amounts of each angle trial for each day.
A significant source for variation and error came from the wind velocity of the force sensor as it varies slightly from trial to trial. To solve this problem, a timer was used to start collecting data exactly ten seconds after turning on the fans, but the velocity still appeared to vary somewhat. To account for this, an anemometer was attached to the back fan of the wind tunnel as shown above in picture three. The velocity was the recorded for every following trial in the observations with the intention of providing an explanation for any extreme values.
Although these problems did slow down pre-trials, they were not significant enough to result in a necessity to change the problem statement.
Pre-Trial Results
Despite the minuscule problems which occurred, the results of pre-trials were highly favorable. The problems described above facilitated improvement in the setup and procedure which will allow for more consistent testing in data-trials. Ninety trials were completed over the course of the four days of pre-trials with thirty trials of each attachment angle. An initial ANOVA was calculated from the ninety data points collected to determine the progress of the results. The test yielded a significant p-value of 0.004, less than the alpha level of 0.05. This will not be interpreted as significant results as of now, but it does show promise to the potential of the experiment as it currently seems to support the hypothesis with the fifteen degree angle possessing the lowest drag force and the zero degree angle possessing the highest drag force.