Team Victory Lap chose to address the need of implementing an efficient, user-friendly, and accurate operating system for the subsonic wind tunnel at the University of Nevada, Reno. Faculty members, undergraduate and graduate students, will be able to utilize these upgrades in order to conduct wind tunnel experiments in much greater depth. The first issue addressed was how the wind is supplied by a traditional centrifugal blower without the use of a honeycomb structure. This structure straightens out the supplied airflow to provide more accurate and consistent testing conditions. Furthermore, the system was not equipped with a computer data acquisition program, making data collection a long, tedious, and manual process. The team tackled this challenge by developing a continuous data acquisition system, integrated with the existing electrical instrumentation, to measure and record all of the signal forces simultaneously. The last component of the project was to integrate an airflow visualization system, useful when demonstrating experiments for a group, or, supplementing numerically recorded aerodynamic data with corresponding visual representations of airflow. Team Victory Laps additions to the wind tunnel will provide a lasting and valuable upgrade packed with key resources for engineers developing a greater understanding of aerodynamic properties associated with fluid mechanics.
Proof of Concept:
To prove that a party style fog machine could be used for smoke streamline generation, a fog machine was purchased and modified. The fog machine, produces 20,000 cfm of fog but this fog is dispersed in thick clouds rather than fine lines. Using Aluminum duct material and metal tape, the team fabricated an attachment to the smoke generator output which condensed the smoke and funneled it through tubing to a steel rod with pin holes drilled into it. This design can be seen in figures 2 and 3 below. Running the generator produced the streamlines as desired but caused potentially dangerous heat in the contraction cone. To reduce the potential for skin burn or a fire hazard, the team insulated the ducting and contraction cone. Once it was determined that the generator was safe to run, the team was able to insert the streamline apparatus in the wind tunnel and test it. Faint but visible streamlines about six inches long were observed. It was also observed however that a considerable amount of fog was condensating before it reached the pin holes. It was decided that condensation reduction would help improve the quality of the streamlines and an effort to do this would be implemented into the final design.
After conducting the proof of concept, the team was able to move forward with the final design. The main issues that the team addressed in the final design were condensation build up and the high temperatures of the contraction cone and dispersal system. It was decided to still use the fog machine that was used in the proof of concept but make changes to the dispersal system design. The final design completely eliminated the contraction cone and instead incorporated a settling chamber and vacuum pump as can be seen in figure 4 below. The team chose to add a settling chamber to the design in order address the condensation and heat issues that arose during the proof of concept. The settling chamber allows the smoke to build up and become thick while also allowing it to cool down. This helped reduce the amount of condensation dramatically and also helped produce much thicker smoke. The smoke is then sucked out of the settling chamber using a vacuum pump that is installed on top of the settling chamber. From the vacuum pump, smoke is pumped through a plastic corrugated tube, then a small contraction cone made of pipe fittings, a vinyl tube, and finally into an aluminum smoke inlet tube. All of these components of the smoker assembly can be seen in figure 4. The vacuum pump provides enough pumping power to pull the smoke from the settling chamber and push it up to the inlet at speeds that can match the wind tunnel speed. The inlet tube is what is inserted into the wind tunnel and is where the final smoke streamline flows out of. The tube is bent at a 90 degree angle in order to direct the smoke in the direction of air flow inside the tunnel. In order to get the best looking streamline, the flow of smoke needed to be laminar flowing out of the pipe. In order to make sure the flow of smoke is laminar after the pipe bend, it is necessary to have the length of the pipe after the bend be 10x the diameter of the pipe. The pipe used in this design is ½ inch so a five-inch pipe length is necessary after the bend in order to make the flow laminar. The entire smoker assembly is installed directly below the wind tunnel as shown in figure 4. The inlet tube is inserted on the underside of the tunnel through a pre-drilled hole.
For the smoke generator, the team’s goal was to have one well defined 3ft. smoke stream that maintained its visibility across objects being tested. Prior to these tests, the honeycombs were installed. This is was done because a more laminar airflow allows for the smoke streams to maintain their integrity for a longer duration. Because the smoke streams are purely a visual effect, the results were based on our visual determination of smoke thickness. The test went as follows:
- Turn on fog machine
- Pump smoke into settling chamber
- Turn pump on at low setting
- View results inside the tunnel
The smoke streams during this test were visible for only a few inches after the inlet dispersion tube. This led the team to several resolutions These were as follows:
- Spray paint the inside of the test section black
- Provide black lights to the test section to illuminate the smoke
- Short bursts of smoke were shown to improve smoke density
After these three alterations were made, a prominent smoke stream was created that could maintain its integrity for well over 3 ft. These results can be seen in figure 6 which shows the smoke stream across an air foil.
Proof of Concept:
In order to prove that a honeycomb mesh could help create laminar airflow inside the wind tunnel, the team used a smaller honeycomb that could be easily installed in the wind tunnel test section. The mesh used was a piece of scrap outside of the machine shop in the Palmer Engineering building. The mesh was cut to size using a bandsaw for a precise, tight fit. The dimensions of the honeycomb tested in the proof of concept were 24 inch x 24 inch x 4 inch (LxWxT) with a cell size of 0.375 inches. 1 inch x 1 inch aluminum L-brackets were cut to six inch lengths and glued around the inside of the wind tunnel on the front and back of the mesh to prevent it from moving while air is flowing. Ribbon was then tied to a platform inside of the test section and the wind tunnel was ran on a low, medium, and high speed with and without the mesh installed to compare the visualization of airflow. The ribbon was observed to be much more stable once the honeycomb was installed, thus the concept was proven. From here the team was able to conclude that adding a second honeycomb right before the contraction section that meets the specifications should cause laminar airflow.
Based on a study the team found, in order to have laminar airflow, the honeycomb mesh needed to have 150 cells across the diameter of the section where it is installed. The area where the mesh will be installed is 4×4 ft. so to have 150 cells, each cell needed to be 0.32 inches long with an average cell diameter of 0.25 inches. Figure 7 below details the different dimensions of a honeycomb that need to be specified in it’s design. To have laminar airflow, the cell length should be six to eight times longer than the diameter of the cell, or 1.5 to 2 inches. A scrap honeycomb mesh was used in the proof of concept and was 2×2 feet and 4 inches thick with 0.375 inch cells. This did not meet the requirement of 150 cells across the diameter. Because the team designed for functionality, this design was modified with the addition of a 4×4 foot mesh installed at the contraction section that did meet the teams calculated specifications. The aluminum that the honeycomb was made of proved to be strong enough to withstand the 60 mph winds inside the test section. The aluminum honeycomb that was installed in the 4×4 section also easily withstood the 60 mile per hour wind force based on the team’s calculations. To insure safety and security, the final design also included adding brackets that held all meshes in place.
Designing for safety was a concern, so the smaller mesh needed to be secured so that it would not slide into the test section and hit the mounting plate potentially causing harm to the system or the user. Based on calculations, the 2×2 foot mesh needs to withstand a force of 6.27 pounds and the 4×4 foot section needs to withstand a force of 1.57 pounds. The team decided to use 1/4-inch long wood screws to attach steel brackets around the edge of the tunnel to prevent the meshes from moving. A total of 8 screws, 4 bolts, 4 nuts and 4 L- brackets were used to secure the small mesh. The large mesh required more effort to secure in its final location. A total of 32 screws, 4 bolts, 4 nuts and 16 L- brackets were used for this part of construction. Each wood screw used can hold a shear load of 415 pounds giving our design a factor of safety of over 2000. The new 4×4 foot mesh was designed by the team and then manufactured by a professional manufacturing company. The mesh arrived in a compressed state which was not easily expanded without proper tools. It was decided that the mesh be returned to be expanded professionally to ensure no harm. The mesh was shipped back in two 4×4 foot sheets that were 1 inch thick which, when sandwiched together, were perfect compared to our design specifications. The mesh used in the smaller section was donated by the machine shop in the Palmer Engineering building. Both installed honeycombs can be seen in figures 8 and 9 below.
For the honeycomb mesh, the team’s goal was to create laminar airflow in the test section. To test whether this was achieved, the team had to first remove the small honeycomb that was previously installed for the proof of concept. This allowed the team to test the wind tunnel under its initial turbulent conditions. The test consisted of the following:
- Attach a string to the pyramidal force balance with a ribbon at its end
- This was done as opposed to the ribbon tests done for the proof of concept because those ribbons tended to twist which hampered results
- Turn on video camera to record results
- Slowly turn up wind speed and look for turbulence in ribbon
After this test was done, the honeycombs were installed and the same tests were conducted. Reviewing the videos of each showed that the string remained much more stable once the honeycombs were installed. The team is confident that the honeycombs had created a more laminar airflow and no further adjustments were made. The results of this test can be seen in videos 1A and 1B below.
Video 1A: Airflow characteristics of wind tunnel without rigid honeycomb mesh installed
Video 1B: Airflow visualization with honeycomb mesh installed in wind tunnel
Proof of Concept:
The proof of concept designed for the computer data acquisition system implemented a NI USB-6009 DAQ obtained from graduate research assistant Rachel Green at the University of Nevada, Reno. Team Victory Lap tested whether or not the voltage outputs of the wind tunnel are measurable by the DAQ. The readings for roll, pitch, yaw, drag, lift, and side force are all measured in millivolts, which are too small for an accurate reading. To address this, the team conducted a preliminary analysis to see if the DAQ alone has the capability to measure such small input voltages. The results proved negative, so the team constructed a simple circuit on a breadboard with a non-inverting amplifier that would be wired into the system to amplify the voltage signal received from the wind tunnel apparatus. The design of this breadboard is shown in Figure 10 below.
In addition to this, another simple circuit was designed using a transimpedance amplifier to convert a current signal input into a voltage signal output that could be used to measure the wind speed within the wind tunnel. The model DAQ obtained through donation for proof of concept testing and final implementation has an accuracy range of ± 7mV. This accuracy range, when receiving the amplified signal, will be sufficient for consistent and adequate testing results. This proof of concept was constructed and the input signals from the wind tunnel measurement apparatus were wired into the breadboard. Using the non-inverting amplifier circuit the output from the breadboard was measured using a multimeter. Results confirmed the teams’ expectations and a readable amplified voltage signal was received by the multimeter. This method of testing was easier for conducting multiple preliminary tests than integrating a fully licensed copy of LabView and a computer, which had not yet been installed in the wind tunnel lab. A circuit design schematic of the non-inverting and transimpedance amplifiers used in the proof of concept testing are shown in figures 12 and 13 respectively.
To switch from recording and calculating each force and wind speed manually, the team chose to use an NI-DAQ to transmit the signals to the computer. The strain gauges on the test apparatus output a millivolt signal. Because the DAQ has a range of -10V to 10V and an accuracy of ±1.53 mV, a huge error would be present on each recording. To address this problem the team used an AD627AN op-amp as shown in the amplification circuit depicted in figure 15. The team needed to amplify the voltage at least 1000 times to output in millivolts. The exact amplification is not important since the signals will be brought to zero in LabVIEW for the start of any test. The proof of concept, which used a generic 741 op-amp, proved that this method was viable but might not be accurate enough for any significant calculations. After switching to the AD627AN, the team found that this method was sufficient. The installation of the DAQ entailed running the jumper wires from the blue box selector knob and pressure transmitter to the amplification and conversion circuit designed by the team. The full design schematic for the DAQ is shown in figure 14. This circuit board will be encased in the blue box so it will not need a separate housing. The output from each op-amp in the circuit is then ran to the DAQ which is plugged into the computer. Once the system was physically in place, the team wrote the LabVIEW program which acts as the user’s data recording interface. When designing this portion of the project, the team wanted to design for functionality so any person can obtain data with little help. The team created a program that allows the user to have the data graphically represented in real time while doing an experiment. The DAQs, computer, and LabVIEW software were all donated to the team. The only cost of this part of the design was the resistors, breadboard and op-amps. The data acquisition system is designed for cost and functionality. More expensive DAQs exist that would not require any amplification however purchasing one would have put the team over budget.
For the data acquisition system, the team’s goal was to simultaneously measure and record the outputs for roll, pitch, yaw, drag, lift, side force, and wind speed. This sub-project either worked or did not, so the initial testing process consisted of manipulating either the circuit or the LabVIEW program when results did not meet specifications. Once it appeared that the data acquisition system was accurately measuring its inputs, testing went as follows:
- Wire the system as it was originally
- Zero the lift output on the blue box
- Gradually record the output data while gradually placing weights on the force balance
Once these results were recorded, the new data acquisition system was wired and the same tests were taken. Comparing the data showed identical outputs as the original system for each weight. The system worked excellently giving accurate measurements of forces with an uncertainty of about +/- .013 N or about 2 grams. It was also witnessed during this test that all outputs were also being measured. This proved that we were getting simultaneous readings for each output. The wind speed measurement was tested in a similar manner. The old system was tested against the new system at varying wind speeds. The results showed the measurements once again to be identical. These tests were also complemented by actual use of our installed equipment. The intermediate fluid mechanics class used the new data acquisition system for a lab. The reception of it was very positive. The team received positive feedback from both the lab instructor (Dr. Fu) and the students who had used the wind tunnel prior to its renovation. The final LabVIEW output panel and VI can be seen in figures 16 and 17 respectively.
For the fabrication of team Victory Lap’s final design, the team manufactured or created three subprojects to meet our designed specifications and the needs of the university and its students. The three subprojects manufactured by Team Victory Lap were; a smoke generator system for airflow visualization, a staged rigid honeycomb airflow laminarization system, and a continuous multifunctional data acquisition system. To fabricate the smoke generator system intended for airflow visualization, a settling chamber and suction fan were implemented to create a denser smoke outflow in the test section. In addition, the straight portion of piping after the radius’ outlet was extended to ten times the diameter of the outlet pipe to ensure the relaminarization of smoke filled air exiting the piping. The final smoke generator system design is shown in figure 18.
The fabrication of the amplification circuit was fairly simple. Each op-amp is powered by the same source of +-9V DC. Every op-amp also shares a common ground with the DAQ. A 0.1uF capacitor is wired across the input terminal of the op-amp. A separate variable resistor controls the gain of each op amp as well. The wind speed is read as a voltage drop across a 100-ohm resistor. The final circuit can be seen in figure 19 below.
Jumper wires were used to connect the signal from the six resistors on the signal selector knob to the input terminals of the op-amp. The outputs of each amplifier are wired to two seperate DAQs each capable of simultaneously reading 4 signals. The final wiring can be seen in figure 20 below.
The input signals, read by an NI DAQ, were converted and filtered using a developed LabVIEW program to intuitively display the resultant forces for each signal received in the appropriate units.
In order to install the rigid honeycomb sandwich panels in the wind tunnel, an access door was needed to meet the necessary space requirements. The cut was made such that no structural support members were affected. Figure 21, shown below, illustrates the trap door cut into the underside of the wind tunnel.
Once completed, the manufactured honeycomb was installed in the wind tunnel. To keep the sandwich panels aligned, a bolt was installed through each corner mount of honeycomb. In total, twelve L-bracket style mounts were created to secure the manufactured honeycomb at the contraction cone inlet. An additional eight mounts were used to secure a smaller donated honeycomb at the test section entrance. This can be seen in figure 22 shown below.
Ryan was born in San Diego, California in 1992. He moved to the Danville, California in 1995 where he lived most his life. He graduated high school in 2010 and then began attending school at the University of Nevada, Reno. He is currently a senior at the university studying mechanical engineering and plans on graduating with a bachelors degree in the spring. Ryan has some engineering experience outside of school from an internship for a company called USS-POSCO. He interned there the last four summers working in the technology department doing mostly engineering design and CAD drawings.
Wolf was born in Pleasanton California and moved to the small town of Cool California when he was two. He grew up in Cool enjoying activities such as sports, hunting, and fishing. He began attending the University of Nevada as a computer science major but decided that the mechanical engineering field allowed for more creativity and design which led him to switch majors. He has experience working for the University at the Mobile Engineering Lab, which is a outreach program that attempts to interest students in the field of engineering.
Brian was born in 1992 in Reno, Nevada. He went to Reno High school and graduated in 2010 before attending the University of Nevada Reno. He plans on graduating in the spring of 2015 with a bachelors degree in mechanical engineering and a minor in renewable energy. He has worked for Electratherm, a waste heat recovery organic Rankine cycle company, since January 2014. At Electratherm, he has gained experience with test instrumentation that will be valuable towards the teams project.
Ryan was born in Honolulu, Hawaii and grew up in Carson City, Nevada playing football and wrestling. He has always enjoyed problem solving and figuring out how the real world works. He graduated high school with a 4.0 GPA and started attending the University of Nevada in fall of 2010. He plans to finish his degree in Mechanical Engineering and minor in Business and autonomous systems. He currently enjoys working on bikes and doing small projects.
Jayme was born in Reno, Nevada and grew up outside of Sacramento, California in a small town called Lincoln. He grew up racing motorcycles all across the country while maintaining a 4.0 GPA. He started attending the University of Nevada in Fall of 2009 as a mechanical engineering undergraduate student and has not lost his focus since. He enjoys mechanical engineering because of the background and knowledge it has given him and likes to look at the world around us and conceptualize the amount of work and engineering knowledge that goes into systems, processes, and structures that appear everyday in the world around us. Jayme has experience working in the engineering field as an intern for Western Sierras Inc., an overhead electrical parts and components distributor for light rail vehicles (LRVs) throughout the US and Canada. This opportunity has provided him skills necessary to manage projects, understand the concepts and work that go into project estimating, and review and analyze technical drawings using various engineering software applications.
Team Victory Lap would like to give a special thank you to Rachael Green for her help and contributions to the teams project. She made the project possible by providing the team with two NI-DAQs for free that were used in the data acquisition system. Without her donations, the team would have most likely not met the budget. She was also always willing to help and answer any questions the team had which helped make the project a success.
Dr. Henry Fu
The team would also like to thank Dr. Fu for being a great team mentor. He was always easy to get in contact with and was able to provide the team with some good advice on the project. He also donated a computer to the team which made the data acquisition system possible without going over budget.